Process science: the interdisciplinary study of socio-technical change

Brocke, Jan vom et al. Article — Published Version Process science: the interdisciplinary study of sociotechnical change Process Science Provided in Cooperation with: Springer Nature Suggested Citation: Brocke, Jan vom et al. (2024) : Process science: the interdisciplinary study of socio-technical change, Process Science, ISSN 2948-2178, Springer International Publishing, Cham, Vol. 1, Iss. 1, https://doi.org/10.1007/s44311-024-00001-5 This Version is available at: https://hdl.handle.net/10419/316986 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by/4.0/ Open Access © The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creativecommons.org/licenses/by/4.0/. EDITORIAL Brockeetal. Process Science (2024) 1:1 https://doi.org/10.1007/s44311-024-00001-5 Process Science Process science: theinterdisciplinary study ofsocio-technical change Jan vom Brocke1,2,3,4* , Wil M. P. van der Aalst5 , Nicholas Berente6 , Boudewijn van Dongen7 , Thomas Grisold8 , Waldemar Kremser9 , Jan Mendling10,11 , Brian T. Pentland12,13 , Maximilian Roeglinger14 , Michael Rosemann15 and Barbara Weber8 Abstract Process science is the interdisciplinary study of socio-technical processes. Sociotechnical processes involve coherent series of changes over time, entailing actions and events that include humans and digital technologies. The ubiquitous availability of digital trace data, combined with advanced data analytics capabilities, offer new and unprecedented opportunities to study such processes through multiple data sources. Process science is concerned with describing, explaining, and intervening in socio-technical change. It is based on four key principles; it (1) puts socio-technical processes at the center of attention, (2) investigates socio-technical processes scientifically, (3) embraces perspectives of multiple disciplines, and (4) aims to create impact by actively shaping the unfolding of socio-technical processes. Introduction We live in an age of process. Many core phenomena of our time speak to complex dynamics involving change: globalization, the platformization of economies, the rise of (generative) AI, as well as societal movements, or political decisions, have in common that we can learn a lot more about them if we think of them as ongoing processes, rather than stable entities or systems. To study the emergence and development of contemporary phenomena, we need to embrace that contemporary processes are socio-technical. They comprises actions and events that involve humans (individuals or collectives) and digital technologies (e.g., mobile devices, organizational systems, AI). Socio-technical phenomena unfold, evolve and wane, they occur at macro, meso and micro levels and continually incorporate an ever-evolving frontier of technological innovation. Further, a process view sees the world primarily as flowing as opposed to being in a stable state (Rescher 1995). Taking this perspective is not a trivial; it goes against our deeply ingrained assumption that the world embodies stability and permanence (Chia 1999). An orientation towards socio-technical processes embraces a view of the world that is becoming and in a constant state of change (Baygi etal. 2021; Tsoukas & Chia 2002). *Correspondence: [email protected] 1 European Research Center for Information Systems, Münster, Germany 2 University of Münster, Münster, Germany 3 University of Liechtenstein, Vaduz, Liechtenstein 4 National University of Ireland, Maynooth University, Maynooth, Ireland 5 RWTH Aachen, Aachen, Germany 6 University of Notre Dame, Notre Dame, USA 7 Eindhoven University of Technology, Eindhoven, The Netherlands 8 University of St, Gallen, St. Gallen, Switzerland 9 Johannes Kepler Universität Linz, Linz, Austria 10 Humboldt University of Berlin, Berlin, Germany 11 Vienna University of Economics and Business, Vienna, Austria 12 Weizenbaum Institut, Berlin, Germany 13 Michigan State University, East Lansing, USA 14 University of Bayreuth, Bayreuth, Germany 15 QUT Brisbane, Brisbane, Australia Page 2 of 14 Brockeetal. Process Science (2024) 1:1 Process science as a field is particularly relevant because socio-technical processes produce digital traces. Digital traces represent recorded activities that are left behind when humans use digital technologies (Freelon 2014). Typically, digital traces are equipped with temporal information, allowing for granular insights into the dynamics through which socio-technical phenomena take shape (Lazer etal. 2020). Furthermore, an evolving frontier of computational techniques, such as process mining, enable increasingly sophisticated analyses of such data. Analysis of digital traces using these contemporary techniques can support data driven theorizing and generate new theories about all sorts of phenomena (Berente etal. 2019). Process science aims to integrate data from diverse sources, including organizations, environments, the human body, and others. Process science pursues two broad types questions. First, it encourages researchers to