Management of a modern IT сompany: theoretical and technological aspects
Abstract
The object of the study is the processes of functioning and maintenance, which together determine the stage of operation of a web-based information system for managing an enterprise or organization.During the study, the problem of developing elements of the methodology for lifecycle management of a web-based information system was solved. Modern research in this area is mainly aimed at analyzing and developing individual aspects of managing the operation of complex systems. The main attention is paid to solving individual tasks of managing the operation of systems based on IoT or DevOps. General issues of creating and improving the System Lifecycle Management approach remain little studied.During the study, the main provisions of the effective management concept of the life cycle of a web-based information system were proposed. This concept is based on a multi-level representation of the information system and the possibility of effective management of the system according to its properties. For further formal description of the proposed concept, the main concepts were defined and the ontology of this concept was developed. Based on this ontology, a generalized model of effective management of the operation of a web-based information system and a set of theoretical and categorical models that detail the description of the elements of this model were developed.The resulting generalized model was taken as a basis for developing a formal problem statement of effective life cycle management of an operated information system as an IT service system. A formal description of the objective function and constraints of the problem of classical (permanent) life cycle management of an operated information system was proposed. On its basis, a formal description of the objective function and constraints of the problem of effective life cycle management of an operated web-based information system was proposed. It was determined that the problem of classical (permanent) life cycle management of an operated information system is a special case of the proposed problem of effective management.The practical application of the obtained research results allows improving systems for life cycle management of an operated information system without global reengineering of existing systems and technologies for data storage and processing.
Full text
MANAGEMENT OF AMODERN IT COMPANY: THEORETICAL AND TECHNOLOGICAL ASPECTS Collective monograph Tallinn Edited by Maksym Yevlanov
Published in November 2025 by Scientific Route OÜ® Parda tn 4, Kontor 526, Tallinn, Harju maakond Estonia, 10151 www.route.ee Management of a modern IT company: theoretical and technological aspects Maksym Yevlanov (Editor) This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The Open Access version of this book, available at monograph.route.ee, has been made available under a Creative Commons Attribution 4.0 International License. Cover photo: "Tech icons over a green server background" © Canva.com. The cover was created using a Canva's Content License. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. DOI: 10.21303/978-9908-9706-8-4 ISBN 978-9908-9706-8-4 (eBook) ISBN 978-9908-9706-9-1 (ePub) This publication has been peer reviewed. ISBN 978-9908-9706-8-4 (eBook) ISBN 978-9908-9706-9-1 (ePub) © The Author(s) of individual chapters, 2025 This is an open access book under the Creative Commons Attribution 4.0 International License (CC BY 4.0)
iii AUTHORS Chapter 1 Viktor Levykin Doctor of Technical Sciences, Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-7929-515X Maksym Yevlanov Doctor of Technical Sciences, Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-6703-5166 Oleksandr Petrychenko Senior Researcher Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-1319-5041 Olga Neumyvakina Senior Researcher, Leading Engineer Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0001-6936-6543 Chapter 2 Maksym Yevlanov Doctor of Technical Sciences, Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-6703-5166 Viktor Levykin Doctor of Technical Sciences, Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-7929-515X Borys Moroz Doctor of Technical Sciences, Professor Department of Software Engineering Dnipro University of Technology ORCID: https://orcid.org/0000-0002-5625-0864 Dmytro Moroz PhD, Associate Professor Department of Software Engineering Dnipro University of Technology ORCID: https://orcid.org/0000-0003-2577-3352 Ivan Iuriev PhD, Associate Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-5178-519X Chapter 3 Tetiana Borysenko PhD, Associate Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0001-6915-6861 Maksym Yevlanov Doctor of Technical Sciences, Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-6703-5166 Konstantin Petrov Doctor of Technical Sciences, Professor, Head of Department Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0003-1973-711X Viktor Borysenko PhD, Associate Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0001-9372-6449 Chapter 4 Maksym Yevlanov Doctor of Technical Sciences, Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-6703-5166
iv Management of a modern IT company: theoretical and technological aspects Nataliia Vasyltsova PhD, Associate Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-4043-487X Iryna Panforova PhD, Associate Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0001-7032-9109 Dmytro Osypchuk Independent Researcher ORCID: https://orcid.org/0009-0005-2808-2145 Chapter 5 Maksym Yevlanov Doctor of Technical Sciences, Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-6703-5166 Nataliia Vasyltsova PhD, Associate Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0002-4043-487X Iryna Panforova PhD, Associate Professor Department of Information Control System Kharkiv National University of Radio Electronics ORCID: https://orcid.org/0000-0001-7032-9109 Anastasiia Popova Independent Researcher ORCID: https://orcid.org/0009-0007-4780-7556
v The monograph is devoted to solving scientific and applied problems that arise in the process of managing modern IT companies. The rapid development of IT products in the 21st century has caused significant changes in the processes and management system of an IT company as a commercial enterprise providing services. These changes have led to the perception of a modern IT company as an enterprise that combines the most successful practices, models and methods of planned and project management in its activities. But this view of an IT company is detailed only at the conceptual level. The issues of formal description, design and development of systems and technologies that would be based on this concept remain largely unresolved. The vast majority of modern IT products used to manage an IT company and its processes are focused on the concept of exclusively project management and do not take into account the features of planned practices of modern service management. The monograph considers the solution of theoretical and theoretical-applied issues related to the formulation and solution of important tasks of managing the general activities of an IT company and its individual processes. Such tasks include the task of managing the operation of web-based information systems, the task of assigning tasks to IT company employees, as well as the task of assessing the costs of IT company resources during its activities. The research results are formal models, methods, and elements of information technology that can be used to build management systems for modern IT companies. The research used mathematical tools of category theory and set theory; models and methods of statistical analysis and Data Mining, as well as existing models and methods of assessing the costs of efforts and resources in IT projects. The obtained research results and their experimental verification indicate the possibility of improving the processes of modern IT companies. Such improvement can be carried out through the consistent development of individual elements of the IT company management system, without global reengineering of systems and technologies for storing and processing data currently used in the IT company. This monograph can be useful to researchers, teachers, graduate students and higher education students in the field of computer science, information management systems and IT project management. Keywords IT project, information system, operations management, description of a system element, incomplete information, task, performer, evaluation, localization, video game, change, Beckhard and Harris model, descriptor, sprint. ABSTRACT
vi CIRCLE OF READERS AND SCOPE OF APPLICATION The first chapter presents the results of the development of elements of the method ology, models and formal formulation of the problem of effective management of the life cycle of the operation of a web-based information system (IS). These results can be used to build an IS of multi-level management of IT services provided by an IT company. Such use is appropriate when developing a strategy for creating and gradually improving an IT company management system. The materials of the chapter will be useful to scientists in the fields of computer science, information systems, as well as management of the activities of an IT company and its individual IT projects. The second chapter presents the results of solving the problem of modeling losses when designing a description of an IS element in conditions of incompleteness or lack of information about this element. These results can be used in the process of managing IT project decisions in conditions of partial uncertainty. Such use is appropriate for automated assessment of possible costs in the process of forming and changing task backlogs for IT project performers. The materials of the chapter will be useful to scientists and specialists in IT project management of the development of IS management of enterprises and organizations. The third chapter proposes the results of solving the problem of assigning IT project tasks to its performers. These results can be used to automatically assign IT project tasks to its performers, taking into account the previous work experience of these performers. Such use is appropriate for creating services that can interact with both existing automated management tools for an IT company and those being developed. The materials of the chapter will be useful to scientists and practitioners in the fields of IT project team management and the use of artificial intelligence tools for managing IT projects. The fourth chapter proposes the results of solving the problem of assessing the complexity of an IT project for video game localization. These results can be used when planning IT projects for video game localization. Such use is appropriate as amethod of automated assessment of efforts to implement an IT project for video game localization. The materials of the chapter will be useful to scientists and practitioners in the fields of IT project management for the localization of various computer games. The fifth chapter proposes the results of solving the problem of quantitative assessment of changes that occur during the implementation of a long-term IT project. These results can be used to manage changes that occur in IT projects that last more than a year. Such use is appropriate for automated assessment of time spent on implementing changes during individual iterations of an IT project. The materials of the chapter will be useful to scientists and practitioners in managing teams of performers and changes in IT projects.
vii CONTENTS List of Tables.............................................................................................................................................x List of Figures ........................................................................................................................................xii Introduction .............................................................................................................................................1 CHAPTER 1 Basics of the effective management methodology of web-based information systems ............................................................................................................................4 1.1 Introduction .......................................................................................................................5 1.2 Analysis of modern research in the field of effective management ofweb-based information systems ..........................................................................7 1.3 Research concepts and assumptions ....................................................................11 1.4 Results of the development of the elements of the methodology for effective operation management of a web-based information system ......................................................................................................16 1.4.1 Results of the development of the concept and ontology for effective operation management of a web-based information system .......................................................................................16 1.4.2 Results of development of a set of theoretical and categorical models of effective operation of a web-based information system .......................................................................................20 1.4.3 Results of the development of a formal statement of the problem ofeffective management of the operation of a web-based information system ........................................................25 1.5 Results of the development of a theoretical and categorical model of the property of an information system ...........................................................28 1.6 Discussion of the research results .........................................................................36 1.7 Conclusions .....................................................................................................................39 References .................................................................................................................................41 CHAPTER 2 Cost management of an information system IT project in the conditions of incomplete information about its elements ..................................45 2.1 Introduction ....................................................................................................................46 2.2 Analysis of current research in the field of IT project decision management ...................................................................................................................47 2.3 Research models ...........................................................................................................50
Management of a modern IT company: theoretical and technological aspects viii 2.4 Results of adapting the causal scheme of production losses to the features of the design processes of an information system element .......54 2.5 Results of developing detailed models of losses from lack of information ................................................................................................................57 2.6 Experimental verification of the obtained results ...........................................66 2.6.1 Description of the initial data ....................................................................66 2.6.2 Description of the assessment of losses due to the replacement ofincomplete information with information about a similar element of the information system at the stage of IT project initiation ..........................................................70 2.7 Discussion of the results of the loss assessment for the considered solution options .............................................................................................................79 2.8 Conclusions .....................................................................................................................82 References .................................................................................................................................83 CHAPTER 3 Method of automatic assignment of tasks to IT project performers ....86 3.1 Introduction ....................................................................................................................87 3.2 Overview of the current state of solving the problem of automated planning of IT projects .................................................................................................88 3.3 Materials and methods ...............................................................................................94 3.4 Results of developing a method for solving the problem of assigning ITproject tasks to its performers ........................................................................100 3.4.1 Results of adapting the polynomial naive Bayes classifier tothespecifics of the problem of assigning IT project tasks toitsperformers .........................................................................................100 3.4.2 Description of the results of developing an algorithm forimplementing the developed method ..........................................104 3.5 Experimental verification of the improved method .....................................111 3.6 Discussion of the research results ......................................................................120 3.7 Conclusions .................................................................................................................. 122 References ..............................................................................................................................124 CHAPTER 4 Method for assessing the complexity of an IT project for video game localization .........................................................................................................128 4.1 Introduction ................................................................................................................. 129 4.2 Analysis of current research in the field of software and video game localization ...........................................................................................130 4.3 Basic research method.............................................................................................136
© The Author(s) of individual chapters, 2025 DOI: 10.21303/978-9908-9706-8-4.introduction INTRODUCTION A modern IT company is an enterprise that stands out among all other types of enterprises. The main goal of an IT company is to provide services for the development, implementation, support, maintenance and development of IT products for various purposes on behalf of various stakeholders. Therefore, an IT company is forced to combine the best practices of planned management of IT services and project adaptation of these IT services to the requirements of individual customers during its own activities. Such a combination requires, in turn, the creation and development of an appropriate methodology, models, methods, systems and technologies for managing an IT company and its individual processes. Unfortunately, the current state of science allows to describe in detail the methodological issues of creating and using an IT company management system and its processes mainly at the conceptual level. A significant number of scientific studies in recent years are aimed at solving only individual issues of managing the activities of an IT company. This can be explained by the complexity of the problem of building a methodological and formal basis for such a management system. Such a lag is also observed in the field of development and use of specialized information systems and technologies for managing an IT company. The vast majority of modern IT products used to manage an IT company and its processes are focused on the concept of exclusively project management and do not take into account the peculiarities of planned practices of modern service management. Some work on the creation of information systems and technologies for multilevel management of IT services was started in the early 2010s. But still the main results of these works are too large IT products, the cost and complexity of which significantly limits their distribution. Solving the considered problems of managing a modern IT company requires a large amount of scientific and applied work that cannot be performed by a single team of researchers. But the gradual solution of scientific and applied issues, coordinated with the modern concept of IT company management and with the results obtained by individual research groups, can provide the governing bodies of IT companies with results that have significant applied value. Therefore, conducting research in this area is relevant from a theoretical and applied point of view. The monograph offered to your attention contains the results of research by employees of the Department of Information Control Systems of the Kharkiv National University of Radio Electronics, the Department of Computer Systems Software of the National Technical University "Dnipro Polytechnic" and individual
Management of a modern IT company: theoretical and technological aspects 2 independent researchers working in the field of management of IT companies and individual IT projects. This monograph considers the solution of theoretical and theoretical-applied issues related to the formulation and resolution of important problems and tasks of managing the general activities of an IT company and its individual processes, namely: –problems of effective management of the operation of web-based information systems(Chapter1); –problems of managing the costs of an IT project of an information system in conditions of incomplete information about its elements(Chapter2); –tasks of assigning tasks to IT project performers(Chapter3); –tasks of assessing the complexity of an IT project for video game localization(Chapter4); –tasks of quantitative assessment of changes in the long-term IT project management system(Chapter5). During the research described in the chapters of the monograph, mathematical tools of category theory and set theory, models and methods of statistical analysis and Data Mining, as well as existing models and methods for estimating the costs of efforts and resources in IT projects were used. The main theoretical results of the research presented in this monograph are: –the concept of the problem of effective management of the life cycle of an operated information system as an IT service system; –a set of models for effective management of the operation of a web-based information system; –a formal description of the problem of effective management of the life cycle of an operated information system; –a theoretical and categorical model of the properties of an information system; –detailed models of losses in the activities of an IT company due to the lack of information; –a method for solving the problem of assigning IT project tasks to its performers; –a new method for assessing the complexity of IT projects for video game localization; –an improved method based on the Beckhard and Harris model for quantitative assessment of changes in a long-term IT project. The obtained theoretical results were verified during the implementation of relevant IT projects or on data from IT projects implemented using other methods and techniques. The results of the tests confirm that the application of the models and methods proposed by the authors of the monograph is better than the use of existing models, methods and tools for solving similar problems of IT company management.
Introduction 3 This monograph may be useful to researchers, teachers, postgraduate students and higher education students in the field of computer science, information management systems and IT project management. The authors express special gratitude to the first reader of the monograph – PhD, Leading Researcher of the Department of Information Control Systems of the Kharkiv National University of Radio Electronics, Olha Neumyvakina – for valuable recommendations, attention, and versatile erudition, repeatedly revealed during the creation of the monograph.
© The Author(s) of individual chapters, 2025 DOI: 10.21303/978-9908-9706-8-4.ch1 Abstract The object of the study is the processes of functioning and maintenance, which together determine the stage of operation of a web-based information system for managing an enterprise or organization. During the study, the problem of developing elements of the methodology for lifecycle management of a web-based information system was solved. Modern research in this area is mainly aimed at analyzing and developing individual aspects of managing the operation of complex systems. The main attention is paid to solving individual tasks of managing the operation of systems based on IoT or DevOps. General issues of creating and improving the System Lifecycle Management approach remain little studied. During the study, the main provisions of the effective management concept of the life cycle of a web-based information system were proposed. This concept is based on a multi-level representation of the information system and the possibility of effective management of the system according to its properties. For further formal description of the proposed concept, the main concepts were defined and the ontology of this concept was developed. Based on this ontology, a generalized model of effective management of the operation of a web-based information system and aset of theoretical and categorical models that detail the description of the elements of this model were developed. The resulting generalized model was taken as a basis for developing a formal problem statement of effective life cycle management of an operated information system as an IT service system. A formal description of the objective function and constraints of the problem of classical (permanent) life cycle management of an Viktor Levykin Maksym Yevlanov Oleksandr Petrychenko Olga Neumyvakina CHAPTER 1 Basics of the effective management methodology of web-based information systems
5 Basics of the effective management methodology of web-based information systems Chapter 1 operated information system was proposed. On its basis, a formal description of the objective function and constraints of the problem of effective life cycle management of an operated web-based information system was proposed. It was determined that the problem of classical (permanent) life cycle management of an operated information system is a special case of the proposed problem of effective management. The practical application of the obtained research results allows improving systems for life cycle management of an operated information system without global reengineering of existing systems and technologies for data storage and processing. Keywords Web-based information system, system property, operation management, ontology of effective management, theoretical-categorical model, attributive model. 1.1 Introduction The first quarter of the 21st century is characterized by the rapid development of web-based information systems (IS) for managing enterprises and organizations. The term "web-based IS" will be used here and further to refer to IS that uses web technologies to provide information and services to users or other information systems/applications[1]. One of the most important consequences of this definition is the perception of web-based IS as systems, which functions can be accessed from anywhere in the world (provided there is a working Internet connection) via a browser. The modern view of the architecture of web-based IS recognizes the existence of several different approaches to describing this architecture. However, to reflect the above-mentioned perception of web-based IS, it is best to use the description of the architecture of such a system proposed in[2] (Fig.1.1). In Fig.1.1, the following designations are adopted for the elements of the description of the architecture of web-based IS: DNS– Domaine Name System, a domain name system that ensures the operation of the Internet; CDN– Content Delivery System, a content delivery system that provides acceleration of the delivery of static HNML, CSS, JavaScript files and images (dotted lines separate the elements of the web-based system that are located on the system's web servers and can be located practically anywhere on the globe). Detailed descriptions of these elements are given in[2]. Creation, support and development of web-based IS with the architecture described in Fig.1.1 significantly complicates the solution of the problem of managing such IS within their life cycle (LC). Such management is necessary for any IT company that promotes web-based IS of its own production on the market of IT products
6 Management of a modern IT company: theoretical and technological aspects intended for the management of enterprises and organizations. The goal of such management in general should be to increase the competitiveness of web-based IS by maintaining the characteristics of efficiency and quality of this system at the level most desirable for all stakeholders. DNS Load Balancer CDN User's browser Cloud storage Copy of data Database Web app servers Data warehouse Data "firehose" Caching service Job Queue Services Job servers Full text search service 1 11 210 9b 9a 9c 3 4 56a 6b 8 7 Fig. 1.1 Description of the architecture of a web-based system Source:[2] The modern approach to the creation and functioning of IS management of enterprises and organizations conceptually considers such systems as a type of system for providing IT services to operators and users[3]. Here and further, the term "service" will be understood as the performance of actions, work or duties, which is self-sufficient, consistent, discrete and can be composed of other services[4]. Based on this definition, the term "IT service" was proposed to be used to refer to an independent functional task in IS, the use of which is economically and technically feasible for performing a separate activity of the process of an enterprise or organization or for managing this activity[5]. The results of implementing this approach in the IT industry are[6]: –the concept of an IT service management system as a system for managing and controlling the activities of managing IT objects of an IT company;
7 Basics of the effective management methodology of web-based information systems Chapter 1 –an approach "Product Lifecycle Management" (PLM); –an approach "Application Lifecycle Management" (ALM). The term "PLM" will be understood as an approach to managing processes and methods that are used or created during the LC of products or systems. Its scope mainly covers the hardware parts of the system[7]. The term "ALM" will be understood as an approach focused on the software LC. It is designed to monitor, control and manage artifacts and processes that arise during the program LC. These processes include requirements development, source code management, test management, release management, etc.[8]. To solve the problem of managing services during their LC, a separate type of management system was created– service management systems. According to the modern understanding, such systems should be perceived as systems for managing and controlling the activities of the service provider from the point of view of service management[3]. However, although the PLM and ALM approaches arose on the basis of the same concept of an IT service management system, the ISs created within these approaches developed in different ways. In order to eliminate the difference between these ISs, a new approach was developed, the main difference of which is the integration of PLM and ALM into the interdisciplinary management of the system's LC (SLM)[9]. Unfortunately, the creation and development of the SLM approach occurs mainly at the conceptual level. The reasons for this should be considered different approaches to the structuring of hardware and software, as well as to the integration of elements of this software into systems. In addition, the phases of the software and hardware LCs also differ, even if there are similarities[10]. Therefore, conducting theoretical and theoretical-applied research devoted to the development of methodological foundations of the SLM approach at the conceptual and formal levels is relevant. 1.2 Analysis of modern research in the field of effective management ofweb-based information systems Modern information technologies (IT) and IS based on PLM are aimed at managing large amounts of information and artifacts that arise and accumulate during the product LC[7]. Such PLM-IT and PLM-IS are considered as a center that ensures the existence of compatible data flows (for example, data from CAD products, modeling or architecture data, etc.) and solving the problems of managing these data flows. This is due to the fact that PLM grew out of CAD (Computer Aided Design) and PDM(Product Data Management). For this reason, PLM-IT and PLM-IS are only limitedly suitable for displaying and supporting the software LC[9].
