Schedule and Cost Control in Dwelling Construction Using Control Charts

Send Orders for Reprints to [email protected] The Open Construction and Building Technology Journal, 2014, 8, 63-79 63 1874-8368/14 2014 Bentham Open Open Access Schedule and Cost Control in Dwelling Construction Using Control Charts Madelyn Marrero1,*, Antonio Fonseca2, Raúl Falcon3 and Antonio Ramirez-de-Arellano1 1University of Seville, Department of Building Construction II, Ave. Reina Mercedes 4A, Seville, Seville 41012, Spain; 2Self-employed Technical Architect; 3University of Seville, Department of Applied Mathematics I, Ave. Reina Mercedes 4A, Seville, Seville 41012, Spain Abstract: Methods to monitor the schedule and to control cost in dwelling construction projects are numerous and varied but commonly constitute an obstacle to a fast and agile response by construction managers, whose decisions require information to be comprehensive and summarized. A simple model to monitor these projects is proposed that can easily be implemented within control systems that are already in place. For the first time, process control charts are combined with cost control in dwelling construction in order to prevent overruns in terms of time and/or cost. The model facilitates the production supervision of construction contracts by regularly providing information on the work completed and the incurred cost of the production processes per period, through charting and/or summarizing this information in a manner consistent with statistical control charts. Finally, the manager can easily identify those processes which are off target by consulting control charts. Keywords: Cost control, dwelling construction, project management, scheduling, statistical process control, work breakdown system. 1. INTRODUCTION Over the last decade, an abundance of project control methods have been developed and a variety of software packages have become available to support their application, [1-3]. However, project planning and cost control still remain an open issue. Olawele and Sun [4] performed a survey to establish the current common practice of time and cost control in the UK construction industry, and include control methods and software applications in use. They identified the importance of cost and time control; already widely recognized by construction professionals. Their questionnaire survey reveals that 58% of respondents always apply time controls to their project and an overwhelming 84% implement cost control methods. These authors also identify the most popular time-planning and control techniques. The Gantt Bar Chart [5] is the most widely used, closely followed by the Critical Path Method [6]. Another commonly used technique is the Program Evaluation Review Technique, PERT [6]. The use of software support is widespread. Olewale and Sun [4] also detected that the cost control techniques most commonly used in practice, in about 70% of surveys, are project cost-value reconciliation, overall profit and loss, profit and loss at valuation dates, and reconciliation of actual versus forecast labour/plant/material. Despite the wide variety of control techniques and software available, construction projects are still subject to cost and time overruns. Researchers have been adapting previous techniques to more precise prediction levels through the *Address correspondence to this author at the University of Seville, Department of Building Construction II, Ave. Reina Mercedes 4A, Seville, Seville 41012, Spain; Tel: (34) 954-55-6667; Email: [email protected] implementation of stochastic instead of deterministic models. For example, approximations are used in the estimation of the mean and variance of the completion time for a PERT network and the task durations are approximated by employing normal distributions [7]. Other work related to stochastic models focuses on the planning and execution of construction projects, and account for the variability inherent in the duration and cost of the scheduled activities by simultaneously applying range estimation and probabilistic scheduling to historical data [1]. More recently, stochastic analysis by a multiple simulation analysis technique using Monte Carlo simulation has been used to evaluate the impact of noncritical activities for which the deterministic critical path method is insufficient [2]. In industrial processes, certain factors of stochastic behaviour exist whose influence on the process determines variations in the quality of the final product. In order to predict and control this influence, several methods have been developed, such as Statistical Process Control (SPC) [8] which uses statistical indicators to identify common causes of quality variation in manufacturing processes and to reduce the cost. An introductory survey on the subject appears in [9]. Specifically, the result of applying a repetitive process can be represented by a normal distribution whose mean and standard deviation are related to a single quality parameter. Multivariate analyses based on SPC, which are more common in practice, have also been developed over the last two decades [10-12]. In both uni- and multi-variate analysis, statistical measures, such as mean and standard deviation, determine the tolerance levels of the variation with respect to the target, which can be represented in a control chart with regard to time. These charts clearly and promptly show abrupt fluctuations and indicate whether they are due to ei- 64 The Open Construction and Building Technology Journal, 2014, Volume 8 Marrero et al. ther a random or an unknown factor which should consequently be taken into account [13]. Thus, managers can adopt corrective decisions in order to reach the final target. More specifically, in the construction industry, SPC has successfully controlled road