develop a scientific understanding that is concerned with “what is changing?” and “how is it changing?” At the same time, advancing our understanding of phenomena in terms of their underlying processes provides us with new opportunities for generating and influencing change. Hence, since change both occurs naturally and can be constructed artificially, we need to further ask: “(how) can and should we influence change?” If we know why, how, and when certain changes occur, we can design and study interventions. This is important since many suggest that scientists should take on the roles of realworld problem solvers (Gaieck etal. 2020). In this paper, we conceptualize process science as the interdisciplinary study of sociotechnical processes in terms of their evolution, transition, and change at various levels of abstraction. The goal of process science is to reconcile methods, theories, and approaches of different scientific fields to establish a comprehensive understanding of socio-technical processes as well as means to design interventions to them. Process science is a platform for disciplines to jointly advance the study of processual dynamics and find ways to change them. Conceptualizing process science An interdisciplinary focus onprocesses The term ‘process’ has been appropriated by various disciplines in different ways (Mendling etal. 2021; Pettigrew 1997; Van de Ven & Poole 1995). These include sociology, engineering, computer science, the natural sciences, or organizational science (e.g., Cornwell 2015; Dumaset al.2018; Leenders etal. 2016). Regardless of the discipline, focusing on processes in a temporal sense provides an encompassing view on the dynamics of phenomena as they evolve over time (Langley & Tsoukas 2017; Rescher 2000). While processes might be constantly changing, people conceptualize processes in discrete ways. This happens at different levels of granularity. The interest around processes in computer science, information systems, and management, often lies in the identification and analysis of sequences of activities and events, not only in the sense of how they give rise to and constitute phenomena (Blagoevet al.2023), but also in terms of how the underlying dynamics change and take different shapes over time (Cloutier & Langley 2020). Yet, there is a continuum to what extent the conceptualization of processes would account for dynamics, either very closely (as in process mining research) or more remotely (as e.g. in business model research). As different research communities have applied a process perspective to different phenomena, Page 3 of 14 Brockeetal. Process Science (2024) 1:1 they developed different methods to study them. Recent arguments have stressed that cross-fertilization among these fields may lead to new methods in order to study how and why certain phenomena evolve and change over time (Lazer etal. 2020; Mendling etal. 2021; Simsek etal. 2019). Therefore, process science strives to be an interdisciplinary field for studying socio-technical phenomena, providing a platform to foster continuous exchange across various specialized fields. Embracing socio‑technical processes The acute relevance of process science is tied to the changes and shifts associated with digital technologies (Mendling etal. 2020). Regardless of whether we think of private, public, or work-related contexts, the majority of our everyday activities are performed with digital technologies. All forms of communication, collaboration and coordination are supported by or enabled through digital technologies, be it sensor technology, personal digital assistants, or smart environments. From a broader point of view, many global societal phenomena—such as social-media “shit storms,” crowdsourcing, or cryptocurrency—are made possible only through digital technologies. Taking these developments together, it is important that process science be sociotechnical. It involves studying activities of humans – individual or collective – whose activities are tied to the use of digital technologies (Sarker etal. 2019). Since both human activities as well as digital technologies continuously change and evolve, process science embraces the study of this change in socio-technical processes. If processes appear stable, they are only stable “for now” (Feldman etal. 2016). Leveraging digital traces andcomputational techniques The socio-technical nature of these processes leads to new opportunities to study them and their underlying dynamics. This is because the use of digital technologies produces digital traces: digital records of activities that are typically equipped with temporal information (Freelon 2014; Pentland etal. 2020). Key is that such traces are produced even if actors do not decide so intentionally or willingly. In times of increasing connectivity between humans and digital technologies, “there is no opt-out; there is no way to be invisible.” (Leonardi & Treem 2020, p. 1603). In other words, digital traces are produced in the performances of any kind of socio-technical process, and they are produced as naturally