8 Management of a modern IT company: theoretical and technological aspects The use of PLM-IT and PLM-IS for managing the IS LC in enterprises and organizations is complicated by the need to manage changes that arise during the creation and operation of such IS. Thus, in[11] the reasons for these complications are the lack of compatibility of product change notifications, low data quality, and the lack of direct traceability and tracking of changes to the final product. To eliminate these complications,[11] proposes to integrate PLM-IS and IT asset administration(Asset Administration Shell). Such integration should improve cooperation between PLM-IS and IS of enterprise and organization management, improve data quality, ensure traceability and increase efficiency within engineering processes. However, this approach is not very suitable for LC management, because it focuses mainly on hardware management. Unlike PLM, ALM was originally considered as an approach that can theoretically be used without the use of appropriate tools. ALM was mainly proposed to be used to integrate all processes, methods and data used and generated within software development[8]. The modern vision distinguishes three main aspects within ALM: governance, development and operations[9]. The governance aspect considers the tasks of conducting software business analysis, project portfolio management and program portfolio management. The development aspect considers the tasks of software generation (from requirements definition through design and source code development to testing and release). The operation aspect considers the tasks of software monitoring and management[12]. Modern research suggests understanding SLM as a technical and organizational basis for LC management of all system artifacts that are created or improved during the existence of this system, and the tools used to work with these artifacts. This basis provides clear tracking of all system elements. Thus, interdisciplinary SLM is considered the basis for complex digital engineering[13]. But IT products that are the results of this complex digital engineering differ significantly from each other. Examples of such IT products include: –Microsoft products (Microsoft System Center 2016, Microsoft Azure Automation & Control in Microsoft Operations Management Suite); –specialized products of individual IT companies (Cortex, ActiveBatch Workload Automation, Ignio, VMware Vrealize Orchestrator, etc.); –products that are descendants of CASE systems and visual modeling environments (CA Process Automation). As can be seen, the main difference between IT products designed to implement the SLM approach is caused mainly by the features and requirements of specific IT companies, on which orders these products were created. It should be noted that there is no fixed classification of IT products that can be used to implement the SLMapproach.
9 Basics of the effective management methodology of web-based information systems Chapter 1 But it should be recognized that these and similar IT products generally meet the requirements for service management systems defined in[3]. In general, the set of functions of such products is aimed at achieving the following goals: – increasing the efficiency of the functioning of IT processes of the enterprise-consumer of IT services (technical and economic goal); –reducing the operational risks of the enterprise-consumer of IT services (technical and economic goal); –reducing the costs of IT processes of the enterprise-consumer of IT services throughout the entire LC (economic goal); –compensation for the negative consequences of the complexity of managing the IT services system of the consumer enterprise, in particular, eliminating the effect of "IT blindness" (economic goal); –increasing the level of compliance with standards, including to facilitate further certification (economic and organizational and methodological goal). Unlike applied developments, most theoretical research in the field of PLM, ALM and SML is aimed at developing individual aspects of IT service management of large industrial systems based on the Internet of Things. For example, in[14] a description of the information layer of the reference architectural model RAMI 4.0 is proposed, which is supported by the Zachman Framework concept. The proposed description, according to the authors, should ensure the practical use of RAMI4.0, allowing stakeholders to use model-based systems engineering, on the one hand, and include information engineering, on the other. However, the description proposed in[14] can be used to develop only individual aspects of industrial systems and does not cover the task of industrial IS management as a whole. Similar research is being conducted in the IT industry. Thus, in[15] the results of mixed modeling and modeling of continuous service delivery pipeline scenarios as a separate aspect of DevOps are described. But these studies also do not cover the overall task of managing IS and IT, which an IT company uses in its activities. The formation of integral assessments of the progress and results of the functioning of individual IS services currently does not involve the use of formal models[16]. At the same time, studies have shown that the success of projects aimed at improving individual processes is not a coincidence, but a predictable result with clearly defined and measurable characteristics[17]. To solve various IT service management problems, approaches based on solving multi-criteria optimization problems are proposed in[18]. However, the difficulties of applying such approaches in managing real webbased IS force to look for other options for solving such problems. Thus, in[19] it is proposed to consider an approach to web service management based on knowledge. However, this approach is not yet fully developed and requires additional research.
10 Management of a modern IT company: theoretical and technological aspects A significant number of studies consider the problem of managing the IS LC mainly through the prism of tools for its solution. In this case, the multi-criteria decision-making process is usually proposed as the main tool. To implement this process, decision support systems based on multidimensional data warehouses are proposed. An example of such a platform designed to automate the management of the stages of the LC describing the development of modern software is considered in[20]. Asignificant part of the publications is devoted to describing the results of research into problems that arise when using tools for solving such multi-criteria optimization problems. Algorithms[21] and tools[22] for analyzing and optimizing hyperparameters are considered as similar tools in modern research. Studies devoted to the use of simulation modeling tools for LC management of individual aspects of IS[23] are not left without attention. The following directions for further research on the development of SLM are indicated in[9]: –improving the implementation of the knowledge management process in the business (in particular, ensuring the interconnection of business goals and indicators with current processes to identify awareness of complaints and the potential for business improvement[24]); –developing an integrated or common ALM-PLM data model (in particular, to identify and minimize differences between the descriptions of the hardware and software development processes[25]); –extending the requirements and use cases of the ALM-PLM system to other stages of the Life Cycle; –urther detailing and refining the SLM, PLM and ALM approaches, developing these approaches through new methods, ideas and processes to support the development of software and hardware; –defining KPIs for ALM-PLM integration (in particular, to further evaluate and control the use of these approaches to assess their success or potential for improvement[26]). But the SLM development is impossible without eliminating an important contradiction in the field of IT service management. The essence of this contradiction is as follows. A typical IT service management system, defined in[1], recognizes the goal of management as maintaining user satisfaction with existing IT services at the level desired by the Provider and Consumer of IT services. At the same time, the main quantitative criterion by which it is possible to determine the degree of achievement of this goal is the indicator of the number of incident reports that remain unresolved. But these incident reports arise as a result of the interaction of operators and users with existing IT services and the IT infrastructure that ensures the operation of these
17 Basics of the effective management methodology of web-based information systems Chapter 1 Recognition of these points of view allowed to formulate the basic concept of effective management of the operation of web-based IS as a set of the following provisions[27]: –effective management of the operation of IS and its CIs from the consumer's point of view is considered as a set of design and/or operational measures, the result of which is the achievement of global IS efficiency from the point of view of all employees of the Consumer while maintaining the costs of these measures at the level desired by the Consumer; –effective management of the operation of IS and its CIs from the Supplier's point of view is considered as a set of design and/or operational measures, the result of which is the achievement of global IS efficiency from the point of view of all employees of the Supplier while maintaining the costs of these measures at the level desired by the Supplier; –effective management of the IS operation as a whole while respecting the interests of the Consumer and the Supplier of this IS is considered as a special case of the multi-criteria optimization problem, the solution of which will be Pareto-optimal IS; –effective management of the operation of individual CIs of the IS while respecting the interests of the Consumer and the Supplier of this IS is considered as a set of design and/or operational measures to transform a set of current values of the indicators of the properties of the CIs of the IS (operational characteristics of individual CIs of the IS) into a set of planned values of these CIs, at which the IS as a whole will be Pareto-optimal. In this study, the main attention is paid to the development of elements of the methodology for effective operation management of web-based IS precisely from the point of view of the Supplier. The role of such a Supplier can be any IT company that provides (sells or leases) its own web-based IS for operation by enterprises or organizations acting as Consumers. Based on the developed concept of operation management of web-based IS, an ontology of the methodology for effective management of IT services of webbased IS was proposed. This ontology was obtained by improving the ontology of the existing service management methodology. The diagram of the relationships between the main concepts of the ontology of effective IT service management is shown in Fig.1.3[27]. To improve perception, Fig.1.3 does not show the main slots of frames describing the main concepts. In red in Fig.1.3, concepts that are introduced into the ontology as a result of improving the service management methodology are high lighted. In italics in Fig.1.3, concepts that expand the service management methodol ogy in the course of solving applied service management problems are highlighted.
18 Management of a modern IT company: theoretical and technological aspects Provider of IS Customer of IS Requirement of IS Information System Configuration of IS Configuration Item State of IS Planned State Current State Efficiency RFC IS Indicator Library of Indicators Presentation Layer Transaction Property 1 1 1 11 1 1 1 1 11 1 11 11 1 11 1 1 Fig. 1.3 Diagram of the relationships between the main concepts of the ontology of the methodology for managing IT services of web-based IS Source:[27] The scheme shown in Fig.1.3 allows to distinguish groups of main entities of the subject area, on the basis of which it is possible to form the following descriptions[27]: –description of the operated IS at different levels of representation; –description of the states of the operated IS; –description of indicators that characterize the operated IS and its CIs; –description of the efficiency of the operation of the IS by its individual properties and in general. The description of the operated IS at different levels of representation will consist of the following concepts shown in Fig.1.3[27]: –"Information System" concept;
19 Basics of the effective management methodology of web-based information systems Chapter 1 –"Provider of IS" concept; –"Customer of IS" concept; –"Requirement of IS" concept; –"RFC IS" concept; –"Configuration of IS" concept; –"Presentation Layer" concept; –"Configuration Item" concept; –"Transaction" concept. These concepts define the main points of view that should be taken into account during a detailed formalized description of the operated IS at different levels of representation as a set of individual CIs and transactions created to meet the requirements of the IS Supplier and Consumer. The description of the states of the operated IS will consist of the following concepts, shown in Fig.1.1[27]: –"Configuration of IS" concept; –"Presentation Layer" concept; –"Configuration Item" concept; –"Transaction" concept; –"State of IS" concept; –"Planned state" concept; –"Current state" concept. These concepts define the main points of view that should be taken into account during a detailed formalized description of the operated IS at different levels of representation of both planned and current states of individual CIs and transactions. The description of indicators that characterize the operated IS and its CIs will consist of the following concepts, shown in Fig.1.3[27]: –"State of IS" concept; –"Indicator" concept; –"Library of Indicators" concept. These concepts define the main points of view that should be taken into account during a detailed formalized description of the operated IS by a set of individual indicators based on a specific library of indicators. The description of the efficiency of the IS operation by its individual properties and as a whole will consist of the following concepts shown in Fig.1.3[27]: –"State of IS" concept; –"Planned state" concept; –"Current state" concept; –"Indicator" concept;
20 Management of a modern IT company: theoretical and technological aspects –"Presentation Layer" concept; –"Property" concept; –"Efficiency" concept; –"RFC IS" concept. These concepts define the main points of view that should be taken into account during a detailed formalized description of the efficiency of the operated IS at different levels of representation. Partial duplication of concepts in different descriptions is necessary to ensure the connections of these descriptions with each other. 1.4.2 Results of development of a set of theoretical and categorical models of effective operation of a web-based information system Based on the groups of concepts of the subject area identified in Subsection1.4.1, it becomes possible to develop generalized mathematical models of effective management of the operation of web-based IS. Since these descriptions are sets of individual concepts defined in a different way, they can be formally represented in the form of the following categories[27]: –description of the operated IS at different levels of representation will be represented by the LIS category; –description of the states of the operated IS will be represented by the LSt category; –description of indicators characterizing the operated IS and its CIs will be represented by the Lind category; –description of the efficiency of the IS operation by its individual properties and in general will be represented by the LEff category. The LI category, taking into account the concepts of the ontology of web-based IS service management and the connections between these concepts, has the form L Ob Ob Ob Ob Ob Ob Ob Ob Ob H IS IS Cust recRFC Conf PL CI Tr O ,, ,, ,,,,, _Pr bb Ob Ob Ob Ob Ob Ob Ob Ob Ob Ob IS Cust IS rec IS recIS Cust H HHH H Pr ,, ,, , Pr CConf rec Conf PL Conf CI Conf Tr Tr CI Ob Ob Ob Ob Ob Ob Ob Ob Ob HHHH,,,,, _HHHH Ob Ob Ob Ob RFCRFC Cust Ob Pr ,, , 1 (1.1) where ObIS– a subset of LIS category objects that describes the "Information System"concept; ObPr– a subset of LIS category objects that describes the "Provider of IS" concept; ObCust– a subset of LIS category objects that describes the "Customer of IS"concept; Obrec– a subset of LIS category objects that describes the "Requirement of IS" concept; ObRFC– a subset of LIS category objects that describes the "RFC IS" concept; ObConf– a subset of LIS category objects that describes the "Configuration of IS"
21 Basics of the effective management methodology of web-based information systems Chapter 1 concept; ObP_L– a subset of LIS category objects that describes the "Presentation Layer" concept; ObCI– a subset of LIS category objects that describes the "Configuration Item" concept; ObTr– a subset of LIS category objects that describes the "Transaction" concept; HOb Ob IS Pr – a subset of LIS category morphisms defined between subsets ObPr and ObIS; HOb Ob Cust IS – a subset of LIS category morphisms defined between subsetsObIS and ObCust; HOb Ob rec IS – a subset of LIS category morphisms defined between subsetsObIS and Obrec; HOb Ob rec Pr – a subset of LIS category morphisms defined between subsets ObPr and Obrec; HOb Ob IS Cust – subset of LIS category morphisms defined between subsets ObCust and ObIS; H Ob Ob Conf rec– subset of LIS category morphisms defined between subsets Obrec and ObConf; H Ob Ob Conf PL _ – subset of LIS category morphisms defined between subsets ObP_L and ObConf; H Ob Ob Conf CI – subset of LIS category morphisms definedbetween subsets ObCI and ObConf; H Ob Ob Conf Tr – subset of LIS category morphisms defined between subsets ObTr and ObConf; HOb Ob Tr CI – subset LIS category morphisms defined between subsets ObCI and ObTr; HOb Ob RFC Pr – a subset of morphisms of the LIS category defined between the subsets ObPr and ObRFC; HOb Ob RFC Cust – a subset of morphisms of the LIS category defined between the subsets ObCust and ObRFC; HOb 1– a subset of single morphisms defined on selected subsets of objects of the LIS category. The LSt category, taking into account the concepts of the web-based IS service management ontology identified in[27] and the relationships between these concepts, has the form L Ob Ob Ob Ob Ob Ob Ob H St Conf PL CI Tr St Pl St CStOb Ob Conf P ,,,,,,, ___ _LL Conf CI Conf Tr Tr CI St Conf Pl St H HHHH Ob Ob Ob Ob Ob Ob Ob Ob Ob O ,, ,, ,_ bb Ob Ob St CSt St Ob HH,, , _1 (1.2) where ObSt– a subset of objects of the LSt category, which describes the concept of "State of IS" concept; ObPl_St – a subset of objects of the LSt category, which describes the concept of "Planned State" concept; ObC_St– a subset of objects of the LSt category, which describes the concept of "Current State" concept; HOb Ob St Conf – a subset of morphisms of the LSt category, defined between subsets ObCont and ObSt; H Ob Ob Pl St St _–asubset of morphisms of the LSt category, defined between subsets ObSt and ObPl_St; H Ob Ob CSt St _– a subset of morphisms of the LSt category, defined between subsets ObSt and ObC_St; HOb 1– a subset of single morphisms, defined on selected subsets of objects of the LSt category. The LInd category, taking into account the concepts of the web-based IS service management ontology highlighted in[27] and the connections between these concepts, has the form LO bObOb HHH IndStInd LibOb Ob Ob Ob St Ind Lib Ind Ob ,,,,, , 1 (1.3)
22 Management of a modern IT company: theoretical and technological aspects where ObInd– a subset of objects of the LInd category, which describes the "Indicator" concept; ObLib– a subset of objects of the LInd category, which describes the "Library of Indicators" concept; HOb Ob St Ind– a subset of morphisms of the LInd category, defined between the ObInd and ObSt subsets; HOb Ob Lib Ind– a subset of morphisms of the LInd category, defined between the ObInd and ObLib subsets; HOb 1– a subset of single morphisms, defined on selected subsets of objects of the LInd ategory. The LEff category, taking into account the concepts of the web-based IS service management ontology highlighted in[27] and the connections between these concepts, has the form L Ob Ob Ob Ob Ob Ob Ob Ob H Eff St Pl St CStInd PL op EffRFC O ,,,, ,,,, __ _Prbb Ob Ob Ob Ob Ob Ob Ob Ob Ob Pl St St CSt St St Ind op St op P H HH H __ Pr Pr _ ,, ,, LL op Eff Pl St Eff CSt Eff RFC St Ob HHHHH Ob Ob Ob Ob Ob Ob Ob Ob ,,,,, Pr __ 1 , (1.4) where ObProp– a subset of objects of the LEff category that describes the "Property" concept; ObEff– a subset of objects of the LEff category that describes the "Effi ciency" concept; H Ob Ob op St Pr – a subset of morphisms of the LEff category defined between the ObSt and ObProp subsets; H Ob Ob op PL Pr _– a subset of morphisms of the LEff category defined between the ObP_L and ObProp subsets; H Ob Ob Pl St Eff _– a subset of morphisms of the LEff category defined between the ObEff and ObPl_St; subsets H Ob Ob CSt Eff _– a subset of morphisms of the LEff category defined between the ObEff and ObPl_St subsets; HOb Ob RFC St – a subset of morphisms of the LEff category defined between the ObSt and ObRFC subsets; HOb 1– a subset of unit morphisms defined on selected subsets of objects of the LEff category. According to the provisions of[3], the management of effective operation of web-based IS is based on the cycle "Plan– Do– Check– Act" (PDCA). In this management cycle: –at the "Plan" stage, the development of web-based IS operation plans is carried out; –at the "Do" stage, work is carried out to implement web-based IS operation plans; –at the "Check" stage, work is carried out to verify the results of the implementation of web-based IS operation plans; –at the "Act" stage, work is carried out to correct web-based IS operation plans. Then the generalized model of effective management of web-based IS operation can be represented as a supercategory of the following form[27] ML LL LF FFFF OPDChAL L L L L L L L L L D P Ch D D Ch A Ch P A ,, ,, ,,,, , (1.5) where LP– a supercategory that formally descries the "Plan" stage; LD– a supercategory that formally describes the "Do" stage; LCh – a supercategory that formally describes the "Check" stage; LA– a supercategory that formally describes the
23 Basics of the effective management methodology of web-based information systems Chapter 1 "Act"stage; FL L D P– a single covariant functor that establishes a connection between the LP and LD supercategories; FL L Ch D– a single covariant functor that establishes a connection between the LD and LCh supercategories; FL L D Ch – a single covariant functor that establishes a connection between the LCh and LD supercategories; FL L A Ch – a single covariant functor that establishes a connection between the LCh and LA supercategories; FL L P A– a single covariant functor that establishes a connection between the LA and LP supercategories. The functors FL L D P, FL L Ch D, FL L A Ch and FL L P A connect the supercategories LP, LD, LCh and LA according to the control cycle of the operated web-based IS. The functor FL L D Ch establishes a connection between the supercategories LCh and LD to describe situations when the operation of the web-based IS is carried out according to the developed plans and there is no need to adjust these plans. To detail the formal description of the supercategories LP, LD, LCh and LA, the category of description of the states of the operated IS LSt (1.2) was divided into two subcategories: the subcategory of description of the planned state of the operated IS LSt P and the subcategory of description of the current state of the operated IS LSt C . The LSt P subcategory has the form[6, 27] L Ob Ob Ob Ob Ob Ob HH St PConf PL CI Tr St Pl St Ob Ob Ob Conf PL C ,,,,,,, __ _ oonf CI Conf Tr Tr CI St Conf Pl St St Ob Ob Ob Ob Ob Ob Ob Ob Ob HHHHH , ,, ,, _ 1 OOb , (1.6) and the subcategory LSt C has the form[6, 27] L Ob Ob Ob Ob Ob Ob HH St CConf PL CI Tr St CStOb Ob Ob Conf PL Co ,,,,,,, __ _ nnf CI Conf Tr Tr CI St Conf CSt St Ob Ob Ob Ob Ob Ob Ob Ob Ob Ob HHHHH , ,, ,, _1 . (1.7) Then, taking into account the previously developed categorical descriptions of the exploited ISs, the LP supercategory is proposed to be described as follows[6, 27] LL LL FF PISSt P IndL L L L St P IS Ind St P ,, ,, , (1.8) where F L L St P IS – a single covariant functor that establishes a connection between the LIS categories LIS and LSt P ; FL L Ind St P – a single covariant functor that establishes a connection between the categories LSt P and LInd. The LD supercategory is proposed to be described as follows[6, 27] LL LL FF DISSt C IndL L L L St C IS Ind St C ,, ,, , (1.9)
24 Management of a modern IT company: theoretical and technological aspects where F L L St C IS – a single covariant functor that establishes a connection between the LIS categories LIS and LSt P ; FL L Ind St P – a single covariant functor that establishes a connection between the supercategories and LInd. The LCh supercategory is proposed to be described as follows[6, 27] LL LLL FFFF Ch IS St P St C IndL L L L L L L L St P IS St C IS Ind St P Ind S ,,, ,,,,tt C . (1.10) The LA supercategory is proposed to be described as follows[27] LL LF AISEff L L Eff IS ,, , (1.11) where F L L Eff IS – a single covariant functor that establishes a connection between the supercategories LIS and LEff. The diagram of the relationships between the elements of the generalized model of effective management of the operation of web-based IS (1.5) is shown in Fig.1.4[6]. Fig. 1.4 Diagram of relationships between elements of a generalized model of effective management of the operation of a web-based information system LP LDLCh LA P D L L F A P L L F Ch A L L F D Ch L L F Ch D L L F This diagram establishes transitions between the main states of the cycle "Plan– Do– Check– Act" of effective management of the operation of a web-based IS. In this diagram, states are defined as supercategories LP, LD, LCh and LA, which describe the corresponding stages of the mentioned cycle. Transitions between these
25 Basics of the effective management methodology of web-based information systems Chapter 1 states are shown as functors FL L D P, FL L Ch D, FL L D Ch , FL L A Ch and FL L P A. Thus, model (1.5) can be represented as a finite state machine, transitions between the states of which are associated with changing the alphabets of the description of these states, although they retain their structural features. 1.4.3 Results of the development of a formal statement of the problem ofeffective management of the operation of a web-based information system The use of the developed set of theoretical and categorical models (1.1)–(1.11) allowed to formally describe the global problem of managing the operation of a webbased IS as an IT service system. In doing so, it is possible to proceed from the following assumption. Assumption5: the main approach to managing an IT service system is the "Plan– Do– Check– Act" cycle. As noted above, classical management of the operation of an IT service system is management aimed at minimizing the number of incident reports that remain unresolved. Therefore, it is proposed to consider the target state of the operated IT service system as a state in which the number of incident reports and RFC of the managed IT service system tends to 0. Let's call such a state of the operated IS as an IT service system stable. The aspiration of the operated web-based IS to a stable state is proposed to be described by an objective function, which has the form[6] () () , ,, Tr RFC ii njjm 11 0 (1.12) where Tri– the transaction, the execution of which generates the i-th incident; n is the number of incidents that occur during the operation of the web-based IS; RFCj– the j-th RFC; m – the number of RFCs that occur during the operation of the web-based IS. The objective function (1.12) of the classical (permanent) LC management r of the operation of a web-based IS is limited by the following conditions[6]: –for each incident during its elimination, the planned and current states in which the operated IS may be during the control of its operation must be determined Tr Ob LL LLLL iTrISD St P Ch St C Ch ,; (1.13)
26 Management of a modern IT company: theoretical and technological aspects –for each RFC during its implementation, the planned and current states in which the operated IS may be during the control of its operation must be determined RFCObLLLLL L iRFC IS DSt P Ch St C Ch ,; (1.14) –the current state of the IS in the process of controlling its operation must fully correspond to its planned state (based on the features of the "Plan– Do– Check– Act" cycle and model (1.5)) ()().LLLL LLLL St P Ch St C Ch St C Ch St P Ch 0 (1.15) Such a formulation of the task of managing the operation of a web-based IS will allow obtaining an IS that will be best adapted to the requirements of its users. But such adaptability does not guarantee effective operation of the IS, because the set of requirements and RFC itself may not be aimed at achieving effective planned and current states of the operated IS. Therefore, an attempt to achieve goal (1.12) may lead to an increase in RFC. This increase is due to the fact that the number of RFCs that will arise during the operation of the IS can increase only because previous solutions to the management task will not give the desired coincidence of planned and current states. This situation will be especially pronounced in cases where the business processes of the automation object, the IS itself and the IT infrastructure in which the IS is operated are in a state of constant change. Unlike the considered concept of classical (constant) management, the proposed concept of effective management of the operation of web-based IS uses an additional concept of "system property"[31]. At the same time, the interest for the Consumer and the Supplier of IS is only those properties of the system that characterize the manifestations of individual qualities of the web-based IS. The qualities of the web-based IS, in turn, are determined by a set of requirements (functional and non-functional) that are imposed on the IS before the start of its operation, as well as a set of RFCs of the IS and its individual CIs formed during the operation of the IS. In this case, effective management of the operation of the web-based IS will be defined as actions to achieve the planned results for each individual property of the IS and its CIs. Then the task of achieving the global goal of effective management of the operation of the web-based IS will be considered as the task of achieving optimal characteristics of this IS for each of its specific properties and the minimum probability of the existence of unsolved incidents and RFCs during the operation stage of this IS. The formal description of the objective function of this task will be as follows[6]
33 Basics of the effective management methodology of web-based information systems Chapter 1 Then, it is proposed to represent the basic attributive model of process attributes used during measurement by a tuple that has the form MP ANamePAVal PA DT As PAM B jm ls lj = = = 11 ,..., ;, ..., _,_,__ , (1.31) where PA Name PA ValPADTAs l j _,_,__ – a tuple of attributes that describe the j-th attribute of the process used in measurement; PA_Name– an attribute that describes the name of the process attribute used in the measurement; PA ValPADTAsl _,__ –atuple of attributes that describe the result of the l-th measurement of the value of the j-th process attribute; PA_Val– an attribute that describes the value of the j-th process attribute, which is the result of the l-th measurement; PA_DT_As– an attribute that describes the date and time of the l-th measurement of the j-th process attribute; s– the number of measurements planned within the implementation of the PDCA cycle iteration. The basic attributive model of a transaction as a IS CI, the state of which is measured, is proposed to be represented by a tuple that has the form MT rNameTrDTAsCINam Tr BAs tt cc As actact , ,..., ;,..., _,__,_= = = 11 eeAs ct , (1.32) where Tr Name Tr DT As CI Name As As ct _,__,_ – a tuple of attributes describing the t-th transaction as a IS CI participating in the measurement; Tr_NameAs– an attribute describing the name of the transaction participating in the measurement; Tr_DT_As– an attribute describing the date and time of the transaction's participation in the measurement; CI NameAs c _ – a tuple of attributes describing the list of IS CIs as elements of the transaction participating in the measurement; CI_NameAs– an attribute describing the name of the IS CI as an element of the transaction participating in the measurement; c– a designation of the IS CI that is an element of the t-th transaction; cact– the maximum number of IS CIs that make up the t-th transaction on the date and time of the measurement; t– a designation of the transaction as a IS CI participating in the measurement; tact– the maximum number of transactions as CIs that make up the IS configuration option that is current on the date and time of the measurement. The basic attribute model of indicators used to measure the status of a transaction as an IS CI is proposed to be represented by a tuple, which has the form M CI Name IndNameInd DT MInd V Ind BAs qI nd As As act , ,..., _,_,__,_ = =1 aal q, (1.33) where CI Name IndNameInd DT MInd Val As As q _,_,__,_ – a tuple of attributes that describe the q-th indicator used to measure the status of a transaction as a IS CI;
34 Management of a modern IT company: theoretical and technological aspects CI_NameAs– an attribute that describes the name of the IS CI (transaction or its individual element) participating in the measurement; Ind_NameAs– an attribute that describes the name of the q-th indicator used to measure the CI status; Ind_DT_M– an attribute that describes the date and time of measurement of the value of the q-th indicator; Indact– an attribute that describes the value of the q-th indicator as the result of the measurement, Indact– the maximum number of indicators used to measure the status of a transaction as a IS CI. Based on the representations of the set of objects of the LProp category (1.22), (1.24), (1.26) and (1.30), it is proposed to represent the set of morphisms of the LProp category as a collection of the following subsets MorHHH H op PG AbLPAM AM PMF Pr _ ,, , , (1.34) where HP_G– a subset of morphisms that determine the subordination of descriptions of the process measurement scheme and model by the property of the IS to descriptions of the general characteristics of this property; HAbL– a subset of morphisms that determine the subordination of descriptions of process attributes and rating scales to descriptions of the levels of capabilities of the IS property; HPAM– a subset of morphisms that determine the subordination of descriptions of transactions as IS CIs, the state of which is measured, and indicators of these transactions to descriptions of process attributes used during measurement; HAM PMF – a subset of morphisms that determine the subordination of descriptions of the process measurement model to descriptions of the process measurement scheme. The HP_G subset of morphisms generally consists of the following morphisms HHH generalPMF PG AM PG __ , , (1.35) where HPMF PG_ – a morphism that determines the hierarchy of subordination of descriptions of the process measurement scheme by the IS property to descriptions of the general characteristics of this property; HAM PG_ – a morphism that defines the hierarchy of subordination of descriptions of the process measurement model by the IS property to descriptions of the general characteristics of this property. HAbL morphisms subset in the general case consists of the following morphisms HHH AbLPA AbL RS AbL , , (1.36) where HPA AbL – a morphism that defines the hierarchy of subordination of descriptions of process attributes to descriptions of the corresponding capability level;
35 Basics of the effective management methodology of web-based information systems Chapter 1 HRS AbL – a morphism that defines the hierarchy of subordination of descriptions of the rating scale to descriptions of the corresponding capability level. HPAM morphisms subset in the general case consists of the following morphisms HHH PAMTr PAM Ind PAM , , (1.37) where HTr PAM – a morphism that defines the hierarchy of subordination of descriptions of transactions as IS CIs, the state of which is measured, to descriptions of process attributes used during measurement; HInd PAM – a morphism that defines the hierarchy of subordination of descriptions of indicators used to measure the state of a transaction as IS CI descriptions of process attributes used during measurement. HAM PMF morphisms subset in the general case consists of a HPAM PA morphism that defines the subordination of descriptions of process attributes used during measurement to descriptions of process attributes of the measurement scheme. A diagram of the structure of the category-theoretic model LProp, which describes the "Property" concept, is shown in Fig.1.5. Fig. 1.5 Structure diagram of the theoretical-categorical model LProp, which describes the concept of "Property" ObAbL ObPA ObPMF ObP_G ObAM ObPAM LProp ObRS As Tr Ob As Ind Ob PAM Tr H PAM Ind H _ P G PMF H _ P G AM H AbL PA H AbL RS H PA PAM H Modern standards and IT are focused on solving the problem of classical (permanent) management of IS operations, in which deviations between the planned and
36 Management of a modern IT company: theoretical and technological aspects current values of state indicators of individual IS CIs are measured. The application of the LProp model in the process of solving the problem of effective management of the operation of web-based IS made it possible to formulate and formally solve the following variants of this problem: –the problem of minimizing deviations in the values of the levels of capabilities of IS transactions at any of the levels of representation of this IS (the problem of optimizing the levels of maturity of IS and its elements); –the problem of minimizing deviations in the values of the attributes of IS transaction processes at any of the levels of capabilities and levels of representation of this IS (the problem of optimizing the states of IS and its elements within the selected maturity levels); –the task of minimizing deviations in the values of IS transaction indicators and their CI within individual process attributes at any of the levels of capabilities and levels of representation of this IS (the task of classical (constant) IS operation management). 1.6 Discussion of the research results In the process of the research, a concept of effective management of the LC of an operated web-based IS was proposed. The proposed concept consists in a multi-level representation of a web-based IS as a product with a set of individual properties. In this case, each property of the IS is considered as a manifestation of the individual qualities of this system. Each quality, in turn, is proposed to be considered as the result of fulfilling a set of requirements and RFCs of the IS and its SI. This concept allows to formally describe the task of effective management of the LC of an operated IS as a task of achieving optimal characteristics of this IS for each of its specific properties and the minimum probability of the existence of unresolved incidents and RFCs during the operation stage of this IS. For a formal description of the proposed concept, a generalized model of effective management of the operation of web-based IS (1.5)–(1.11) and a complex of theoretical and categorical models (1.1)–(1.4) were developed, which detail the description of the elements of this model. The use of the apparatus of category theory allows to further present the generalized model (1.5)–(1.11) as a finite state machine, the transitions between the states of which are associated with the change of the alphabets of the description of these states, although they retain their structural features. Such a representation allows to consider the developed models (1.1)–(1.11) in further studies as the basis for a formal description of the
37 Basics of the effective management methodology of web-based information systems Chapter 1 automated effective management of the operation of web-based IS based on the "Plan– Do– Check– Act" cycle. This description can exist provided that it is possible to prove the possibility of transforming the corresponding finite state machine into a consistent set of simpler finite state machines that describe IT processing of data structures at each stage of the cycle[6]. Based on the developed generalized model (1.5)–(1.11), a formal description (1.12) of the purpose of classical (permanent) LC management of the operated IS and limitations (1.13)–(1.15) was developed. Taking into account these formal descriptions and the results of the development of the concept of effective LC management of an operated web-based IS, a formal description of the objective function of effective management of the LC of an operated IS (1.16) and systems of constraints (1.17)–(1.20) was developed. The obtained results allow to define the problem of classical (permanent) LC management of an operated IS (1.21) as a special case of the problem of effective management. To clarify the formal statement of the problem of effective LC management of an operated web-based IS, a theoretical-categorical model of the properties of an operated web-based IS (1.22)–(1.37) was developed. The use of the developed theoretical-categorical model LProp allowed to significantly change the features of the statement and solution of the problem of effective management of the operation of a web-based IS. Solving this problem using the developed model (1.22)–(1.37) will allow not only to maintain the operated IS in a stable state, but also to determine the optimal levels of maturity of the IS and its individual transactions from the point of view of the Consumer and the Provider of IT services. This allows stakeholders to plan and manage the recognition of the feasibility and implementation of the following measures during the operation of a web-based IS: –standardization (typification) of individual IS transactions and IS as a whole at any of the representation levels; –to ensure the manageability of individual transactions of the IS and the IS as awhole (in real time or time close to real); –to ensure the optimal operation of individual transactions of the IS and the IS as a whole. It should be noted that a significant number of modern studies in the field of PLM, ALM and SLM use the mathematical apparatus of graph theory for the formal description of the tasks to be solved. Examples of such studies are[40,41]. Incontrast, in this study, the mathematical apparatus of category theory was used for the formal description of the tasks of LC management of the operated IS. The theoretical and categorical models of the operated IS and the tasks of LC management of this IS used and developed in this study are based on the ontology of web-based
38 Management of a modern IT company: theoretical and technological aspects IS services management proposed by one of the authors of the study in[27] and allow to establish: –the boundaries of domains and data structures used to describe individual concepts of the ontology; –the rules for transforming domains and data structures when creating connections between different concepts of the ontology. The applied application of the models developed in this study makes it possible to transition to effective LC management of an operated IS without global reengineering of systems and technologies for data storage and processing. The vast majority of IT products for LC management of an operated IS are highly specialized and relatively new products that occupy a small market segment. Therefore, systems for LC management of an operated IS are most often the result of integrating a large number of such highly specialized products. The proposed theoretical and categorical models allow in such cases[6]: –to determine domains and data structures for technologies for integrating heterogeneous IT products into a single system; –to ensure the transition from classical to effective LC management of an ITservice system without serious changes in the list of highly specialized products inoperation. Also important from an applied point of view is the proposed presentation of the existing problem of classical (permanent) LC management of an operated IS as a separate case of the problem of effective management. This representation allows to consider the design and implementation of an effective LC management system of operated IS as a gradual evolutionary development of the previously created and already operating system for the classical LC management of IT services. The main limitation of this study is the need to use complete descriptions of all elements of this IS to solve the problem of effective LC management of an operatedIS. This means that during at least one iteration of the "Plan– Do– Check– Act" cycle, IS administrators must interact with a complete digital model of this IS at all levels of representation to solve individual tasks of LC management. The development and maintenance of such a model is a complex and costly task. But this limitation is methodological in nature and is valid for any work in the field of change management of developed or operated IS (which is confirmed, for example, by studies[42,43]). Another feature of the obtained results is their orientation on the LC management of the operated IS in conditions of constant changes in business processes, functions and elements of IT infrastructure. This feature is a consequence of the generalization of applied experience in supporting and ensuring the functioning of IS management of enterprises and organizations in conditions of martial law, which is in force in Ukraine[6].
39 Basics of the effective management methodology of web-based information systems Chapter 1 Although the obtained research results are methodological in nature, they should be recognized as important for further theoretical and applied research in the field of creating PLM-, ALMand SLM-IS and IT. The main further directions of development of this research are[6]: –determination of the set of properties of the operated IS and efficiency indicators for these properties; –development of new and improvement of existing models, detailing the formal description of the task of effective management; –development and implementation of methods and algorithms for solving problems of effective management of various IS and IT options. 1.7 Conclusions The concept of the task of effective LC management of an operated IS as an IT service system is formulated. The formulated concept allows to formally describe the task of effective LC management of an operated IS as the task of achieving optimal characteristics of this IS for each of its specific properties and the minimum probability of the existence of unresolved incidents and RFCs during the operation stage of this IS. The result obtained takes into account the points of view of the Supplier and Consumer of IT services as the main stakeholders of web-based IS. During further research, the main attention was paid to the development of elements of the methodology for effective management of the operation of webbased IS precisely from the point of view of the Supplier as an IT company that provides (sells or leases) its own web-based IS for operation by enterprises or organizations acting as Consumers. A generalized model of effective management of web-based IS operation (1.5)–(1.11) and a set of theoretical and categorical models (1.1)–(1.4) that detail the description of the elements of this model have been developed. The developed models are a formal basis for creating IS and IT for effective management of webbased IS operation without global reengineering of systems and technologies for data storage and processing. Unlike existing models of a similar purpose, the developed models allow establishing the boundaries of domains and data structures used to describe the elements of the generalized model (1.5)–(1.11), as well as the rules for transforming domains and data structures when creating connections between different elements of this model. A formal statement of the problem of classical (permanent) LC management of operated IS as an IT service system has been developed. A formal description of
40 Management of a modern IT company: theoretical and technological aspects the objective function (1.12) and constraints (1.13)–(1.15) of this problem is proposed using the apparatus of category theory and set theory. The developed formal formulation is based on the existing definition of the goal of such management as maximizing the satisfaction of IS users based on the results of each iteration of the "Plan– Do– Check– Act" cycle of LC management of the operated IS. It is proposed to consider the IS state as a formal sign of such satisfaction as the number of incident reports and RFCs of the operated IS tending to 0. The results obtained allow to formally describe the task of LC management of the operated IS in conditions of stable(unchangeable) business processes and IT infrastructure of enterprises and organizations as objects of automation. A formal description of the objective function (1.16) and constraints (1.17)–(1.20) of the task of effective LC management of the operated IS is proposed using the apparatus of category theory, set theory and elements of probability theory. It is determined that the task of classical (permanent) LC management of the operated IS is a special case of the proposed task of effective management. Practical application of the proposed formal description of the task of effective LC management of an operated IS allows improving SLM-systems for LC management of an operated IS without global reengineering of existing systems and technologies for data storage and processing. A theoretical and categorical model of the IS property (1.22)–(1.37) has been developed. This model allows formally describing any IS property taking into account the current requirements set by modern standards for the implementation of the processes of assessing the LC of the system. Using the developed model (1.22)–(1.37) allows solving tasks for automating effective planning and management of the operation of web-based IS. Conflict of interest statement The authors declare that there is no conflict of interest in relation to this paper, as well as the published research results, including the financial aspects of conducting the research, obtaining and using its results, as well as any non-financial personal relationships. Use of artificial intelligence statement The authors declare that they did not use artificial intelligence tools in preparing this manuscript.