occurring (as opposed to research provoked) data (Lazer etal. 2020). Digital traces offer insights into activities of actors that would not have been possible to study before (Akemu & Abdelnour 2020), since manually obtaining traces is not feasible at large scales. The growing availability of increasingly sophisticated computational tools affords new and more powerful approaches for analyzing trace data. Techniques involving process mining, sequence analysis, network analysis, and text and video analysis are continually progressing in terms of accuracy and precision in temporal data analysis. Machine learning applications of all sorts are capable of identifying ever more interesting patterns that would remain invisible to human sense-making (Lindberg 2020). Digital traces offer new opportunities to study how phenomena evolve in terms of underlying sequences of events (Pentland etal. 2017). This opens up a powerful view to understand and predict how phenomena change and behave over time (Lazer etal. 2020; Page 4 of 14 Brockeetal. Process Science (2024) 1:1 Pentland etal. 2017), as well as to change them through these insights (Grisold etal. 2024; Oliver etal. 2020). With the use of digital trace data, we can study socio-technical phenomena at different levels, including the micro-, meso-, and macro-level (e.g., individual and organizational level). This can complement established theories (Lazer etal. 2020), or it can enable the development of new, more robust theories altogether (Tremblay etal. 2021). This opportunity has led to a recent movement to strengthen theories of socio-technical phenomena using digital trace data through “computationally intensive theory construction” (Miranda etal. 2022), which offers a step-by-step process for making theoretical contributions from trace data (e.g., Berente etal 2019; Lindberg 2020). Embracing such opportunities and establishing a dialogue across disciplines to study socio-technical processes from an integrated viewpoint is at the core of process science. Defining alanguage tostudy change overtime When studying processes, a critical step involves defining the language used to describe a process. This language enables the researcher to structure the temporal ordering of activities or events that influenced entities causing them to change state. Such languages can range from natural language to formal computer-executable specifications. Regardless of the specific focus, however, language defines the level of abstraction at which a process is studied. A key issue when determining the language for socio-technical change involves deciding on an approach for temporally punctuating process data, and the researcher must decide how to reconcile discrete and continuous change. Especially in the context of digital technologies, the level of abstraction is often discrete. Rather than describing the continuous change process, the language describes the discrete moments in time when the change is significant enough to look at the process in two different states: one before and one after this time point. Often, even in a continuous process, such timepoints exist such as when processes move from phase to phase. For example, a criminal process might move from investigation to prosecution to sentencing to rehabilitation. However, when studying continuous processes, sequences of activities or events can be difficult to specify. In such cases researchers must use judgement based on the goals of the research and the established approaches for dealing with similar situations. For example, one might formally specify a continuous process using sets of differential equations expressing the state at any point in time as a derivative of the continuous change process, e.g. when describing the volume of water in a bucket while emptying it. A further choice that has to be made when describing processes in a language is the decision whether one aims to describe a specific instance of a process, or all possible instances of a process. A process can be specified by providing all possible activity sequences, for example, or perhaps by common patterns, or a particular instance. These decisions will be guided by goals for the analysis and the domain of study. For example, in disaster management, no two processes are ever executed in the same way, which makes strict workflow languages (describing the exact sequence of steps each process instance should follow) a poor choice for describing them. Against this background, process science researchers study all aspects of process languages. From formal questions on how to quantify the relation between (the observed part of) a process and the language used to describe it, to more foundational questions Page 5 of 14 Brockeetal. Process Science (2024) 1:1 on the suitability of a specific language for describing processes in general or even in a specific context, or higher level questions on how to best describe the evolution of the continuous changing process under consideration. A definition forprocess science Using the term ‘process science’, we draw