41 Basics of the effective management methodology of web-based information systems Chapter 1 References 1. What is Web (Based) Information System. IGI Global. Available at: https:// www.igi-global.com/dictionary/web-based-information-system/32189#:~: text=An%20information%20system%20that%20utilizes,or%20other%20information%20systems%2Fapplications 2. Fulton, J. (2017). Web Architecture 101. Medium. Available at: https://medium. com/storyblocks-engineering/web-architecture-101-a3224e126947 3. ISO/IEC 20000-1. Information technology – Service management – Part 1: Service management system requirements (2018). Geneva: ISO Copyright Office, 96. 4. ISO/IEC/IEEE Standard No 15288:2015 (2015). Systems and software engineering – System life cycle processes. ISO/IEC/IEEE International Standard. https://doi.org/10.1109/IEEESTD.2015.7106435 5. Levykin, V. M., Evlanov, M. V., Kernosov, M. A. (2014). Patterny proektirovaniia trebovanii k informatcionnym sistemam: modelirovanie i primenenie. Kharkiv: OOO "Kompaniia "Smit", 320. 6. Levykin, V., Ievlanov, M., Levykin, I., Petrychenko, O. (2025). Development of a concept for the task of life cycle effective management of an operated information system. Technology Audit and Production Reserves, 2 (2 (82)), 66–73. https://doi.org/10.15587/2706-5448.2025.326479 7. Stark, J. (2020). Product Lifecycle Management (Vol.1). Cham: Springer International. https://doi.org/10.1007/978-3-030-28864-8 8. Schwaber, C. (2006). The Changing Face of Application Life-Cycle Management. Forrester Research Inc. Available at: https://www.yumpu.com/en/document/ view/13866040/download-the-changing-face-of-application-life-cycle-mks 9. Wyrwich, F., Kharatyan, A., Dumitrescu, R. (2024). Interdisciplinary system lifecycle management – a systematic literature review. Proceedings of the Design Society, 4, 2765–2774. https://doi.org/10.1017/pds.2024.280 10. Rizzo, S. (2016). Why ALM and PLM need each other. Siemens Whitepaper. Available at: https://polarion.plm.automation.siemens.com/hubfs/Docs/Whitepapers/why-alm-and-plm-need-each-other-whitepaper.pdf 11. Liepert, C., Stary, C., Lamprecht, A., Zügn, D.; Elstermann, M., Lederer, M. (Eds.) (2025). Interoperable Product Change Management Within Engineering: A Digital Twin Approach. Subject-Oriented Business Process Management. Models for Designing Digital Transformations. S-BPM ONE 2024. Communications in Computer and Information Science. Vol. 2206. Cham: Springer. https:// doi.org/10.1007/978-3-031-72041-3_17
42 Management of a modern IT company: theoretical and technological aspects 12. Chappell, D. (2010). What is Application Lifecycle Management? David Chappell and Associates. Available at: http://davidchappell.com/writing/white_papers/ What_is_ALM_v2.0--Chappell.pdf 13. Eigner, M. (2021). System Lifecycle Management: Digitalisierung des Engineering. Berlin, Heidelberg: Springer Vieweg. https://doi.org/10.1007/978-3-66262183-7 14. Binder, C., Neureiter, C., Lüder, A. (2022). Towards a domain-specific information architecture enabling the investigation and optimization of flexible production systems by utilizing artificial intelligence. The International Journal of Advanced Manufacturing Technology, 123 (1-2), 49–81. https://doi.org/10.1007/s00170022-10141-2 15. Colantoni, A., Berardinelli, L., Garmendia, A., Bräuer, J. (2022). Towards Blended Modeling and Simulation of DevOps Processes: the Keptn Case Study. MODELS'22: Proceedings of the 25th International Conference on Model Driven Engineering Languages and Systems: Companion Proceedings, Association for Computing Machinery. New York, 784–792. https://doi.org/10.1145/3550356.3561597 16. Gulzar, K., Ruusu, R., Sierla, S., Aarnio, P., Karhela, T., Vyatkin, V. (2018). Automatic Generation of a Lifecycle Analysis Model from a First Principles Industrial Process Simulation Model. 2018 IEEE 16th International Conference on Industrial Informatics (INDIN). Danvers, 741–746. https://doi.org/10.1109/ indin.2018.8471980 17. Calderon, N. N., Kajko-Mattsson, M., Nolan, A. J. (2015). Successful process improvement projects are no accidents. Journal of Software: Evolution and Process, 27 (11), 896–911. https://doi.org/10.1002/smr.1738 18. Reiff-Marganiec, S., Tilly, M. (Eds.) (2012). Handbook of Research on Service-Oriented Systems and Non-Functional Properties: Future Directions. IGI Global. https://doi.org/10.4018/978-1-61350-432-1 19. Driss, M., Aljehani, A., Boulila, W., Ghandorh, H., Al-Sarem, M. (2020). Servicing Your Requirements: An FCA and RCA-Driven Approach for Semantic Web Services Composition. IEEE Access, 8, 59326–59339. https://doi.org/10.1109/ access.2020.2982592 20. Kienzle, J., Combemale, B., Mussbacher, G., Alam, O., Bordeleau, F., Burgueno, L. et al. (2022). Global Decision Making Over Deep Variability in Feedback-Driven Software Development. Proceedings of the 37th IEEE/ACM International Conference on Automated Software Engineering. New York. https://doi.org/ 10.1145/3551349.3559551 21. Moosbauer, J., Binder, M., Schneider, L., Pfisterer, F., Becker, M., Lang, M. et al. (2022). Automated Benchmark-Driven Design and Explanation of
49 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 of the methodology for optimizing adaptive project management using network economic and mathematical modeling and the feedback principle. The use of this methodology allows taking into account the incompleteness of information associated with the failure or delay of specific project operations. In[9], a review of solutions for computer-aided decision support for offshore software outsourcing in the global context of the development of such systems is provided. In particular, in[9] ten main key factors related to the decision-making process in the IT industry were identified, namely: human communication, cost reduction, organizational and professional maturity, project management methods, IT infrastructure support, language restrictions, knowledge-based support, changes in requirements, legal issues and cultural diversity. However, the data obtained in[9] show that the software industry lacks effective and efficient decision-making models that take into account the specifics of the IT industry and, in particular, the features of management in IT projects. The conclusion about the lack of application of effective and efficient models of decision-making and decision-making in IT project management is also confirmed by the results of other studies. Thus, in[10] an approach to decision analysis is proposed, which allows for the implementation of structured, reproducible and group execution of relevant procedures using a decision support system. However, this approach, as shown in[10], has only a methodological basis. In[11], a study was conducted to support decision-making on the implementation of test automation in the context of Agile-based Software Development. In total, the study identified twenty-one factors that significantly affect test automation in the specified context. But the final result of applying the identified factors in[11] is only a conceptual model designed to help managers practicing Agile make decisions on the implementation of test automation in the specified context. The results of the analysis allow to conclude that it is appropriate to conduct scientific and applied research in the field of developing new and improving existing formal models and decision-making methods for IT project management. These studies, taking into account the results obtained in[4], are proposed to be based on the following assumptions: –the greatest effect from the application of these models and methods should be expected when making decisions on current and operational management of work at the early stages of the project; –it is desirable to use such models and decision-making methods from various aspects of project management that would allow to assess the impact of the studied alternatives of the decision being formed and adopted on the overall assessment of
50 Management of a modern IT company: theoretical and technological aspects the project cost and, accordingly, on the estimates of the costs of performing the work of this project. These assumptions made it possible to determine that the purpose of this study is to develop loss models when designing a description of the elements of the enterprise management information system (IS) in conditions of incompleteness or lack of complete information about these elements. The use of these models will allow formalizing the procedure for choosing a rational option for using personnel and IT infrastructure of an IT company during the design or implementation of IS elements, the descriptions of which are characterized by incompleteinformation. To achieve this aim, it is proposed to solve the following research objectives: –adapt the situational-consequential scheme of production losses due to lack of information to the features of the design processes of IS elements; –develop and implement detailed loss models for the main consequences of production losses and decision-making on managing iterations of the implementation of the IT project of creating IS; –perform experimental verification of the results obtained. 2.3 Research models To determine the mechanism of loss formation during the design and implementation of IS elements, it is proposed to apply the existing causal scheme of loss formation in production due to the lack of information (Fig.2.1)[12]. It should be noted that causal schemes are considered by some modern researchers as one of the means of overcoming uncertainty in project management[13]. In Fig.2.1, the following designations are adopted: Ex1 – "Replacing missing information with other, for example, scattered or indirect" consequence; Ex2 – "Concentration of production resources on performing work not for their direct purpose" consequence; Ex3 – "Transferring production resources to performing work that corresponds to their direct purpose" consequence; Ex4 – "Refusal to use re sources due to the lack of information for making management decisions" consequence; Ex5 – "Any measures to maintain the required duration of the production cycle are not taken" consequence; Ex6 – "Intensification of the production process after the elimination of uncertainty" consequence; P1–P8 and P′ 9–P′ 13 – losses characterizing the corresponding consequences; Psum – total losses of the consequence Ex1; W1–W3 – penalties for failure to complete the planned work on time. A detailed description of these elements is given in[12].
51 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 The situations in the result of a lack of information Situations after receiving information P1P′ 9P′ 10 P′ 11 P′ 12 P′ 13 W1W2W3W1W2W3 Psum P2P3P4P5P6P7P8 Ex2 Ex1Ex3Ex4Ex5Ex6 Fig. 2.1 Cause-and-effect scheme of production losses due to lack of information The occurrence of a situation where there is no information can lead to one of the consequences Ex1, Ex2, Ex1 and Ex4. The consequence Ex1 attracts the most attention among them, because it allows replacing the description of an IS element for which complete information is missing with reusable descriptions of other IS elements being designed or other IS. The loss models P1 and P2, which characterize the consequence Ex1, have the form[12]: PSRHk add 1 , (2.1) PSR Hk 2 , (2.2) where P1 – losses from the absence of information in the event of the need for further processing with replacement information; Sadd – the average cost of additional processing of replacement information; R(H) – the average number of occurrences of situations of absence of information for the analyzed period of time T with a certain composition of information H={h1,h2,…,hn}; k – the coefficient of quality of management decision-making; P2 – losses from the absence of information in the presence of a difference in the cost of information being replaced and information being replaced; DS – the average difference in the cost of information being replaced and information being replaced. In some cases, as shown in the scheme (Fig.2.1), these losses are summed up.
52 Management of a modern IT company: theoretical and technological aspects According to the result of Ex2, production resources are concentrated on performing work not for their direct purpose. For this case, the loss models P3 and P4, which characterize the consequence of Ex2, have the form[12] PSltHRHk res t indef av st 31 ()() (2.3) provided tHt indefc ()≥ ; PSltHRHk res t c av st 41 ()() (2.4) provided that tHt indefc ()< , where P3, P4 – losses from uncertainty of the situation due to the lack of management information when transferring resources to perform work not for their direct purpose; b – coefficient of losses on the cost of resources when they perform work not for their direct purpose; Sres t av st – average cost of a resource unit for a normalized period of time tst; l – number of conditional units of resources; tH indef() – average time of uncertainty of the situation with a certain information composition H; tc – average cycle of performing a normalized unit of work; k1 – coefficient of quality of work performance. The consequence Ex3 is characterized by the transfer of production re sources to perform work that corresponds to their direct purpose, which is possible in the case when the production process has additional components that ensure the employment of resources for a time not less than the time of uncertainty of the situation. If, to eliminate forced downtime of resources, the work is selected from arandom set of unfinished work (if such is always available) and at the same time the creation of any special reserve is not required, then the losses can be calculated by the formula[12] PRHtS tH Tt WS av nc windef c su mu nfin w 5 () () .. ., (2.5) where t – the time period under consideration; Sav nc w.. – the average cost of an unperformed unit of work as a result of transferring resources; Wsum – the total value of penalties for failure to complete scheduled work on time as a result of transferring resources to reserve work; Sunfinw . – the cost of completing unfinished work. If a special reserve of work is created, then the losses are[12] PRHtS tH Tt WS S av nc windef c sumreserve complw6 () () .. ..rreserve, (2.6)
53 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 where Sreserve – the cost of creating a special reserve of work; Scomplwreserve .. – the cost of the completed reserve of work. Consequence Ex4 is a situation when, due to the lack of information for making management decisions, resources are not used. Losses in this case are proposed to be estimated in[12] as follows PSltHRH res t indef av st 7 ()() (2.7) provided tHt indefc ()≥ ; PSltRH res t c av st 8 () (2.8) provided tHt indefc ()< . After receiving the information necessary for making management decision information, consequences Ex5 and Ex6 arise. They are the results of consequences Ex2, Ex3 and Ex4. Let's define consequence Ex5 as a consequence in which no measures are taken to maintain the required production cycle duration. Naturally, in[12] it is assumed in this case that tc will increase by tindef(H) from the moment of time at which the cycle was interrupted. For the consequence Ex5, if the lack of information does not lead to an increase in the volume of work in progress, the losses are estimated as follows[12] P Wsum9 . (2.9) In the case of an increase in the volume of work in progress, losses are determined as follows[12] PSthRH T W unfinindef su m av 10 ()() , (2.10) where Sunfinav – the average cost of work in progress. The losses associated with the penalty for the consequence Ex5, in the case of an increase in tc, may have different values. If the increase in tc does not lead to an increase in the critical period, W1=0; if tc increases and exceeds the critical period by an amount not greater than that specified in the relevant regulatory documents, then WSRH unfin compl av 2 () , (2.11)
54 Management of a modern IT company: theoretical and technological aspects and in the case of an excess of tc over the critical period by an amount greater than that specified in the standards, WSRH unfin compl av 31 () , (2.12) where α , α 1 – the corresponding penalty coefficients for the specified cases; Sunfin compl av – the average total cost of work in progress. The consequence Ex6 is characterized by the fact that the production process after the elimination of uncertainty intensifies. If the increase in tc due to the uncertainty of the situation can be eliminated without additional costs, for example, by reducing the inter-operational lag, the losses are zero ( P 11 0 ). A possible option is when, in order to reduce tc, the process is intensified by overtime use of resources[12]. Then, if tt H indef() , losses PSltHRH res t indef av st 12 ()() , where γ – the coefficient that increases the cost of resources in the case of their use in overtime; Dt – the maximum possible reduction in the cycle duration. When[12] P SltR H SRHt Ht T W res tunfinindef su m av st av 13 () ()(()) . (2.13) The losses due to penalties when increasing tc are determined by formulas (2.11), (2.12)[13]. The scheme considered in Fig.2.1 was developed for manufacturing enterprises and does not take into account the peculiarities of the processes of designing and implementing IS elements. In addition, the loss and penalty models (2.1)–(2.13) are not detailed, which makes their application in managing an IT project for creating or improving IS much more difficult. 2.4 Results of adapting the causal scheme of production losses to the features of the design processes of an information system element In the course of adapting the causal scheme of production losses due to lack of information[12], it was decided to take into account the following features of the design and implementation processes of IS elements in an IT company: –each specific IS corresponds to a separate IT project, which is performed by ITcompany employees within a predetermined time frame; –an IT company can simultaneously perform several different IT projects;
55 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 –management of IS element development teams is carried out iteratively based on Agile IT project management methodologies; –the description of each IS element being created is performed by one or more developers using a uniform structured template of this description H; –in the process of forming descriptions of individual IS elements, developers strive to achieve the maximum possible reuse of descriptions of existing elements of similar IS; –the content of the task backlog for the development team as a set of IS element descriptions and even the duration T of each iteration (sprint, etc.) are set at the beginning of execution; –the content of the task backlog rarely changes during the iteration being performed. Here and now, the developers of IS elements will be understood as IT service providers who are employees of an IT company that takes responsibility for creating IS according to the requirements of IT service consumers[14] during all relevant stages of the IS life cycle[1]. These features allowed to formulate the definition of the consequences highlighted in the scheme as follows: –consequence Ex1: "Replacement of missing information with information about a similar IS element that is reused"; –consequence Ex2: "Concentration of IT company personnel and IT infrastructure elements on the design or development of other IS elements"; –consequence Ex3: "Transfer of IT company personnel and IT infrastructure elements to the design or development of other elements of the same IS"; –consequence Ex4: "Refusal to use the personnel and elements of the IT infrastructure of the IT company due to the lack of information about the IS elements"; –consequence Ex5: "Failure to take any measures to maintain the required duration of the iteration T"; –consequence Ex6: "Intensification of the processes of implementing the IT project of creating the IS after eliminating uncertainty". Then the elements of the generalized loss formulas (2.1) and (2.2) are proposed to be interpreted as follows: P1 – losses from the lack of information in the event of the need for additional processing with information from the description of the reused IS element; Sadd – the average cost of additional processing of information, which is replaced, from the reused description of the IS element; R(H) – the average number of occurrences of situations of lack of information for the period of time T under consideration, with a priori established template of the description of the IS element H={h1,h2,…,hn}; k – the coefficient of quality of management decision-making;
56 Management of a modern IT company: theoretical and technological aspects P2– loss from the absence of information in the presence of a difference in the value of the information being replaced and the information being replaced from the reused description of the IS element; DS – average difference in the value of the information being replaced and the information being replaced from the reused description of the IS element. The elements of the generalized loss formulas (2.3) and (2.4) are proposed to be interpreted as follows: P3, P4 – losses from the lack of management information when transferring personnel and IT infrastructure elements of an IT company to the design or development of other IS elements; b – coefficient of losses on the cost of personnel and IT infrastructure elements of an IT company when they are used during the design or development of other IS elements; Sres t av st – average cost of a unit of personnel and IT infrastructure elements of an IT company for a normalized period of time tst; l – number of conditional units of resources (personnel and IT infrastructure elements) of an IT company that are used during the design or development of other IS elements; tH indef() – average duration of the uncertainty of the description of an IS element with a priori established template of this description H={h1,h2,…,hn}; tc – average duration of one iteration of the IT project work on the design and implementation of an IS element, tc=T. The elements of the generalized loss formulas (2.5) and (2.6) are proposed to be interpreted as follows: P5 – losses from the transfer of personnel and IT infrastructure elements of the IT company to perform work on the design or development of IS elements in the case when this work is selected from a random set of unfinished work(if such is always available) and the creation of any special reserve is not required; P6 – losses from the transfer of personnel and IT infrastructure elements of the IT company to perform work on the design or development of ISelements in the case when this work is selected from a created special reserve of work; Sav nc w.. – the average cost of a unit of work not performed as a result of the transfer of personnel and IT infrastructure elements of the IT company; Wsum – the total value of penalties for failure to perform scheduled work on time as a result of the transfer of personnel and IT infrastructure elements of the IT company to reserve work. The elements of the generalized loss formulas (2.7) and (2.8) are proposed to be interpreted as follows: P7, P8 – losses from non-use of personnel and IT infrastructure elements of the IT company due to the lack of information for making management decisions. Similarly, the semantics of costs ′ P9 and ′ P 10 was clarified for the consequence Ex5 and costs ′ P 11 , ′ P 12 and ′ P 13 , and for the consequence Ex6. However, this clarification slightly changed the descriptions and interpretations of these costs.
57 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 2.5 Results of developing detailed models of losses from lack ofinformation The results of adapting the situational-consequence scheme of loss generation to the features of the design processes of IS elements, considered in Section2.4, leave open the question of a detailed method of calculating these losses. Therefore, to apply the proposed mechanism of loss generation during the design and implementation of IS elements, it is necessary to develop detailed loss models that would allow obtaining quantitative estimates of losses for the consequences Ex1–Ex6. In the process of formulating a detailed description of the loss models (2.1) and (2.2), it was established that the main problem lies in clarifying the formal description of the parameters Sadd and DS . Based on the results of developing parametric models for estimating labor costs (the modern term is efforts) for the implementation of the COCOMO II IT project[15], the following assumptions were formulated: –Assumption1: at different stages of the IS creation life cycle, different models are used to estimate efforts; –Assumption2: the need for personnel to perform the estimated work during an iteration with a duration of T can be determined by dividing the obtained effort estimate by the value of the parameter T; –Assumption3: the costs of operating IT infrastructure elements in the most common conditions of an IT company's activity can be considered constant and evenly distributed over iterations with a duration of T. These assumptions were proposed to be used as the basis for developing detailed loss models P1 and P2. In the case of using the P1(2.1) loss model to estimate losses in the case of consequence Ex1, the most difficult thing is to determine the value of the parameter Sadd. Taking into account the above-mentioned features of the design and implementation processes of IS elements, as well as the results of adapting the cause-and-effect scheme of losses in production due to the lack of information, this parameter is proposed to be described as follows S kq TStT kq TCIT addzizizi i m zj zj zj j n z R () () () () 11 1 (()H , (2.14) where kzi – the normative value of the cost of wages of a developer who is engaged in the process of creating the z-th IS element in the time period T in the i-th position; qzi(T) – the number of developers who are engaged in the process of creating the z-th IS element in the time period T in the i-th position; Stzi(T) – the operator that
58 Management of a modern IT company: theoretical and technological aspects establishes the fact that developers are in the process of creating the z-th IS element in the time period T in the i-th position; kzj – the normative value of the cost of operating the j-th IT infrastructure element in the process of creating the z-th IS element in the time period T; qzj(T) – the number of j-th IT infrastructure elements operated in the process of creating the z-th IS element in the time period T; CIzj(T) – an operator that establishes the fact of operation of the j-th IT infrastructure elements that are operated in the process of creating the z-th IS element in the time period T. Taking into account Assumption3, expression (2.14) can be written in a simplified form S kq TStT Cz addzizizi i m z RH CI z RH () () () () () 111 , (2.15) where CCI(z) – the constant value of the costs of operating IT infrastructure elements in the process of creating the z-th IS element in the time period T. Thus, it is necessary to determine the number of developers qzj(T), which is required to eliminate the situation that arose as a result of the lack of complete information about the description of the IS element in the IT project iteration of duration T, which is planned or implemented. Based on the COCOMO model, this parameter is proposed to be calculated in the following way qTPT zi z ()= , (2.16) where Pz – the effort to perform additional processing of the replacing information. When using the P2 (2.2) loss model to estimate losses in the case of the consequence Ex1, the most difficult thing is to determine the value of DS . To do this, it is first necessary to establish the peculiarity of the difference in the values of the information being replaced and the information being replaced. This peculiarity arises as a result of using the mechanism for reusing descriptions of previously developed IS elements. According to this mechanism, reuse of the description of an IS element is possible in one of the following cases: –caseI: the description was developed and implemented during one of the previous iterations; –caseII: the description is planned to be developed and implemented during the same iteration as the replaced description of the IS element, but one or more days earlier. In caseI, the value of the information value Srepm, which is replaced, is known to the developers for this element quite accurately from the reports on the progress of the previous iterations. In caseII, the developers can only operate with an estimate
65 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 S kq TStT Cz complwreserveresi resi resi i m res r CI .. () () ( 11 )) , (2.43) where res – the IS element, the description of which is decided to be selected from the backlog reserve of the development team and used instead of the incomplete description of the IS element, which cannot be implemented during the current iteration. The value of the quantity qri is set by a formula similar to (2.16). Based on formulas (2.34), (2.35), (2.38) and (2.40)–(2.43), a detailed description of the loss model P6 (2.6) for the case when the task for replacement is selected from the backlog of the development team is proposed to be presented as follows P RHtkqTSt TkqTCI Tt di di di i m dj dj dj j n 6 11 () () () () () ii ndef c projectpipipi i m pj pj H Tt tk PkqTSt TkqTC () () () () % 1 IIT kq TStT pj j n resi resi resi i m r () () () 1 1 ees r resj resj resj j n res r kq TCIT 111 () () (2.44) provided that tHT indef()≥ . Otherwise, it should be assumed that tHT indef()= . When developing detailed descriptions of costs P7 (2.7) and P8 (2.8), it was taken into account that the product S l res t av st × , which determined the total resource costs for the normalized time period tst, was already described by expressions (2.29)–(2.31). In this case, the difference is that the set of resources that were planned to be used to perform R(H) of the IS A is elements not transferred to the IS B IT project and is not used in any of the IT projects. Therefore, it was proposed to present the detailed descriptions of costs P7 (2.7) and P8 (2.8) as follows P StHRH redp d IS d RH indef A 7 1 , () () () (2.45) provided that tHT indef()≥ ; PSTRH redp d IS d RH A 8 1 , () () (2.46) provided that tHT indef()< .
66 Management of a modern IT company: theoretical and technological aspects In this case, the value of each element of the sum Sredp d IS d R H A , () 1 is calculated using expressions (2.29) or (2.30). For a detailed description of losses due to the consequences Ex5 and Ex6, which are the results of the consequences Ex2, Ex3 and Ex4, it was necessary to conduct additional research. The main goal of these studies was to study the features of planning multiple iterations of an IT project for creating an IS due to an increase or decrease in the duration of a separate iteration T. Therefore, in this study, a detailed description of losses due to the consequences Ex5 and Ex6 was not considered. 2.6 Experimental verification of the obtained results 2.6.1 Description of the initial data To verify the results obtained, it was proposed to use the data of the IT project for the development of the service "Electronic Compulsory Motor Third Party Liability Insurance Policy" (CMTPL). The sale of this type of insurance policy was officially launched in Ukraine in accordance with the changes in the legislation of February7, 2018. In this regard, the problem of developing and implementing IT services that allow automating such activities arose. The IT company ProfITsoft, a leading software development company that also has deep expertise in the insurance domain and innovations in the InsurTech industry, was engaged in solving this problem. It has been implementing IT projects for the development and maintenance of software for more than 20years. In the Ukrainian market, ProfITsoft is known as the developer of the IS "Comprehensive System for Automation of Insurance Company Work – ProfITsoft" (KSASK), which is successfully used by leading insurance companies[16]. The KSASK development began in 2006, when the functional module "Frontoffice" was introduced. The following functions were developed within this module[17]: "Calculators", "Sales network", "Forms", "Commission", "Security". In 2009, the system was supplemented with the functional module "Back-office". The following functions were developed within this module[17]: "Metadata-based interface", "Finance", "Settlement", "Reinsurance". In 2013, KSASK implemented designers – tools that allow to configure any calculations, documents and output forms without the need to modify the entire system or its individual modules. In addition, the functional modules "Underwriting" and "Import of contracts"[17] were implemented during this period.
67 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 In 2014–2019, KSASK gradually transitioned from a modular to a service architecture. In addition, the following modules were developed in the system[17]: "Online-shop" (aka "Internet store"), "Settlement designer", "Integrations" (a module that supports data exchange via API with external platforms (EWA, PrivatBank, etc.)). By 2021, KSASK had developed as follows[17]: –calculator designer was developed; –"Financial monitoring" and "Quote requests" services were developed; –mobile version of KSASK software was developed; –transition from a service to a microservice architecture of the system began. In 2022–2024, KSASK was adapted to the new requirements of the National Bank of Ukraine, which came into force in 2024 in accordance with the Law "On Insurance"(No.1909-IX). As a result of the adaptation, KSASK began to provide automation of all business processes of insurance companies, in particular, accounting of contracts, financial monitoring, formation and submission of reports in accordance with the requirements of the regulator. The system supports electronic signatures and integration with key state registers and services, ensuring reliability, trans parency and control of all operations. KSASK has already been updated to comply with new classifications of insurance products and accounting by business lines, which makes it a reliable solution for meeting the requirements of the National Bank of Ukraine[18]. At the time of the initiation of the IT project for the development of the "Electronic CMTPL Policy" service (2017), the basic version of KSASK consisted of the following functional modules: –"Insurance Calculator"; –"Accounting for Contracts"; –"Import of Data Lists"; –"Underwriting"; –"Accounting for Forms"; –"Commissions"; –"Finances"; –"Settlement"; –"VMI" (voluntary medical insurance); –"Reporting to the Motor (Transport) Insurance Bureau of Ukraine (MTBU)"; –"Online Store"; –"Administration"; –"CRM"; –"Business Processes"; –"Releases"; –"Help".
68 Management of a modern IT company: theoretical and technological aspects The service "Electronic CMTPL Policy", the development of which was the main goal of the IT project, according to the requirements set for it, was to consist of the following functions: –"Filling out an electronic policy"; –"Checking an electronic policy"; –"Issuing an electronic policy"; –"Selling an electronic policy"; –"Printing an electronic policy agreement"; –"Checking the validity period of an electronic policy"; –"Paying for an electronic policy"; –"Calculating the commission fee for an electronic policy"; –"Authorization via SMS". The emergence of a business opportunity to sell an electronic CMTPL policy has led to the emergence of a corresponding business need in insurance companies. Therefore, ProfITsoft has identified the minimization of the time for the development and implementation of this service as an additional condition for the successful implementation of the IT project for the development of the "Electronic CMTPL Policy" service. It was believed that the fulfillment of this condition would allow insurance companies-owners of KSASK to gain a temporary business advantage over competitors (until the implementation of similar services from other developers). The main way to fulfill this condition during the initiation of the project was to reuse the results of the development of other functional KSASK modules. To search for functional KSASK modules and their functions that can be reused in the IT project for the development of the "Electronic CMTPL Policy" service, the information technology of the IT service provision management system was used[19]. The theoretical foundations and features of the implementation of this technology are discussed in detail in[20]. At the moment, this technology in the IT company ProfITsoft has received further development and has become a specialized mechanism "Functional Exchange Fund". This mechanism of interaction between insurance companies-users of KSASK provides the possibility of transferring, exchanging, reusing and collectively developing the modified functionality of the system[21]. The results of the search for functional modules and individual functions of KSASK suitable for reuse in the IT project for the development of the "Electronic CMTPL Policy" service are given in Table2.1. As a result of the application of information technology for managing the IT service provision system during the initiation of the IT project, an assessment of the profitability of independent service development or reuse of one of the KSASK IT service search results (functions) listed in Table2.1 was carried out.
69 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 The ISO/IEC1926:2001 criteria system was used for the assessment[22]. The assessment was carried out according to the IT service selection method proposed in[20], which is based on the hierarchy analysis method. Table 2.1 Results of the search for functional modules and individual functions of KSASKsuitable for reuse in the IT project for the development of the "Electronic CMTPL Policy" service Search object IT service search result Search result for the name/part of the name that matches, IT service function Full match with the list of functional requirements No IT services found that fully implement all functional requirements IT services that fully implement individual functional requirements "Online store" "Authorization via SMS" IT services that implement individual parts of the functional requirements "Online store" –"Filling"; –"Policy registration"; –"Registration"; –"Sale"; –"Payment"; –"Print"; –"SMS"; –"Authorization via SMS" "Accounting for Contracts" –"Verification"; –"Contract verification"; –"Print"; –"Term"; –"Print contract" "Commissions" –"Commissions"; –"Calculation"; –"Print"; –"Verification" "Underwriting" –"Verification"; –"Policy verification"; –"Policy" Source:[20] The assessment resulted in the following recommendations[20]: –the best solution according to the selected criteria is the development of the "Electronic CMTPL Policy" service "from scratch";
70 Management of a modern IT company: theoretical and technological aspects –for reuse during the implementation of the IT project for the development of the "Electronic CMTPL Policy" service, the most suitable are the functions of the "Contract Accounting" and "Online Store" modules. To experimentally verify the results obtained in the study, it was proposed to check the correctness of these recommendations at different stages of the IT project life cycle for the development of the "Electronic CMTPL Policy" service. In particular, it was proposed to consider solutions for the formation and adjustment of IT project plans, based on estimates of losses that may arise as a result of identifying the fact of incompleteness of the description of a separate IT project function. 2.6.2 Description of the assessment of losses due to the replacement ofincomplete information with information about a similar element ofthe information system at the stage of IT project initiation First of all, the assessment of losses due to the replacement of incomplete information about the functions of the "Electronic CMTPL Policy" service with information about similar functions of the KSASK at the stage of IT project initiation was considered. For comparison, an assessment of the costs of developing the "Elec tronic CMTPL Policy" service "from scratch" was also carried out. To calculate the costs of developing the "Electronic CMTPL Policy" service "from scratch", it was proposed to use: –COCOMO model[15]; –a simplified method of functional points[23], which allows to estimate the efforts for the development of IS elements in conditions of almost complete absence of information about new IS functions; –a backfiring technique for converting the number of functional points into the number of lines of source code (as an argument of the COCOMO model). The values of the parameters of the simplified functional point method, the rollback coefficient k and the COCOMO model are given in Table2.2. As a result of calculating the number of functional points of the "Electronic CMTPL Policy" service, it was obtained FP CCC ()() .( ... 123 235235 235 9915 33 3702 61 functional pooints) . The results of calculating the estimates of efforts, time costs and personnel requirements for the IT project for developing the "Electronic CMTPL Policy" service "from scratch" using the COCOMO model are given in Table2.3.
71 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 Table 2.2 Values of the parameters of the simplified functional point method and the rollback coefficient k and the COCOMO model Element name Element value Explanation C19 Estimation of the project scale ("Custom application") C29 Estimation of the users of the design object ("Commercial project") C315 Estimation of the type of design object ("Software for publicly available services") kDB 16 Maximum normative number of SQL commands per one functional point kSW 46 Maximum normative number of Java commands per one functional point Bd1.12 Locked mode of operation (application system development) Source:[23] Table 2.3 Results of calculating the characteristics of the IT project for developing the "Electronic CMTPL Policy" service "from scratch" Characteristic name Characteristic value Units of measurement Service database development EDB 174.37 Person-months TDB 17.77 Months SSDB 9.81 Full-time and part-time employees Service software development ESW 524.49 Person-months TSW 27.08 Months SSSW 19.51 Full-time and part-time employees It should be remembered that the estimates of the characteristics of an IT project at the stage of its initiation may significantly deviate from the real characteristics of the same project. According to the research results given in[15], such estimates may exceed the value of the real characteristic by four times. Based on this, it was proposed to adjust the characteristics of the IT project given in Table2.3 by dividing the number of functional points by four (the result is 925.6525 functional points). The results of the adjustment are given in Table2.4. The data in Table2.4 show that even in the case of organizing parallel development of the database and software for the "Electronic CMTPL Policy" service, the duration of the IT project will be approximately 1year and 4months. This duration was recognized as unacceptable from the point of view of meeting the
72 Management of a modern IT company: theoretical and technological aspects business needs of insurance companies-users of KSASK (even considering that the decision to develop the "Electronic CMTPL Policy" service "from scratch" was recognized as the best according to the criteria for the quality of software product development). Table 2.4 Adjusted results of calculations of the characteristics of the IT project of the development the "Electronic CMTPL Policy" service "from scratch" Characteristic name Characteristic value Units of measurement Service database development EDB 40.67 Person-months TDB 10.22 Months SSDB 3.98 Full-time and part-time employees Service software development ESW 123.28 Person-months TSW 15.58 Months SSSW 7.914 Full-time and part-time employees Next, an assessment of possible losses due to replacing incomplete information about the functions of the "Electronic CMTPL Policy" service with information about similar functions of the "Online Store" module of KSASK at the stage of IT project initiation was carried out. This option was chosen because, according to the search results given in Table2.1, it showed the largest number of matches. It was decided to exclude from further consideration those search results given in Table2.1, which are parts of other search results for functions of the same module. The result of excluding duplicate search results and further comparison of the names of the functions of the service "Electronic CMTPL Policy" and the functions of the module "Online Store" is given in Table2.5. In addition, Table2.5 indicates the estimates of the number of functional points for each of the functions of the service being developed and the value of the number of functional points for the functions of the module. Since at the stage of IT project initiation, the executors only know the names of individual functions of the service being developed, it was proposed to consider the distribution of efforts for the development of individual functions of the service as uniform. Therefore, for further calculations, the value of the effort spent on the development of each individual function of the service was taken as 102.85(functionalpoints).
73 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 Table 2.5 Comparison of the names of the functions of the "Electronic CMTPL Policy" service and the "Online Store" module No. Service function name Number of service function points Name of the module function highlighted in the search results Number of functional points of the module function 1 "Filling out an electronic policy" 102.85 "Filling out an insurance contract" 97 2 "Issuing an electronic policy" 102.85 "Issuing an insurance policy" 84 3 "Selling an electronic policy" 102.85 "Selling an order" 157 4 "Printing an electronic policy agreement" 102.85 "Printing the source document" 48 5 "Paying for an electronic policy" 102.85 "Paying for the insurance contract" 142 6 "Authorization via SMS" 102.85 "Authorization via SMS" 134 For further calculations, it was proposed to use: –the COCOMO model[15]; –the improved functional points method[24], which allows to estimate the efforts for the development of IS elements in the conditions of reuse of individual system functions; –the backfiring technique for converting the number of functional points into the number of lines of source code (as an argument of the COCOMO model). To determine the number of developers qzj(T) required to eliminate the situation under consideration, the expression (16) was used, which in this case took the form q TKLOC T kFP T zi z B zadd z B zz () .( ) .( ) , = = 24 24 1000 , (2.47) where KLOCz – the number of thousands of lines of source code, the creation of which is necessary for additional processing with information that replaces; kz – the rollback coefficient, which determines the normative number of lines of source code for the implementation of one functional point of the z-th IS function; FPadd,z – the number of functional points as an estimate of the efforts for additional processing with information that replaces an incomplete description of the z-th IS function; Bz – an indicator, the value of which is determined by the selected mode of additional processing with information that replaces an incomplete description of the z-th IS function.
74 Management of a modern IT company: theoretical and technological aspects To calculate the value of FPadd,z in[24] it was proposed to use the expression FP FP Name Name Name addz z dz z , , (2.48) where FPz – the number of functional points, which is an estimate of the efforts to develop the z-th function of the developed functional problem; Named – the name of the d-th function of the developed problem, the description of which is incomplete; Namez – the name of the z-th function of the developed functional problem. Based on formula (2.48), to calculate the FPadd,z value it was necessary to perform stemming of the names of individual service functions and their corresponding module functions with the subsequent removal of stopwords. The results of these operations using the Porter stemmer are given in Table2.6. The FPz values for each of the module functions, the descriptions of which are proposed to be reused in the IT project for the development of the "Electronic CMTPL Policy" service, are given in Table2.5. These values, as established for caseI, are one of the results of the analysis of previous IT projects of the company ProfITsoft. Then, according to expression (2.48) for z=1, there is the following number of functional points FP add, "" "" " 197 Fill electron policiFill insur contract Filll insur contract functional " .( 97 2 364 67 ppoints). The results of calculating the FPAdd,z values for all functions are given in Table2.7. The values of the elements of formula (2.47), which were used to experimentally verify the results obtained during the assessment of losses in the situation under consideration, are given in Table2.8. As a result of the calculation according to formula (2.47) for z=1, it is possible to obtain q T 1 112112 2446 64 67 1000 2 24297482 2 8136 2 406() .(. ).(. ).. .. 88( ). man The results of the calculations of the qz(T) values given in Table2.9.
81 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 –consequence Ex2: "Concentration of IT company personnel and IT infrastructure elements on the design or development of other IS elements"; –consequence Ex3: "Transferring IT company personnel and IT infrastructure elements to the design or development of other elements of the same IS"; –consequence Ex4: "Refusal to use IT company personnel and IT infrastructure elements due to the lack of information about IS elements". The results obtained take into account the iterative nature of planning the activities of IT project teams and can be used for the majority of modern Agile and hybrid IT project management methodologies. Unlike modern models and methods of IT project financial management[2], the implementation of the developed detailed models allows to estimate losses and make decisions on IT project cost management in the absence of a priori defined descriptions of individual IS elements (functions), which is the expected result of the project. The use of these models allows to choose solutions during the implementation of the IT project that minimize additional losses that arise in the following situations: –a priori incompleteness of the results of the collection and analysis of functional requirements for the created system; –the occurrence of a significant number of changes in the descriptions of functional requirements and IS architecture and its elements due to changes in the business processes of the automation object; –the occurrence of changes in the descriptions of functional requirements and IS architecture and its elements due to the elimination of errors made during the design and implementation of individual elements of the created IS. However, the use of the developed detailed loss models is not free from some limitations. The main limitation of these is the orientation of the developed detailed loss models to the assessment of direct losses exclusively during the period of time considered at the current stage of the project life cycle (the total duration of the ITproject during the initiation and start of its planning; the total duration of the current iteration during the planning of the activities of the team of performers and the implementation of the IT project). The developed models practically do not take into account indirect losses that arise after making a decision on the selected consequence. Such a limitation during the practical application of the developed models may lead to the selection of a solution option that will be a local, rather than a global minimum of losses for the entire IT project or its individual iterations. Another limitation of the application of the obtained models is the implicit assumption of the constant availability of the relevant IT project performers and other resources. It is believed that these resources can be used at any time to eliminate the specified consequences of detecting incomplete information. In fact, the results of accounting for
82 Management of a modern IT company: theoretical and technological aspects the workload of personnel and other material resources of an IT project can significantly change the estimates of possible losses that arise during the elimination of the identified incompleteness of information about the elements of the created IS. The main drawback of the results of the study is the lack of technological sophistication of the developed detailed models. It should be remembered that the use of the obtained mathematical models in the IT company management system is impossible without conducting scientific and applied research in the field of developing relevant information technologies. It should also be recognized that the existing information technologies for managing an IT company and its IT projects practically do not allow using the obtained detailed models for their own improvement. There is a need to develop fundamentally new information technologies for managing an IT company, which would combine the use of the obtained detailed loss models with the capabilities of managing the work of an IT company in a time regime close to real. Therefore, the following main areas of further prospective research in the field of IT project cost management were proposed to be identified: –research on the improvement and development of the obtained detailed loss models and the general concept of cause-and-effect management of IT project costs, taking into account the characteristics of IT projects and programs; –research on the possibility of integrating detailed loss models into existing IS and information technologies for managing an IT company and its IT projects; –research on the development of fundamentally new IS and information technologies for managing an IT company, including using modern methods and tools of artificial intelligence. 2.8 Conclusions As a result of the study, the features of the design processes of IS elements in modern IT companies were identified. The situational-consequential scheme of the formation of losses in production as a result of the lack of information was adapted to these features. The adaptation results show that the use of the adapted scheme for estimating costs and making decisions on managing iterations of the implementation of the IS creation IT project in conditions of incompleteness or absence of descriptions of individual IS elements is possible. Detailed models of losses from the lack of information have been developed if additional processing of the information that replaces is necessary, as well as losses from the lack of information if there is a difference in the cost of the information that is replaced and the information that is replaced. The obtained models can be
83 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 used with various models and methods of assessing efforts to implement the IS creationITproject. An experimental verification of the obtained results has been carried out. The course and results of the calculations confirm the possibility of using the adapted scheme and the developed detailed models for estimating losses during the initiation, planning and implementation of the IS creation IT project. Conflict of interest statement The authors declare that there is no conflict of interest in relation to this paper, as well as the published research results, including the financial aspects of conducting the research, obtaining and using its results, as well as any non-financial personal relationships. Use of artificial intelligence statement The authors declare that they did not use artificial intelligence tools in preparing this manuscript. References 1. ISO/IEC/IEEE Standard No 15288:2015 (2015). Systems and software engineering – System life cycle processes. ISO/IEC/IEEE International Standard. https://doi.org/10.1109/IEEESTD.2015.7106435 2. Nastanova do zvodu znan z upravlinnia proiektamy. Nastanova PMBOK (2021). Project Management Institute, Inc. Available at: https://learn.ztu.edu.ua/ pluginfile.php/274061/mod_resource/content/1/PMBOK7_Ukr_ForPersonalUseOnly.pdf 3. Valadares, F. S., Moura, N. C. S., Pereira, T. N. F., Arantes, M. D. O. (2024). Identification of the Main Traditional Project Management Methods Through a Systematic Literature Review. International Journal of Advanced Computer Science and Applications, 15 (6). https://doi.org/10.14569/ijacsa.2024.0150697 4. Coali, A., Gambardella, A., Novelli, E. (2024). Scientific decision-making, project selection and longer-term outcomes. Research Policy, 53 (6), 105022. https:// doi.org/10.1016/j.respol.2024.105022
84 Management of a modern IT company: theoretical and technological aspects 5. Khosravizadeh, O., Maleki, A., Ahadinezhad, B., Shahsavari, S., Amerzadeh, M., Tazekand, N. M. (2022). Developing decision model for the outsourcing of medical service delivery in the public hospitals. BMC Health Services Research, 22(1). https://doi.org/10.1186/s12913-022-07509-1 6. Zhou, M. (2024). Optimization of University Scientific Research Project Management Resources Based on Genetic Algorithm. Disruptive Human Resource Management, 128–138. https://doi.org/10.3233/atde240422 7. Chen, K., Ou, W., Zeng, C., Wu, H., Ye, M. (2024). Research on Cost Management of Power Transmission and Transformation Project Based on BIM Technology. 2024 International Conference on Machine Intelligence and Digital Applications, 141–147. https://doi.org/10.1145/3662739.3670223 8. Shorikov, A. F., Butsenko, E. V.; Kahraman, C., Cevik Onar, S., Cebi, S., Oztaysi, B., Tolga, A. C., Ucal Sari, I. (Eds.) (2024). Description of the Structure and Functions of an Intelligent Software System for Optimizing Adaptive Project Control with Fuzzy Data. Intelligent and Fuzzy Systems, 488–495. https://doi. org/10.1007/978-3-031-67192-0_55 9. Rahman, H. U., da Silva, A. R., Alzayed, A., Raza, M. (2024). A Systematic Literature Review on Software Maintenance Offshoring Decisions. Information and Software Technology, 172, 107475. https://doi.org/10.1016/j.infsof. 2024.107475 10. Sakka, A., Kourjieh, M., Kraiem, I. B. (2023). An IT projects' conceptual model to facilitate upstream decision-making: project management method selection. International Transactions in Operational Research, 30 (6), 3687–3718. https:// doi.org/10.1111/itor.13231 11. Butt, S., Khan, S. U. R., Hussain, S., Wang, W.-L. (2023). A conceptual model supporting decision-making for test automation in Agile-based Software Development. Data & Knowledge Engineering, 144, 102111. https://doi.org/10.1016/ j.datak.2022.102111 12. Svyrydov, V.V., Moroz, B.I. (1992). Orhanyzatsyia protsessov obrabotkyynformatsyy po kryteryiam tsennosty y starenyia v ASU. Kharkiv: Vydavnytstvo"Osnova" pry Kharkivskomu derzhavnomu universyteti, 112. 13. Calderon-Tellez, J. A., Bell, G., Herrera, M. M., Sato, C. (2023). Project management and system dynamics modelling: Time to connect with innovation and sustainability. Systems Research and Behavioral Science, 41 (1), 3–29. https://doi. org/10.1002/sres.2926 14. Levykin, V. M., Ievlanov, M. V., Kernosov, M. A. (2014). Patterny proektirovaniia trebovanii k informatcionnym sistemam: modelirovanie i primenenie. Kharkiv: KhNURE, 320.