on, pursue, and extend an existing intellectual trajectory. From a computer science perspective, van der Aalst and Damiani (2015) have used the term to denote “the broader discipline that combines knowledge from information technology and knowledge from management sciences to improve and run operational processes.” (p. 2). By this account, process science extends data science which is “an inter-disciplinary field that uses scientific methods, processes, algorithms and systems to extract knowledge and insights from many structural and unstructured data.” Furthermore, Mendling (2016) used the term in the context of business process management to call for more scientific and empirical research in the field. The term process science has also been used as a specific field of engineering that is concerned with fluids and circulation (Judd & Stephenson 2002; Velis etal. 2009). While these works approach process science from within the frame of a specific discipline, they share (for example) an interventional perspective. In turn, we intend to emphasize process science in terms of an interdisciplinary study of processes. Process thinking is put center stage and its use should not be limited to a specific research discipline. Thus, we define process science as follows: Process science is the interdisciplinary study of socio-technical processes over time. By socio-technical process, we mean a coherent series of changes that involve actions including humans and technologies, and occur at different levels. Key tenets ofprocess science Process science emphasizes the following key characteristics; (1) socio-technical process are in the focus; (2) socio-technical processes are investigated scientifically; (3) investigation occurs through an interdisciplinary lens, and (4) process science aims to change processes and create practical impact. We will explain these tenets in the following. At the core of process science is the study of socio-technical processes (focus). It aims to describe, explain and intervene in these processes (objective). Thereby, it embraces an interdisciplinary viewpoint, integrating contributions from various disciplines (perspective); some of these disciplines are exemplified here. Socio‑technical processes are thefocus Process science inverts a basic assumption: by focusing on the processes themselves, this view backgrounds the entities involved with the process. Across a wide range of disciplines, we have been trained to think of entities first. For example, computer science and information systems adopt the stance that processes change the properties of entities that exist a priori (Wand & Weber 1993). Influential process modeling languages such as UML and BPMN share this commitment to representing entities first (Chinosi & Trombetta 2012; Fowler 2004). Other disciplines, such as biology, are beginning to question the “entity-first” perspective and consider a “process first” perspective. Page 6 of 14 Brockeetal. Process Science (2024) 1:1 Nicholson and Dupré (2018, p. 3) propose that “the living world is a hierarchy of processes, stabilized and actively maintained at different timescales.” They argue that the entities we recognize (e.g., cells or organisms) are the result of those processes. In organization studies, the “process first” perspective has also been proposed (Langley & Tsoukas 2017; Tsoukas & Chia 2002). In practice, processes and entities always co-exist: the fire burns the wood and the wood fuels the fire. However, the shift in perspective from entity-first to process-first affords a novel way to think about familiar problems. For our purposes, the process-first perspective provides a useful way to see analogies across domains that have different entities but similar processes. Fundamental to the interplay between entities and processes is the concept of time, the ultimate process. Processes evolve as entities change over time. While the changes are continuous in nature, the chosen abstraction level to describe the process may make the entire process discrete for the purpose of the study. Within process science, we take different perspectives to study such socio-technical processes, which can be informed by e.g. social sciences, such as organizational sciences, or technical research, such as computer science. Table1 exemplifies that process science is concerned with a variety of socio-technical processes (Rescher 2000). We distinguish between different forms of socio-technical processes according to broader criteria and specific types of socio-technical processes, which all fall under the proposed definition of process science. We also provide specific examples for each type of process. To this end, we suggest that processes can have different structures (e.g., causal, related to problem-solving or social practices, and performatory). Furthermore, socio-technical Table 1 Distinctions of process relevant for process science (drawing on and extending Rescher 2000) Criterion Distinction of process Types of Process Example Structure of process Causal processes (one event or process