85 Cost management of an information system IT project in the conditions of incomplete information about its elements Chapter 2 15. COCOMO II Model Definition Manual. Available at: https://athena.ecs.csus. edu/~buckley/CSc231_files/Cocomo_II_Manual.pdf?ref=thestack.technology 16. ProfITsoft. Available at: https://profitsoft.dev/ua/index.php 17. Product history. ProfITsoft. Available at: https://ksask.profitsoft.dev/en/#history 18. Key features. ProfITsoft. Available at: https://ksask.profitsoft.dev/en/#characteristics 19. Iuriev, I. O. (2021). Developing IT Service Management Information Technology. Management Information System and Devises, 177, 57–63. https://doi. org/10.30837/0135-1710.2021.177.057 20. Iuriev, I. O. (2019). Methods, models and information technology of control of IT-services provision system[Extended abstract of PhD thesis; Kharkiv National University of Radio Electronics]. Available at: https://openarchive.nure.ua/ entities/publication/117efb1f-4d38-40f9-b3fa-dac4cb783736 21. Polozhennia pro fond obminu funktsionalom. ProfITsoft. Available at: https:// ksask.profitsoft.dev/polozhennya-pro-fond-obminu-funkczionalom/ 22. ISO/IEC/IEEE Standard No 9126-1 (2001). Software engineering – Product quality – Part 1: Quality model. ISO/IEC/IEEE International Standard. 23. Na start! Vnimanie! I? (2005). ITCua. Available at: http://itc.ua/articles/na_start_ vnimanie_i_21814/ 24. Levykin, V., Ievlanov, M., Neumyvakina, O., Levykin, I., Nakonechnyi, A. (2024). Estimation of IT-project efforts for information system creation in the conditions of re-use of its functions. Eastern-European Journal of Enterprise Technologies, 2 (2 (128)), 6–19. https://doi.org/10.15587/1729-4061.2024.301227 25. Technical support. ProfITsoft. Integrated Insurance Company Automation System. Available at: https://ksask.profitsoft.dev/en/#prices 26. Online store and customer access (IMCD). ProfITsoft. Available at: https://ksask. profitsoft.dev/client-access/
© The Author(s) of individual chapters, 2025 DOI: 10.21303/978-9908-9706-8-4.ch3 Abstract The aim of the study is to develop a method for solving the problem of assigning IT project tasks to its performers. The object of the study is the process of planning an IT project. During the study, the problem of assigning IT project tasks to its performers was solved. It is shown that in recent years heuristic solutions to this problem have gained popularity. The main requirements for such a solution are determined. The results of the analysis of modern research confirm the relevance of scientific and applied works devoted to solving the problem of assigning IT project tasks to its performers. According to the results of the study, it is proposed to represent the problem of assigning IT project tasks to its performers as a type of classification problem. This representation allowed using a polynomial Bayesian classifier to solve this problem. In the process of the study, the classification rule with a minimum error and calculations of the main elements of this rule were adapted to the specifics of the problem. An additional classification condition was established that prevents the maximum load of the IT project performer from being exceeded. Based on the results of the adaptation, the algorithm for solving the classification problem, which uses this adapted classifier, was modified. A general description of the method for solving the problem of assigning IT project tasks to its performers was developed. An algorithm for implementing this method was developed for a detailed description of the content of individual stages. The scheme of this algorithm and the proposed descriptions of the main steps of the algorithm determine the features of its implementation both as a methodology Tetiana Borysenko Maksym Yevlanov Konstantin Petrov Viktor Borysenko CHAPTER 3 Method of automatic assignment of tasks to IT project performers
87 Method of automatic assignment of tasks to IT project performersChapter 3 for applying the obtained solutions in the current management of an IT project and as a specialized information technology. To verify the operability of the obtained results, an experimental test of the method and its implementation algorithm was conducted during the management of one of the IT projects of an outsourcing IT company. The test results indicate the feasibility of using the developed method to solve the problem of assigning IT project tasks to its performers. The developed method contributes to better project planning, minimizes the administrative burden and helps to avoid delays and errors in project implementation. Keywords IT project, task, performer, polynomial Bayesian classifier, effort, team, sprint. 3.1 Introduction The main difference between IT project management and project management in any other field of human activity is the recognition of personnel as the single most used resource of the project. The emergence and widespread implementation of Agile project management methodologies in the IT industry have not led to significant changes in the existing attitude towards IT project personnel. As early as 2007, it was established that direct costs for an IT project are almost exclusively determined by labor costs[1]. Therefore, personnel management problems remain one of the main problems facing IT project management. Among these problems, it is worth highlighting the problems that arise during ITproject planning. In general, IT project planning includes: –definition of tasks; –assessment of task duration; –placement of tasks on a timeline that visually reflects their sequence, duration, start and end dates; –according to the results of the placement– assigning resources to each task from available jobs, divided by skill set[2]. Assigning tasks to IT project performers is a complex and time-consuming activity, especially in large and complex projects. During the implementation of this activity, important issues arise regarding the correct distribution of tasks between performers, ensuring the optimization of human resources for task performance, and maximizing the productivity of performers. It should be noted that modern IT companies use a significant number of various information technologies (IT) to support the IT project planning process. However,
88 Management of a modern IT company: theoretical and technological aspects the use of such IT is severely limited by the fact that the personnel of the majority of IT projects is a limited resource. Recognition of this fact led to the formulation of the Resource Constrained Project Scheduling Problem (RCPSP), which belongs to NP-hard problems[3]. Such problems for large and medium-sized IT projects are very often intractable due to the excessive growth of the volume of required calculations. Therefore, basic research suggests two main ways to solve this problem[4]: –using models and methods that provide accurate solutions to the project planning problem taking into account scarce constraints; –using heuristic solutions. As a result, the most common IT planning of IT projects mostly leaves the implementation of the activity of distributing project tasks between performers to the discretion of project management specialists. But this approach leads to subjective errors, irrational use of resources and overloading of individual performers. This often happens in large projects, where the complexity of coordination increases in proportion to the number of tasks and performers. Therefore, the automation of the activity of assigning tasks to performers during the planning and implementation of an IT project remains an extremely important and relevant scientific and applied task. Solving this task will undoubtedly contribute to the optimization of planning processes, avoid excessive administrative burden, ensure more accurate and efficient distribution of work, and also reduce the risk of errors or delays in an IT project. 3.2 Overview of the current state of solving the problem of automated planning of IT projects The existence of RCPSP as an NP-hard problem has been confirmed in recent years at the level of a number of recent versions of PMBOK[5,6]. Therefore, a significant amount of modern research is focused on the search and development of models, methods and technologies that allow either to obtain an exact solution or todevelop a heuristic solution to this problem. An example of research aimed at finding an exact solution to RCPSP is the work[2]. The goal of this research is to develop software that creates an optimal schedule that takes into account both technical and resource constraints of the project in order to prevent resource reallocation. In this case, [2]identifies three main factors that can affect the allocation of resources during the project life cycle: –an increase in the scope of the project, which will require greater efforts to complete it;
89 Method of automatic assignment of tasks to IT project performersChapter 3 –possible unavailability of resources during periods in which they were initially considered available; –an existence of competition for the same resources between several projects that can be performed simultaneously. As a result of the research in[2], the creation of software is claimed, which should help the project manager in creating the optimal schedule of this project. This software provides an accurate RCPSP solution in a reasonable period of time in the conditions of performing within the IT company of several projects simultaneously. At the same time, the developed system takes into account the distribution of ITproject performers according to their skill set for a better assessment of the duration of project tasks[2]. However, the research result given in[2] is not free from shortcomings that complicate the practical application of the developed software. The main of these shortcomings should be recognized as the need to recalculate all IT project schedules as a result of any change in the availability of IT company personnel. This need leads to an increase in the number of computational operations performed during planning or replanning of IT projects. In addition, this need contradicts the principle of the oncoming wave, according to which detailed planning of the IT project schedule until its completion is considered impractical[5]. Using heuristic solutions to solve the RCPSP requires dividing this problem into a sequence of separate IT project planning management tasks and solving each task separately using the appropriate heuristic. An example of such an approach is research[7]. In this work, it is proposed to develop task and developer profiles to solve the problem of assigning each project task to the most suitable developer. After that, the task is solved by finding the best match of these profiles for assigning tasks to developers. A comparative analysis showed that the Sokal and Snit method[7] should be considered the best for such a search. It should be noted that one of the authors of this study also tried to solve this problem in a similar way. In[8], he proposed a solution to create task and project developer profiles based on a data-logical description of the project work packages. The use of these profiles allowed[8] to propose a method for solving the problem of assigning enterprise employees to tasks of a new IT project, the basis of which is the apparatus of clustering and classification tasks. This method made it possible to assign to the task of the new project an employee whose value of the integral indicator of the quality of work performance was maximum. But this method of solving the problem of assigning each project task to the most suitable developer is not free from some shortcomings. First, the solutions proposed in[7] and[8] require the creation and constant maintenance of data warehouses in which it is necessary to store and update historical data on the profiles of completed
90 Management of a modern IT company: theoretical and technological aspects IT project tasks and project performers. Second, the use of complex data logical descriptions to build profiles leads to significant expenditure of money and time on additional work on the creation, support and maintenance of such profiles during the planning and performing of an IT project. Third, this method leaves open the question of the optimal distribution of tasks among project personnel from the point of view of maximizing and equalizing the employment of project performers. It must be recognized that the approach to profiling project tasks or work packages and storing the corresponding historical information is characteristic not only of IT projects. In[9], a similar approach is considered for creating a hierarchy of construction project task organization. This study is interesting in that it recog nizes the need to create project task profiles not based on special complex datalogical descriptions, but on the results of pre-processing text descriptions of project work packages. Although such a solution complicates the work of processing information about project work packages, it significantly simplifies the work of forming, maintaining and maintaining similar profiles of these work packages. An attempt to find an exact RCPSP solution using heuristics is considered in[10]. This study proposes to use a genetic algorithm to form a schedule that provides maximum profit from the implementation of an IT project. The following assumptions are taken into account[10]: –capacity constraints can often be systematically changed by temporarily assigning expensive additional production resources or using overtime; –project revenue decreases with increasing duration of its implementation. But such a solution to RCPSP is not free from the shortcomings identified for the software for finding the exact solution to RCPSP proposed in[2]. In addition to these shortcomings, the accuracy of the obtained result is affected by the shortcomings caused by the use of the genetic algorithm, of which it is worth noting in particular[11]: –the tendency to converge to a local optimum instead of searching for a global optimum; –the difficulties that arise when formulating the stopping condition of the genetic algorithm; –the poor scalability of genetic algorithms to the complexity of the problem being solved (which is of particular importance for creating schedules for large complex IT projects). To eliminate the last of the listed shortcomings, in[12] it was proposed to use agenetic algorithm to solve the problem of assigning IT project tasks to members of the development team within one iteration or sprint of the project. At the same time, when solving this problem, the experience of each specific performer is taken into account. As descriptions of IT project tasks, user stories that the development
97 Method of automatic assignment of tasks to IT project performersChapter 3 To solve the problem of assigning IT project tasks to its performers, the following criteria are proposed to be used: –the priority of performing that each task has; –the possible belonging of the task to an epic that is assigned to one or several teams of performers; –the total efforts for one sprint of each performer (depending on the position), which should not be exceeded when distributing tasks; –the possible gradation of performers by positions (competences) in the team. Of all the NB options, the multinomial naive Bayes classifier (MNB) was proposed to be used by the IT project task assignment task assigners. The choice of this classifier was due to the following considerations: –attributes of IT project task descriptions as classification objects contain discrete features corresponding to a polynomial distribution; –MNB is easy to implement and allows for fast processing of large amounts of data. The MNB effectiveness has been confirmed by studies[30,31]. In particular, [30]describes the results of a comparison of the Bernoulli-Naive Bayes algorithm and MNB and concludes that the polynomial model generally outperforms the Bernoulli model. [31]considers the application of seven different algorithms for the classification and analysis of tweets related to real natural disasters. MNB, in comparison with the Bernoulli-Naive Bayes algorithm and logistic regression, showed the best results in classifying tweets as concise texts with media elements. Let the object of classification X be described by a set of features (X1,X2,…,Xn). Then the posterior probability of belonging of the object X to the class yk from the set of classes Y should be calculated by the formula P yXXX Py PX y PX XX kn ki k i n n |,,..., | ,,..., , 12 1 12 (3.1) where P(yk|X1,X2,…,Xn)– the posterior probability of belonging of a classification object with a set of features (X1,X2,…,Xn) to the class yk; P(yk)– the prior probability of the existence of the class yk; P(Xi|yk)– the probability that a classification object that has a feature Xi with some fixed set of values will be classified as belonging to the class yk; P(X1,X2,…,Xn)– the probability that a classification object X will be described by a set of features (X1,X2,…,Xn). For the case when the prior probabilities P(yk) for each element yk of the set of classes Y are known, the classification rule with minimum error[22] is used, which selects the class with the highest posterior probability
98 Management of a modern IT company: theoretical and technological aspects P yXXX Py PX y knkk ik i n |,,..., argmax | 12 1 , (3.2) where ∝ – the proportionality sign. In many cases, the classification rule (3.2) is translated into logarithmic space. Then it takes the form[22]: lo g|,,..., argmax lo g| Py XX XP yPXy knkk ik i n 12 1 , (3.3) lo g|,,..., argmax lo gl og |Py XX XPyXPX y knkki i n ik 12 1 , (3.4) where Xi in (3.4) is the frequency of occurrence of the feature values in the set of studied data (training set), on the basis of which the prior probabilities P(yk) are calculated. To estimate the parameters of each class, in [22] it was proposed to use asmoothed version of the maximum likelihood estimation P Xy N Nn ik yi y k k | , (3.5) where Nyi k – the number of times the feature Xi occurs in the objects of classification X belonging to the class yk; Nyk – the total number of times all features Xi occur in the objects of classification X belonging to the class yk; n– the number of features describing the objects of classification in the training set; α – the smoothing parameter. Although the value of the parameter α may be different for each feature, in[22] it is recommended to follow the generally accepted practice, setting α =1 for all features. The estimate of the probability value P(yk) is proposed to be calculated as follows P y O O k yk , (3.6) where Oyk – the number of objects of classification present in the training set and belonging to the class yk; O– the total number of objects of classification present in the training set. The UML activity diagram describing the algorithm for solving the classification problem using MNB is shown in Fig.3.2.
99 Method of automatic assignment of tasks to IT project performersChapter 3 [The set of objects for classification is empty] [Yes] [No] 1. Formation of a training set of objects 3. Selecting a set of classes 2. Testing the hypothesis that individual features of an object from the training set are independent of each other 4. Calculating the prior probability P(yk) of a randomly selected object belonging to a class yk 5. Calculating the probability P(Xi|yk) of each feature of the object for each given class yk 6. Calculating the posterior probability P(yk|X1,X2,…,Xn) of the object belonging to each class yk taking into account all its features Searching for a class with the max posterior probability of the object belonging to it and assigning the object to this class Fig. 3.2 UML activity diagram describing the algorithm for solving the classification problem using a polynomial Bayesian classifier The use of MNB for solving applied classification problems is associated with the emergence of a number of problems[20]. Among these problems, it is necessary to highlight: –the problem of reducing the MNB score for a class with a smaller number of training samples in the case when the training set has more training samples for one class than for another; –the problem of the possible appearance of a class that lacks training sets and, as a result, cannot participate in the classification process; –the problem of checking the independence of the features of the classification objects in the training set and those that need to be classified from each other. These problems can be solved in the process of preparing the training set by appropriately selecting the training objects for classification and checking the training set for the independence of the features of these objects. Another problem arises when the classification object, the class of which should be determined in the process of solving the problem, has a feature that has never
100 Management of a modern IT company: theoretical and technological aspects appeared in the training set[20]. This problem can be solved using the smoothing parameter α , 0< α ≤1. The choice of the value of this parameter can significantly affect the results of solving the classification problem. For solving the problem of assigning IT project tasks to its performers, the description of the algorithm for solving the classification problem using MNB given in Fig.3.2 is not entirely suitable due to the following features: –in this algorithm, the classification process continues until there are no objects left that have not yet been classified; –the number of objects that can belong to a class is not limited in any way. In the problem of assigning IT project tasks to its performers, the assignment process must continue until each performer accumulates the necessary total efforts for its position. Therefore, there is a need to modify the method for solving the classification problem using MNB, taking into account this feature. 3.4 Results of developing a method for solving the problem of assigning ITproject tasks to its performers 3.4.1 Results of adapting the polynomial naive Bayes classifier tothespecifics of the problem of assigning IT project tasks toitsperformers The process of adapting MNB to the specifics of the task of assigning IT project tasks to its performers is proposed to be divided into two main stages: –adaptation of the classification rule with minimal error (3.2) to the specifics of the task of assigning IT project tasks to its performers; –modification of the algorithm for solving the classification problem using the adapted MNB. During the adaptation of the classification rule with minimal error (3.2) to the specifics of the task of assigning IT project tasks to its performers, the main elements of this rule were considered and their definitions and calculation rules were clarified. As objects of the MNB classification X as a result of the adaptation, ITproject tasks presented in the form of User Story and Task should be considered. The features(X1,X2,…,Xn) of the classification object X, which are taken into account in rule(3.2), should be the words from which the names of IT project tasks are composed. The names of tasks should reflect the characteristic features of the problem area or the performed part of the IT project. Classes yk, by which the classification objects X are distributed, should describe individual performers who are part of one
101 Method of automatic assignment of tasks to IT project performersChapter 3 or more teams of IT project performers. The training set should contain data on already completed tasks (in particular, their names) and performers of these tasks. Then the classification rule with minimal error (3.2) should be formulated as arule for selecting the performer ek with maximum posterior probability e Pe Pe kk ki k i n argmax | 1 , (3.7) where P(ek)– the prior probability that the IT project task belongs to the performerek; P( ω i|ek)– the probability that the word ω i from the IT project task title will be classified as being distributed to the performer ek; n– the number of words in the IT project task title. The prior P(ek) should be calculated by the formula P y T T k ek , (3.8) where Tek – the number of tasks in the training set of tasks performed by the performer ek; T– the total number of tasks in the training set. The probability P( ω i|ek) should be calculated using formula (3.3) P e N NV ik ew e ki k | , (3.9) where New ki – the number of occurrences of the word ω i in the names of the tasks of the training set performed by the performer ek; Nek – the total number of words in the names of the tasks of the training set performed by the performer ek; V– the size of the vocabulary (the total number of words in the training set); α – the smoothing parameter (allows to solve the problem of the appearance in the names of tasks of such words that have never been encountered in the training set). In our case, it is possible to assume that α =1. But the considered results of adapting the classification rule (3.2) to the specifics of the task of assigning IT project tasks to its performers do not take into account the specifics of the existence of individual performers within the IT project performed by the company. In particular, the adapted classification rule (3.7) does not take into account: –the condition of no excess of the total efforts of each performer ek for one sprint; –the condition of the performer ek being in the team in a specific position during the planning and performing of the sprint, which can be determined by the competencies of this performer.
102 Management of a modern IT company: theoretical and technological aspects To take these conditions into account, an assumption is introduced according to which each individual performer ek can be characterized by the parameter "Maximum effort of the performer in one sprint". This parameter can be formally represented as the indicator "Maximum allowable weight of class ek" Vmaxk. The value of the indicator Vmaxk is measured in story points (Story Points)– conventional units of measurement for expressing the assessment of the total effort required for the full implementation of a product backlog element or any other part of the work[32]. For each performer ek, the value of the indicator Vmaxk is determined based on the position held by the performer ek in the team according to their own competencies. Similarly, to describe each individual task of an IT project, the parameter "Task completion effort" has been introduced, which can be formally represented by the indicator "Task weight" vi. The value of the indicator vi is also measured in Story Points. Then the classification rule (3.7) when assigning each j-th task of an IT project will work only if the condition is met S vV vk i i j k 1 max, (3.10) where Svk– the total weight of the class ek accumulated as a result of the successful assignment of the previous j–1 tasks and the current j-th task of the IT project. If this condition is violated, then the posterior probability of the IT project task belonging to the performer ek should be considered equal to zero. The search for the performer ek with the maximum posterior probability of the task belonging to it in this case should be continued. The introduction of condition (3.10) led to the emergence of a new variant of solving the classification problem. In the derived variant, this problem was solved until all available classification objects were classified. In our case, an additional variant of solving the classification problem should be considered the variant in which the value of the total weight Svk of each of the classes approached the value of the indicator Vmaxk as much as possible. This condition means that each performer received the maximum possible number of IT project tasks to perform, which it can perform during the planned sprint. Based on the considered results of adapting MNB to the specifics of the task of assigning IT project tasks to its performers, the algorithm for solving the classification problem using MNB was modified and shown in Fig.3.2. The UML activity diagram describing the modified algorithm for solving the classification problem using the adapted MNB is shown in Fig.3.3.
103 Method of automatic assignment of tasks to IT project performersChapter 3 [The set of objects for classification is empty] [The total weight of the class together with the complexity of the new task exceeds max] [Yes] [Yes] [No] [No] 1. Formation of a training set of objects 3. Selecting a set of classes 2. Testing the hypothesis that individual features of an object from the training set are independent of each other 6. Calculating the prior probability P(yk) of a randomly selected object belonging to a class yk 7. Calculating the probability P(Xi|yk) of each feature of the object for each given class yk 8. Calculating the posterior probability P(yk|X1,X2,…,Xn) of the object belonging to each class, taking into account all its features 4. Determination of the maximum permissible total weight of the filling of each class (Vmax, k) 5. Determination of the weight of each classification object (vi) 10. Zeroing the posterior probability of belonging to the class ym 9. Searching for a class (ym) with the max posterior probability of the object belonging to it (max P(yk|X1,X2,…,Xn)) 11. Assigning the object to the class (ym) and increasing the total accumulated weight Svk = Svk + Vi Fig. 3.3 UML activity diagram describing a modified algorithm for solving a classification problem using an adapted polynomial Bayesian classifier
104 Management of a modern IT company: theoretical and technological aspects 3.4.2 Description of the results of developing an algorithm forimplementing the developed method The use of the modified algorithm allowed to develop a method for solving the problem of assigning IT project tasks to its performers as a sequence of such stages: Stage1 is the preparatory stage. Stage2 is the formation of the "Product Backlog" artifact as a set of IT projecttasks. Stage3 is the formation of detailed descriptions of IT project tasks, the descriptions of which are included in the "Product Backlog" artifact. Stage4 is the assignment of IT project tasks from the "Product Backlog" artifact to performers. Formation of the "Sprint Backlog" artifact. Completion of the method. But this representation of the developed method is too generalized. Therefore, for the successful implementation of this method, an appropriate algorithm was developed. The UML activity diagram, which describes the scheme of this algorithm, is shown in Fig.3.4. The use of the developed algorithm allowed to describe in detail the features of the implementation of each of the stages of the method as a sequence of steps of this algorithm. Let's consider these sequences of steps in more detail. The implementation of Stage1 of the developed method consists of a sequence of the following algorithm steps: Step1. Formation of the training set. The implementation of Stage2 of the developed method consists of a sequence of the following algorithm steps. Step2. Dividing the initiatives of the current IT project into epics. Step3. Distribution of epics between teams. Step4. Dividing epics into User Stories and Tasks. Step5. Adding Tasks to the task pool outside the epics. The implementation of Stage3 of the developed method consists of a sequence of the following algorithm steps. Step6. Determining the values of the attributes of User Stories and Tasks. Step7. Determining the maximum efforts of each performer. Step8. Sorting tasks by priority. The implementation of Stage4 of the developed method consists of a sequence of the following algorithm steps. Step9. If the "Product Backlog" artifact does not contain unassigned tasks, go to Step17. Otherwise, select the next task from the "Product Backlog" artifact. If this task has the attribute of belonging to an epic, go to Step10. Otherwise, go to Step11.
105 Method of automatic assignment of tasks to IT project performersChapter 3 1. Formation of the training set 14. Zeroing the posterior probability for the selected performer 15. Replenishing the list of tasks of the selected performer 16. Removing the selected performer from the list of performers 4. Dividing epics into User Stories and Tasks 2. Dividing the initiatives into epics 3. Distribution of epics between teams [Need to distribute tasks for sprint] [Task pool is empty] [Task belongs to epic] [List of performers is empty] [Accumulated effort of selected performer's tasks along with new task difficulty = max] [Accumulated effort of selected performer's tasks together with new task difficulty <= max] [Yes] [Yes] [Yes] [Yes] [Yes] [Yes] [No] [No] [No] [No] [No] [No] 8. Sorting tasks by priority 9. Select the next task from the Task pool 11. Dividing the task name into separate words 12. Calculating the posterior probability of the task belonging to each performer 13. Determining the performer with the highest posterior probability of the task belonging to it 6. Determining the values of the attributes of User Stories and Tasks 5. Adding Tasks to the task pool outside the epics 7. Determining the maximum efforts of each performer 17. Combining the tasks of the performers of each team into a package of tasks of the team of performers 10. Form a list of performers belonging to teams that are assigned to the epic to which the current task belongs Fig. 3.4 UML activity diagram describing the algorithm for solving the problem of assigning IT project tasks to its performers
106 Management of a modern IT company: theoretical and technological aspects Step10. Form a list of performers belonging to teams that are assigned to the epic to which the current task belongs. Step11. Dividing the task name into separate words. Step12. Calculating the posterior probability of the task belonging to each performer. Step13. Determining the performer with the highest posterior probability of the task belonging to it. Check the accumulated total weight of the performer Svk. If this total weight does not exceed the value Vmaxk, go to Step15. Step14. Zeroing the posterior probability for the selected performer. Go to Step13. Step15. Replenishing the list of tasks of the selected performer. Check the accumulated total weight of the performer Svk. If this total weight is still less than Vmaxk, go to Step9. Step16. Removing the selected performer from the list of performers. If the list of performers of the planned sprint is not yet empty, go to Step9. Otherwise, end the method. Step17. Combining the tasks of the performers of each team into a package of tasks of the team of performers. Form the artifact "Sprint Backlog". End the method. In Step1 of the developed algorithm, data collection is first performed. It involves the accumulation of data on past tasks of those IT projects that are being performed, as well as on previously completed IT projects of the company. This data should include data on the performers to whom the tasks were assigned, and on the results of the work of these performers. This data must be prepared for further analysis and training of the adapted MNB before the first application of the developed method. When performing data collection, it is important to check the condition under which the number of tasks performed by different performers should be approximately the same. In the event that the use of the developed method is recognized as successful, this data can be updated after the completion of each individual IT project performed by the IT company. After collecting historical data, it is necessary to test the hypothesis of the independence of individual words in the names of tasks. For this, it is proposed to use pairwise correlation analysis of each combination of two words of the name. Such an analysis can be done, for example, by counting how many times each pair of words occurs in the collected task names and comparing this number with the total number of word pairs in the collected task names. If the correlation percentage is lower than a priori set threshold, the correlation between these words can be considered absent. For preprocessing of the collected data, it is necessary to remove stop words[18]. Stop words are words that do not add meaning to the phrase on their own, such as prepositions, articles, conjunctions, pronouns, etc.
113 Method of automatic assignment of tasks to IT project performersChapter 3 Table 3.2 List of completed tasks and their performers No. Task Task performer (gradation level) 1 Frontend Design Update Ivan Petrov (Junior) 2 UX/UI Improvement Ivan Petrov (Junior) 3 Frontend Framework Update Ivan Petrov (Junior) 4 Backend API Integration Oleksiy Kovalek (Middle) 5 Web Page Load Optimization Oleksiy Kovalek (Middle) 6 Business Intelligence Report Creation Oleksiy Kovalek (Middle) 7 Backend Data Processing Optimization Oleksiy Kovalek (Middle) 8 Real-time Data Analytics Dashboard Oleksiy Kovalek (Middle) 9 Database Performance Tuning Maria Ivanova (Senior) 10 User Authentication Implementation Maria Ivanova (Senior) 11 Data Encryption Setup Maria Ivanova (Senior) 12 Data Migration from Legacy Systems Maria Ivanova (Senior) 13 DevOps Pipeline Automation Maria Ivanova (Senior) 14 Security Vulnerability Assessment Gana Sergienko (Middle) 15 Cloud Resource Allocation Gana Sergienko (Middle) 16 Continuous Deployment Setup Gana Sergienko (Middle) 17 Network Configuration Management Gana Sergienko (Middle) 18 Mobile App Bug Fixing Dmytro Orlov (Junior) 19 Automated Testing Script Development Dmytro Orlov (Junior) 20 System Backup Configuration Dmytro Orlov (Junior) Table 3.3 Number of occurrences of tokens in the names of the tasks of each of the performers No. Word (token) Ivan Petrov (Junior) Oleksiy Kovalek (Middle) Maria Ivanova (Senior) Gana Sergienko (Middle) Dmytro Orlov (Junior) 1 2 3 4 5 6 7 1 Allocation 0 0 0 1 0 2 Analytics 0 1 0 0 0 3 API 0 1 0 0 0 4 App 0 0 0 0 1 5 Assessment 0 0 0 1 0 6 Authentication 0 0 1 0 0 7 Automated 0 0 0 0 1 8 Automation 0 0 1 0 0 9 Backend 0 2 0 0 0
114 Management of a modern IT company: theoretical and technological aspects 1 2 3 4 5 6 7 10 Backup 0 0 0 0 1 11 Bug 0 0 0 0 1 12 Business 0 1 0 0 0 13 Cloud 0 0 0 1 0 14 Configuration 00011 15 Continuous 0 0 0 1 0 16 Creation 0 1 0 0 0 17 Dashboard 0 1 0 0 0 18 Data 0 2 2 0 0 19 Database 0 0 1 0 0 20 Deployment 0 0 0 1 0 21 Design 1 0 0 0 0 22 Development 0 0 0 0 1 23 DevOps 0 0 1 0 0 24 Encryption 0 0 1 0 0 25 Fixing 0 0 0 0 1 26 Framework 1 0 0 0 0 27 Frontend 2 0 0 0 0 28 Implementation 0 0 1 0 0 29 Improvement 1 0 0 0 0 30 Integration 0 1 0 0 0 31 Intelligence 0 1 0 0 0 32 Legacy 0 0 1 0 0 33 Load 0 1 0 0 0 34 Management 0 0 0 1 0 35 Migration 0 0 1 0 0 36 Mobile 0 0 0 0 1 37 Network 0 0 0 1 0 38 Optimization 0 2 0 0 0 39 Page 0 1 0 0 0 40 Performance 0 0 1 0 0 41 Pipeline 0 0 1 0 0 42 Processing 0 1 0 0 0 43 Real-time 0 1 0 0 0 44 Report 0 1 0 0 0 Continuation of Table 3.3
115 Method of automatic assignment of tasks to IT project performersChapter 3 1 2 3 4 5 6 7 45 Resource 0 0 0 1 0 46 Script 0 0 0 0 1 47 Security 0 0 0 1 0 48 Setup 0 0 1 1 0 49 System 0 0 0 0 1 50 Systems 0 0 1 0 0 51 Testing 0 0 0 0 1 52 Tuning 0 0 1 0 0 53 Update 2 0 0 0 0 54 User 0 0 1 0 0 55 UX/UI 1 0 0 0 0 56 Vulnerability 0 0 0 1 0 57 Web 0 1 0 0 0 Table 3.4 List of IT project tasks, ordered by priority values, planned for current performing No. Task name Priority Effort (SP) Epic ID 1 Backend API performance tuning 1 8 E2 2 Mobile app security audit 1 8 E3 3 Web page speed optimization 1 8 E5 4 Cloud infrastructure setup 1 13 E6 5Network configuration automation 1 8 E9 6 Business intelligence dashboard creation 2 13 E8 7 Backend service integration 2 8 E2 8 Continuous integration pipeline setup 2 5 E7 9 System backup strategy implementation 2 5 E10 10 Automated testing suite development 2 8 – 11 Frontend Feature Extension 2 8 E1 12 Frontend design enhancement 2 5 E1 13 Database schema optimization 2 5 E4 14 Real-time data analysis implementation 2 5 – 15 Mobile app testing 2 5 E3 16 Data encryption security setup 3 3 E11 17 User interface design update 3 3 E1 18 Cloud service migration 3 13 E6 19 Machine learning model training 3 13 E12 Continuation of Table 3.3
116 Management of a modern IT company: theoretical and technological aspects From Table3.4 it can be seen that the total number of IT project tasks planned for the current performing is 19, of which two tasks (No.10 and No.14) are tasks that do not belong to any of the epics and are assigned additionally. The progress of Stage4 of the developed method (the actual solution of the problem of assigning IT project tasks to their performers) was proposed to be considered using the example of its iteration for the task "Frontend design enhancement" and Ivan Petrov (Junior) as its possible performer. Before the start of this iteration, the names of the IT project tasks given in Table3.4 were also tokenized. The parameters for calculating the value of the classification rule (3.7), which establishes the posterior probability of the performance of the task "Frontend design enhancement" by the performer Ivan Petrov, are given in Table3.5. Table 3.5 Parameters for calculating the value of the classification rule (3.7) Parameter Value General parameters Tek 3 T20 N e k 8 α 1 V57 "Frontend" token New ki 2 "Design" token New ki 1 "Enhancement" token New ki 0 (missing from the training set) Then the value of the prior probability Pek according to formula (3.8) is P IvanPetrov(" ") .. = = 3 20 015 The value of the probability P("Frontend"|"Ivan Petrov") according to formula(3.9) is P Frontend IvanPetrov(" "|"" ).. 21 8157 3 65 0046154
117 Method of automatic assignment of tasks to IT project performersChapter 3 The value of the probability P("Design"|"Ivan Petrov") according to formula (3.9) is P Design IvanPetrov(" "|"" ).. 11 8157 2 65 003077 The value of the probability P("Enhancement"|"Ivan Petrov") according to formula(3.9) is P IvanPetrovEnhancement(" "|"" ).. 01 8151 3 65 0015385 The value of the posterior probability according to formula (3.7) is P ePe kik i n |.... 1 0150046154 003077 0015385 0.0000033277370962995 33 10 6 .. The results of the calculations of the posterior probability for each task and performer are given in Table3.6. Table 3.6 Results of the calculations of the posterior probability for each task and performer Task No. Ivan Petrov (Junior) Oleksiy Kovalek (Middle) Maria Ivanova (Senior) Gana Sergienko (Middle) Dmytro Orlov (Junior) 1 2 3 4 5 6 18.4×10–9 3.48×10–8 3.52×10–8 7.44×10–9 6.62×10–9 28.4×10–9 5.81×10–9 8.8×10–9 1.49×10–8 2.65×10–8 38.4×10–9 6.97×10–8 8.8×10–9 7.44×10–9 6.62×10–9 45.46×10–7 4.7×10–7 1.29×10–6 2.14×10–6 4.57×10–7 55.46×10–7 4.7×10–7 1.29×10–6 2.14×10–6 9.13×10–7 68.4×10–9 9.29×10–8 8.8×10–9 7.44×10–9 6.62×10–9 75.46×10–7 2.82×10–6 6.43×10–7 5.36×10–7 4.57×10–7 88.4×10–9 1.16×10–8 3.52×10–8 2.98×10–8 6.62×10–9 98.4×10–9 5.81×10–9 1.76×10–8 7.44×10–9 2.65×10–8 10 8.4×10–9 5.81×10–9 8.8×10–9 7.44×10–9 5.29×10–8 11 1.64×10–6 4.7×10–7 6.43×10–7 5.36×10–7 4.57×10–7 12 3.3×10–6 4.7×10–7 6.43×10–7 5.36×10–7 4.57×10–7 13 5.46×10–7 1.41×10–6 1.29×10–6 5.36×10–7 4.57×10–7
118 Management of a modern IT company: theoretical and technological aspects 1 2 3 4 5 6 14 8.4×10–9 3.48×10–8 5.28×10–8 7.44×10–9 6.62×10–9 15 5.46×10–7 4.7×10–7 6.43×10–7 5.36×10–7 3.65×10–6 16 8.4×10–9 1.74×10–8 1.06×10–7 2.98×10–8 6.62×10–9 17 5.04×10–8 5.81×10–9 1.76×10–8 7.44×10–9 6.62×10–9 18 5.46×10–7 4.7×10–7 1.29×10–6 1.07×10–6 4.57×10–7 19 8.4×10–9 5.81×10–9 8.8×10–9 7.44×10–9 6.62×10–9 The results of applying logarithmic transformation (based on the decimal logarithm) for better perception of the classification results according to formula (3.3) are given in Table3.7. Table 3.7 Results of applying logarithmic transformation of posterior probability calculations for each task and performer Task No. Ivan Petrov (Junior) Oleksiy Kovalek (Middle) Maria Ivanova (Senior) Gana Sergienko (Middle) Dmytro Orlov (Junior) 1–8.076 –7.46 –7.45 –8.13 –8.18 2–8.076 –8.24 –8.06 –7.83 –7.58 3–8.076 –7.16 –8.06 –8.13 –8.18 4–6.26 –6.33 –5.89 –5.67 –6.34 5–6.26 –6.33 –5.89 –5.67 –6.04 6–8.076 –7.03 –8.06 –8.13 –8.18 7–6.26 –5.55 –6.19 –6.27 –6.34 8–8.076 –7.94 –7.45 –7.53 –8.18 9–8.076 –8.24 –7.75 –8.13 –7.58 10 –8.076 –8.24 –8.06 –8.13 –7.28 11 –5.79 –6.33 –6.19 –6.27 –6.34 12 –5.48 –6.33 –6.19 –6.27 –6.34 13 –6.26 –5.85 –5.89 –6.27 –6.34 14 –8.076 –7.46 –7.28 –8.13 –8.18 15 –6.26 –6.33 –6.19 –6.27 –5.44 16 –8.076 –7.76 –6.98 –7.53 –8.18 17 –7.3 –8.24 –7.75 –8.13 –8.18 18 –6.26 –6.33 –5.89 –5.97 –6.34 19 –8.076 –8.24 –8.06 –8.13 –8.18 Continuation of Table 3.6
119 Method of automatic assignment of tasks to IT project performersChapter 3 The results of the IT project task distribution for the performers from teams A and B are presented in Table3.8. They show that none of the performers reached their maximum workload. This is probably due to the lack of tasks with a complexity that would exactly match the remaining maximum workload capabilities of the performers. Table 3.8 Task distribution results User Stories&Task The complexity (SP) Performer Max effort (SP) Remain-der SP 11.Frontend Feature Extension 8 Ivan Petrov (Junior) 20 4 12.Frontend design enhancement 5 17.User interface design update 3 3.Web page speed optimization 8 Oleksiy Kovalek (Middle) 35 1 6.Business intelligence dashboard creation 13 7.Backend service integration 8 13.Database schema optimization 5 1.Backend API performance tuning 8 Maria Ivanova (Senior) 50 8 8.Continuous integration pipeline setup 5 10.Automated testing suite development 5 14.Real-time data analysis implementation 5 16.Data encryption security setup 3 18.Cloud service migration 13 4.Cloud infrastructure setup 13 Gana Sergienko (Middle) 35 1 5.Network configuration automation 8 19.Machine learning model training 13 2.Mobile app security audit 8 Dmytro Orlov (Junior) 20 2 9.System backup strategy implementation 5 15.Mobile app testing 5 In the case of assignment of task No.10, the total weight of assigned tasks for the performer Dmytro Orlov exceeded his maximum allowable weight. Therefore, it was decided to transfer this task to the next performer with the highest posterior probability. This performer turned out to be Maria Ivanova. In the case of assignment of task No.19, the total weight of assigned tasks for the performer Maria Ivanova exceeded her maximum allowable weight. Therefore, it was decided to transfer this task to the next performer with the highest posterior
120 Management of a modern IT company: theoretical and technological aspects probability. This performer turned out to be Ivan Petrov. But in the case of assignment of task No.19 to him, the total weight of assigned tasks also exceeded his maximum allowable weight. Therefore, it was decided to transfer task No.19 to the next performer. This performer turned out to be Gana Sergienko. After distributing tasks among the performers of the IT project, task packages were formed for each team of performers by combining tasks that were distributed to performers of the same team. The formed task packages of each team are given in Table3.9. Table 3.9 Results of using the developed method Team Task A 3. Web page speed optimization 6. Business intelligence dashboard creation 7. Backend service integration 11. Frontend Feature Extension 12. Frontend design enhancement 13. Database schema optimization 17. User interface design update B 1. Backend API performance tuning 2. Mobile app security audit 4. Cloud infrastructure setup 5. Network configuration automation 8. Continuous integration pipeline setup 9. System backup strategy implementation 10. Automated testing suite development 14. Real-time data analysis implementation 15. Mobile app testing 16. Data encryption security setup 18. Cloud service migration 19. Machine learning model training 3.6 Discussion of the research results As a result of the study, a method was developed for automatically solving the problem of assigning IT project tasks to its performers. The developed method is based on the presentation of the problem of assigning IT project tasks to its
121 Method of automatic assignment of tasks to IT project performersChapter 3 performers as a type of classification problem. This presentation made it possible to propose using MNB to solve this problem. This version of the classifier is quite simple to implement and allows for good processing of sets of attribute data, which are descriptions of IT project tasks. During the study, the existing version of MNB was adapted to the features of the problem of assigning IT project tasks to its performers. In particular, the classification rule with minimal error (3.2) as a result of adaptation to the features of this problem took the form (3.7). The results of adapting the methods for calculating the values of the elements of the classification rule (3.7) are given in the form of expressions (3.8) and (3.9). To take into account the limitations that arise as a result of the IT project performer being in the team in a specific position, it was proposed to add condition (3.10) to the classification rule (3.7) of the possibility of assigning a task to a specific performer. The results of adapting MNB to the specifics of the task of assigning IT project tasks to its performers were used as the basis for modifying the algorithm for solving the classification problem. The result of this modification in the form of an activity diagram in the UML language is shown in Fig.3.3. Based on the results obtained, a method was developed for automatically solving the problem of assigning IT project tasks to its performers. Using this method makes it possible to exclude human participation in the process of directly solving the problem of assigning IT project tasks to its performers (in the process of performing Stage4 of the method). This makes it possible to obtain a solution to this problem in a sufficiently short period of time and reduce the time spent on planning individual sprints (iterations) of the IT project. Another significant advantage of the developed method is a fairly high level of objectivity of the results of task distribution. The use of a modified algorithm for solving the classification problem based on the adapted MNB allows making decisions based on statistical data and historical information, which minimizes the possibility of subjective errors and bias. This ensures transparency of the task distribution process, which in turn increases trust on the part of the performers and helps to increase their motivation. Unlike the solution to this problem described in[15], the proposed method allows at least partially to take into account the experience and competence of the performers of the IT project. The experience of the performers when solving the problem is taken into account by selecting for a specific performer those tasks whose names largely coincide with the names of tasks successfully completed by this performer earlier. The competence of the performers when solving the problem is taken into account by assigning to each specific performer their own value of the