contributes to the production of another event or process) Seed germination Problem-solving process (do this, then that) Applying a business improvement method Social practice process Communication Performatory process Playing a video game Form of socio-technical process Equal involvement of social and technical components Office workflows More focus on technical components Car production More focus on social components University teaching Outcome of process Productivity Manufacturing process Solution Solving a business problem Social performance Gaining recognition on social media Language of process Continuous process Assembly line production Discrete process Deciding on a loan application Ad-hoc processes / one-off Disaster management Origin of process Owned process (follows from thing or subject, intentional) Call center agent Unowned process (non-intentional, do not come from subject or thing) Rise of a pandemic Page 7 of 14 Brockeetal. Process Science (2024) 1:1 processes can take on different forms depending on the focus that is placed on social and/or technical components. Socio-technical processes can lead to different outcomes (e.g., productivity, solution or the accomplishment of a social performance). As described in "Defining a Language to Study Change over Time" section, such processes can also be described through different languages (e.g., continuous, discrete and ad-hoc/ one-off processes). A final criterion refers to the distinction between “owned” and “unowned” sociotechnical processes (Rescher 2000). Processes are owned when they involve agency and intention. Unowned processes occur without the intentions of any agent. Owned and unowned socio-technical processes influence one another. Owned processes, such as production processes, influence unowned processes, such as environmental developments. Conversely, unowned processes have an impact on owned processes, as it has been shown dramatically by the Covid pandemic. As process scientists, we aim to study both forms of processes and how they interplay along a continuum where socio-technical processes are owned and unowned to different degrees (see Fig.1). In process science, the notions of owned and unowned processes extend further to owned and unowned activities or events within a socio-technical process. The unowned parts being those that cannot be controlled, changed or influenced by the owner of the process. In the studies of educational processes, for example, the actions of students can almost never be controlled whereas the actions of lecturers or institutes can. Hence, the parts of the processes pertaining to student behavior are unowned. A science ofdiscovering, explaining andintervening intosocio‑technical processes Process science welcomes all approaches to generating new scientific knowledge about socio-technical processes through deduction, induction, and abduction. Its key idea is that a focus on process advances our understanding of various phenomena because it directs our attention to underlying causal-temporal relations constituting a specific phenomenon. When we know why and how a specific process unfolds, we are better prepared to re-direct and change it, if needed. Process science subsumes three broad activities, which are depicted in Table2. Discovery emphasizes the detection of (emergent) dynamics constituting the phenomenon of interest. It can be challenging to detect emerging and evolving processual dynamics and their significance may be understood retrospectively (Chia 1999). The discovery activity capitalizes on the potentials of digital trace data to explore all sorts of phenomena (Lazer etal. 2020). Explanation aims at understanding the dynamics of socio-technical processes. It explains how and why they unfold. Explanation activities seek to identify cause-effect relations (Markus & Rowe 2018), specifically in relation to their situatedness, e.g., in temporal and spatial contexts. Access to a wide range of data sources will be beneficial, and again, the vast potentials associated with digital trace data may come into play (Lazer etal. 2009). Existing theory from about a domain will inform explanations, and analysis of new processes may inform that theory (Berente etal. 2019). An indepth understanding of a process enables predictions about the possible future states of the process. Thereby, one can anticipate patterns arising in the sequence of activities and events in a specific context, or the evolvement of a process in relation to certain Page 8 of 14 Brockeetal. Process Science (2024) 1:1 Discovery ERPBody Social IoT ProcessData … ProcessScience People -Roles -Capabilities -Preferences Task -Objectives -Constraints -Values Technology -Infrastructure -Architectures -Applications -Processing Capabilities Data Processes in Practice Theoretical Foundation -Theories -Constructs -Models -Methods -Instantiation Research Methodologies -Analysis Techniques -Formalisms -Measures -Validation Criteria Knowledge Scientific Knowledge Explanation Intervention Contribution Contribution Fig. 1 Process science research framework