122 Management of a modern IT company: theoretical and technological aspects indicator "Maximum permissible weight of class ek" Vmaxk. This value is set based on the position held by the ek performer in the team based on their own competencies. The use of MNB to solve the problem of assigning IT project tasks to its performers made it possible, in contrast to the solution described in[16], to abandon vector descriptions of IT project tasks and the calculation of the cosine similarity measure. The calculation of the scalar elements of the classification rule (3.7) and condition (3.10) requires less computational resources and requires less time to calculate the result of solving the problem. The main limitation of the application of the developed method is the constant need to maintain the historical data repository, which is used to train the adapted MNB, in an up-to-date state. This data should be collected by analyzing previous ITprojects, assessing the success of individual tasks, and providing feedback from the teams of performers. Regular updating and validation of historical data are critically important for ensuring the accuracy and relevance of the results of solving the problem of assigning IT project tasks to its performers. Among the limitations and disadvantages of the application of the algorithm for implementing the developed method, it is also worth noting some difficulties with the classification of tasks that have similar names, but differ in content. This can lead to erroneous distribution of tasks between individual performers. To improve the accuracy of classification, it is necessary to use a larger amount of historical data for training the model. Regular updating and expansion of the keyword dictionary will help the algorithm better understand the context of the tasks. Therefore, work on the creation and development of a specialized thesaurus, on the basis of which the training set for MNB should be formed, is one of the promising areas of further research on this issue. Another area of further research into solving the problem is the use of modern artificial intelligence tools to increase the accuracy and objectivity of solving the problem of assigning IT project tasks to its performers. In particular, it is about the possibility and feasibility of using lemmatization for pre-processing text names of individual IT project tasks. Although one of the authors of this study has conducted work in this direction[34], it requires further development for successful use in the field of ongoing management of IT projects and their performers. 3.7 Conclusions As a result of the research, a method for solving the problem of assigning ITproject tasks to its performers was developed. The use of this method requires human
129 Method for assessing the complexity of an IT project for video game localizationChapter 4 ITprojects was tested in the project on the Ukrainian localization of the game "Sid Meier's Civilization® VI". The predicted translation duration was 11hours 11minutes. The actual translation duration was 9hours 57minutes (the difference with the predicted value was 12.1%). Additionally, the assessment results obtained using the developed method were compared with the assessment results obtained using the common methodology for assessing time spent on software localization. The results showed that using the developed method allows for the most accurate estimates of time spent on third-level IT projects for video game localization. Keywords Localization, assessment, IT project, translation unit, video game. 4.1 Introduction Developers of modern IT products are increasingly focused on consumers from the global, rather than local market. However, the adaptation of IT software products to local markets(localization) remains a necessity caused by the prevalence of languages, cultural diversity and the specifics of local legislation. One of the most complex types of such IT products, from the point of view of localization, are computer games(or video games). The term "computer game" should be understood as a separate class of application software that involves the interaction of the user( player) with other players or a computer in order to achieve a goal set by the rules of thegame[1]. Video game localization is a relatively new area of activity in the IT industry. Researchers have recognized that since the 2000s. IT companies have begun to apply a new approach known as "deep localization"[2,3]. This approach requires bringing the game closer to the consumer in each individual localization(everything that does not contradict the game world itself and can facilitate player immersion can be revised and adapted in accordance with what is considered a more successful local influence)[3]. Therefore, modern research suggests dividing the global problem of "deep localization" into two separate aspects[2, 4, 5]: –"linguistic" localization (which mainly concerns the translation process); –"product localization" (which consists in adapting a product or service to the culture and language of customers in a specific target market[4,5], even deeply changing the relevant structural and artistic aspects of the product). IT projects for video game localization also differ significantly from other ITprojects for the creation, modification or development of IT products. Among the
130 Management of a modern IT company: theoretical and technological aspects differences of IT projects for video game localization, it is worth noting the special requirements that are imposed on the planning of such projects. In particular, during planning, an important task is to assess the complexity of video game localization. Such an assessment requires a deep understanding of the various factors that affect the successful implementation of relevant IT projects. An objective and accurate assessment of the complexity of video game localization allows for increased accuracy in planning time and personnel needs for an IT project, and also affects the accuracy of assessing financial costs for this IT project. The current state of research in the field of software localization for IT products and video games, as a separate class of such products, demonstrates some progress in the development of methods and tools for assessing the complexity of IT projects. Separate works by such leading institutions and organizations as the Localization Industry Standards Association (1990–2011), the International Game Developers Association (IGDA), and some localization service providers are devoted to solving the issues of effective management and assessment of localization projects. However, the problem of increasing the accuracy of assessing the complexity of IT projects for video game localization has not been fully resolved, in particular, due to the growth of globalization processes and the complexity of video games. At the same time, given the high requirements for translation quality and technical adaptation, correct assessment of complexity becomes critically important for the successful management of modern IT projects. Accurate forecasting of localization operations is critically important, since errors in assessment can lead to underassessment or overassessment of resources(team, budget, etc.) or time required for project implementation. These errors, in turn, negatively affect project planning processes and end-user satisfaction. Therefore, research on the development of new and improvement of existing methods for assessing the complexity of IT projects for video game localization is relevant from both theoretical and applied points of view. 4.2 Analysis of current research in the field of software and video gamelocalization In general, the number of studies in the field of software and video game localization should be recognized as insufficient. Thus, in[2] it was recognized that for the period 1981–2021 in well-known journals, conferences and seminars indexed in the three main scientific search engines (Scopus, WoS and Google Scholar), only 45studies were published on this topic. A similar conclusion was made in[6]: out of 468 researched scientific articles devoted to the processes of game development,
131 Method for assessing the complexity of an IT project for video game localizationChapter 4 localization was considered in only a few of these articles. At the same time, attention was mainly paid to the following issues[6]: –research on regional aspects of game localization; –research on individual concepts related to game localization. An important problem was and remains the problem of effective organization of localization work. An equally important issue is the possibility of connections between localization and other work of an IT project on video game development. Unfortunately, in the early 2010s, the relationship between localization and software development processes was largely ignored in research. In practice, localization was often separated from development[4]. In 2013,[7] proposed an extended video game value chain diagram that included a game localizer (Fig.4.1). Hardware Manufacturer Game Developer Game Localizer Publisher Distributor Retail Consumer Fig. 4.1 Diagram of the extended video game value chain Source:[7] As can be seen from this diagram, the game localizer function is excluded from the main chain of the IT company. In the process of localizing a game, the localizer could interact with the publisher, the game developer, and the hardware manufacturer[7]. The reasons for this separation are the perception of localization as a laborintensive and expensive activity, often performed by external partners of IT software development companies. Such external partners have limited access to developers and rely on documentation and formalized processes. This contrasts with the minimal documentation and on-site presence cultivated in Agile IT project management methods[8]. This situation has hardly improved in the first half of the 2020s. In[9] it is noted that as of 2023, there are almost no studies that analyze localization from the perspective of video game development. According to the results of a survey of developers of video games available in Spanish on Steam, in[9] it was found that the choice of translator is influenced by the size and experience of the development com pany. The earlier the localization is planned, the greater the likelihood that the game will be translated by a professional who cannot play the game before its translation. In general, from the results of the processing of the surveys presented in[9], it is
132 Management of a modern IT company: theoretical and technological aspects possible to conclude that it is desirable to consider the localization of a video game as a subproject or a separate IT project, which is carried out in parallel with the ITproject for the development of the video game. However, the models and methods of managing an IT project for the localization of a video game are formed exclusively on the basis of the private experience of individual specialists and ITcompanies and have hardly been studied by scientists. This conclusion is confirmed by the results of field research by a specialist in the field of video game localization M.Toftedahl. Based on the results of this research, published in[10], he proposed to present the organizational structure of an IT company for the development of games in the indie sector in the form of adiagram shown in Fig.4.2. Internal Environment Art Lead Art Team 2D art 3D art Animation Programming Lead Programming Team Game programming Tools programming Localization Game design Game writing Design Lead Design Team Audio Lead Audio Team Internal Management Community Management Project management Sound design Music Management of External Environment Competencies External Environment Localization Game programming Game writing Music Legal services Sound design PR Marketing Server programming Customers Server infrastructure Fig. 4.2 Structure of an indie game development company based on field research in Sweden and China Source:[10] As can be seen from the diagram in Fig.4.2, game localization is a work or separate IT project that is performed both by the IT development company and by the external (relative to the IT development company) environment. The interaction of the IT development company with such external projects in[10] is proposed to be organized through the management of the competencies of the performers of these external projects. However, in the studied cases of video game localization, the results showed that localization is a more demanding IT project than initially planned.
133 Method for assessing the complexity of an IT project for video game localizationChapter 4 Even if the support of competencies from the external environment is used, management and development should also largely take place in the internal environment of the IT development company[10]. In general, based on the results of the research conducted in[10], the following general conclusions were formulated: –game localization is one of the important development tasks that must be planned in a timely manner; –game localization is a task which complexity is underestimated; –localization is not the main (critical) element of the IT project for developing an indie video game, it is implemented as needed; –in the context of indie game development, game localization is synonymous with translation, more complex aspects of localization related to game mechanics and game design (the so-called culturalization of the gameplay) require additionalresearch. The factors identified as contributing to success in game localization in[10] are: –cultural sensitivity; –effective communication and collaboration; – use of localization tools and technologies. The author of[10] believes that these factors are likely to be relevant to game localization efforts in different contexts and game genres. Among the problems faced by game localization professionals,[10] specificallymentions: –tight localization deadlines; –the need for effective localization project management. These problems, according to the author of[10], are likely common to the entire industry. The main tools used in the process of video game localization will remain machine and automated (computer-aided translation, CAT) translation systems. However, the use of such systems has its own characteristics, which depend on both the translator and the language into which the text is translated. The experience of using machine translation systems in the process of software localization under the conditions of daily work of professional translators for languages with low resources and rich morphology (Czech, Estonian, Hungarian, Latvian, Lithuanian and Polish) is reviewed in[11]. The results of experimental studies showed that the use of machine translation suggestions in addition to the use of CAT increased the productivity of translators in all assessment experiments. However, significant differences were observed in the productivity of different translation tasks and in the results of different translators. In[11] it was proposed to explain these differences
134 Management of a modern IT company: theoretical and technological aspects by the work mode and skills of individual translators. It should also be noted that the error rate increased in all of the experiments (except one), while remaining at the "Good" quality level. This result was explained in[11] by the fact that translators tend to trust the suggestions coming from the machine translation tool and do not check them sufficiently, even if they are marked as machine translation suggestions. It should be recognized that with the development of the capabilities of artificial intelligence tools, interest in the use of such tools in the field of software and video game localization has increased. Thus, in[12] an example of tuning the original neural machine translation model RNN-encoder-decoder with the addition of categorical information related to the problem of rare words (words that relate to a certain subject area, which is a common phenomenon in the text of software) is considered. However, this model has not been tested for video game localization and, according to the authors, provides successful translation results mainly for English and other official languages used by the United Nations[12]. The use of GPT capabilities for video game localization was investigated in[13]. The results show that the GPT-4 model is capable of providing multilingual output in a structured form. The GPT-4 model seems to have the ability to support the resolution of transcreation-related problems such as brainstorming, comparison, and coherence, although the results are not always satisfactory, and there are errors including conceptual repetition, formatting issues, and inappropriate or unusual portmanteau. In terms of output quality, the English-Chinese combination appears to be less creative compared to adaptations from English into French or Spanish. However, to draw any conclusions on this issue, further comparative qualitative and quantitative studies are needed, as well as reception studies among users to find out their preferred translations, as creativity is a highly subjective matter[13]. However, the use of GPT-4 for video game localization is not free from many drawbacks, among which[13] particularly highlight the following: –the strong dependence of localization performance on the specific arrangement of tokens in the GPT-4 model; –the limitation of the amount of context that can be used to improve coherence can lead to transcreations that deviate from previous work or do not correspond to the setting of the video game; –GPT models can produce hallucinatory or nonsensical results or contain grammatical or spelling errors; –GPT models may not take into account culturally sensitive issues, which can lead to the perpetuation, reproduction, and even reinforcement of stereotypes and prejudices that were present in the training data of these models;
135 Method for assessing the complexity of an IT project for video game localizationChapter 4 –the different amount of data available in different languages affects the quality of results for certain language pairs during translation and cultural adaptation; –GPT models risk violating copyright and intellectual property laws, which can expose users to financial and legal consequences; –GPT models are currently limited by paid access, and the number of monthly messages allowed by more powerful models is limited, which negatively affects the localization market and workforce; –opacity, together with the randomness of the results, limits the interpretability and applicability of the GPT model, as well as the reproducibility of the results. However, the authors of the study[13] note that the results obtained and the identified limitations of the application of GPT models for video game localization were established by them on the example of one video game. Therefore, additional research is required to clarify the conclusions regarding the possibility and feasibility of using GPT models for video game localization. In general,[13] recommends using the capabilities of GPT models to interact with such categories of localization specialists as interns and beginners. Thanks to the capabilities of GPT models, these categories of specialists can increase their productivity and creativity, while maintaining control over the result using spoken machine transcreation based on prompts[13]. Summing up the interim results of the analysis of research on game localization tools, it should be recognized that the main localization tools remain the human translator and the CAT and machine translation tools he/she uses. The main advantages of using such tools are their speed and cheapness compared to human trans lation. However, the use of machine translation has a number of disadvantages, among which it is especially worth noting: –the impossibility of a complete and high-quality understanding of the context; –the inability to interpret idioms and cultural features; –the inability to convey the literary features of the text; –difficulties with the translation of highly specialized texts; –the inability to be creative. Therefore, it is impossible to fully localize a video game using machine translation tools, especially while maintaining the consistency of terminology, ensuring its correctness, and taking into account textual and linguistic features. Editing such a text requires a large number of editors, and its result will contain more undetected errors. To combine the advantages of each method, machine translation tools can be used in some cases at the translator's discretion to speed up the translation of individual translation units, rather than the text as a whole. The term "translation unit" (TU) should be understood as the smallest indivisible text that can be represented by a word, phrase, sentence, or set of sentences[14].
136 Management of a modern IT company: theoretical and technological aspects Thus, based on the results of the analysis, it is possible to conclude that there is a need for a modern perception of video game localization as a separate type of IT project. The main resources of these IT projects are game text translation specialists and CAT and machine translation tools. Modern Agile and hybrid methodologies can be used for the general management of such IT projects. However, specific methods and models for assessing, planning, and managing the work of such projects remain almost unexplored. Therefore, the aim of this study is to develop a method for assessing the complexity of video game localization work with increased accuracy. This increase is planned to be achieved by taking into account a larger number of factors affecting the complexity of localization work than in currently existing methods. To achieve the aim, it is proposed to solve the following objectives: –develop a method for assessing the complexity of an IT project for video gamelocalization; –carry out experimental verification of the developed method. 4.3 Basic research method The object of the study is the process of project assessment and control. The purpose of this process is to ensure the balance and feasibility of plans, determine the status of the project, its technical implementation and implementation of processes, ensure implementation according to plans and schedules within the projected budgets for technical tasks[15]. The main hypothesis of the study is the hypothesis of the possibility of increasing the accuracy of forecasting the duration of the IT project for video game localization by means of: –research of factors that affect the complexity of localization work; –development of a method for assessing the complexity of the IT project for video game localization. The main assumption of the study is the assumption about the method of implementing the IT project for video game localization. According to the results of the analysis of modern scientific research presented in Section4.1, this method is considered as the translation of game texts by a human translator using machine translation tools. A study of current experience in quantitatively assessing time costs as an indicator of the complexity of an IT project for video game localization[16–19] showed that the main method of such assessment remains the method based on
137 Method for assessing the complexity of an IT project for video game localizationChapter 4 the number of words. This method is based on calculating the indicator of translation time costs test according to the formula test=w/s, (4.1) where w– the number of words in the text being translated; s– the recommended translation speed. Different variations of this methodology can use different analogues of the indicators of the number of words in the text being translated and the recommended translation speed. Thus, instead of the indicator "Number of words in the text being translated", the following indicators of text volume can be used: –"Total number of words in the localization file" (indicator W1); –"Number of unique words in the localization file" (indicator W2); –"Number of words in unique TUs of the localization file" (indicator W3). The software translation speeds recommended in[16–19] can be characterized by the following indicators: –"Interface translation speed" (indicator S1)[16, 17]; –"Speed taking into account the complexity of the text" (indicator S2)[18]; –"Baseline speed" (indicator S3)[19]. A comparison of the above speed indicators is presented in Table4.1. The considered indicators W1–W3 and S1–S3 can be compared depending on whether they take into account certain factors of localization complexity. The results of the comparison of the complexity factors considered in[16–19], which are taken into account by variations of the method based on the number of words, are given in Table4.2. The variants of the translation duration assessment methodology defined in[16–19] offer slightly different indicators for assessing the translation duration. However, all of these variants state that the duration of editing is equal to half the duration of the text translation work. Table 4.1 Comparison of translation speed indicators proposed in[16–19] Indicators Recommended speed, words/day Recommended speed, words/hour Availability of variability S1 2000 250 Present (depends on the type of content) S21200–4000 200–500 Present (depends on difficulty) S3 2500 300 Absent Source:[16–19]
138 Management of a modern IT company: theoretical and technological aspects Table 4.2 Results of the comparison of the complexity factors, taken into account byvariations of the method based on the number of words Factor Indicator W1W2W3 Text volume Takes into account Takes into account Takes into account Text complexity Doesn't count Doesn't count Doesn't count Speed of the method application Very high High High Need for preliminary data processing Absent Necessary Necessary Taking into account the linguistic features of the text Doesn't count Doesn't count Doesn't count Uniqueness of the text Doesn't count Takes into account Partially takes into account Factor Indicator S1S2S3 Text volume Takes into account Takes into account Takes into account Text complexity Partially takes into account Takes into account Doesn't count Speed of the method application High Low Very high Need for preliminary data processing Absent Necessary Absent Taking into account the linguistic features of the text Doesn't count Partially takes into account Doesn't count Uniqueness of the text Doesn't count Partially takes into account Doesn't count The main drawback of the considered indicators of text volume and translation speed is that they do not take into account the complexity of the text (with the exception of S2). The factors considered in Table4.2 are sufficient for comparing the indicators with each other. However, the list given in Table4.2 is not an exhaustive list of factors that determine the duration and complexity of translation work. According to the data in Table4.1, it can be concluded that in[16–19] the recommended translation speed is set at or close to 300 words per hour. In cases where the translation work is already in progress, the average translation speed can be taken as the recommended speed. However, this indicator will be individual for each individual translator.