construction [14] and, combined with the Earned Value Management (EVM) method, project cost and schedule performance have been controlled [15]. In the current paper, a new model to monitor the schedule and to control cost is proposed which adapts, for the first time, SPC control charts to dwelling construction. Dwelling construction usually consists of simple projects for which repetitive work takes place, which makes it possible for the tools already being used in the control of the manufacturing process to be adapted to control the construction process. Once the schedule is established, by either a deterministic or stochastic approach, it can be controlled and deviations can be corrected. In the present model, schedule and cost control in projects are handled simultaneously. Many models handle resources separately and independently from the time factor, and the cost model is disconnected from the time model, thereby rendering cost and time control difficult and imprecise. Researchers have identified the benefits from integrating cost and schedule control. This is the case of management control by means of the Earned Value Management System (EVMS) which determines the actual costs accumulated and compares them to the earned value [16]. This value is used as a baseline to which the planned schedule (budgeted cost for work scheduled), and the actual cost (actual cost of work performed), are compared in order to measure the schedule performance and cost performance, respectively. The results of performance variances and indices are used for further analysis, which include identifying latent risks, and re-scheduling the remaining work [3]. Following the EVMS approach, the main objectives include the development of a simple tool that allows project managers rapid and agile responses to out-of-budget and/or out-of-schedule processes on the construction site. Three key characteristics are taken into account: a simple structure, ease in data collection, and ease in following the process. 2. METHODOLOGY The main divisions of the model are similar to those used in EVMS: data input, work scheduling, data transformation into reference values, and comparison between the work actually carried out and planned work completion. The steps are the following: • Enter the data of the initial schedule. • Plan work in terms of cost, using the project schedule. • Plan work quantities. • Distribute expected costs and quantities over time periods. • Enter actual values of work performed and cost produced. • Compare actual data values against planned data values. • Calculate cost deviations, and quantity deviations. • Generate reports for each time period. • Edit, revise or modify the schedule according to new forecasts. • Once the schedule is corrected, recalculate the new planned work. The first step, Enter the data of the initial schedule, refers to the original project budget and its planning, (see Fig. 1). The classification of the costs is obtained from this budget which is organized by means of a work breakdown structure (WBS). The schedule is obtained from the first plan of the project, which allows the work to be distributed into steps for each time period: Plan cost and Plan work quantities in (Fig. 1) constitute the "Schedule". This distribution is then used to calculate the accumulative cost and quantities for each period in the step Distribute expected cost and quantities, and is represented in (Fig. 1) as the "Images". Another set of data is obtained by means of surveys at the construction site, and contains information on the quantities of work performed and costs incurred. This data is compared with the planned values, by subtracting the planned values from the actual values, thereby indicating that an overrun is taking place if a positive value is obtained. The differences during each period are transformed into indicators in the "Evaluation" step in (Fig. 1), and are represented in a control chart. Finally, a report is written. The schedule is readjusted accordingly and a new work distribution is generated for future comparisons. In the following sections, we define and study each step of (Fig. 1) in more detail. 2.1. The Data As established in the previous section, the data needed is the original project budget and planning. The costs are obtained from this budget and the schedule from the initial project time chart. One major aspect in process control is data measurement and accuracy. The method, structure, data, and accuracy of detailed measurement may vary depending on the specific characteristics of a project. This situation can lead to misinterpretation of the project status, especially under a multi-project management environment. In order to overcome this issue, Jung and Kang [3] propose a standard progress measurement package which addresses issues for the standardization of the work breakdown structure (WBS). These authors have identified the measurement of the level of progress as a critical factor in terms of the workload required to maintain both the control system and the accuracy. In order to address this issue, the level of detail for progress measurement should be carefully selected as a trade-off between the workload and accuracy, by incorporating strategy, objectives, and management policy of construction projects. All WBSs have the same goals and similar methodologies. The basic concept in all of these systems is to divide a complex problem into simpler parts that can then be aggregated to define the development of a complete construction. Many researchers have been working on the development of construction information classification systems (CICS): Kang et al. [17] for civil work; Eldin [18]; and Jung et al. [3] who address cost and scheduling simultaneously. There are several international CICS, of which the most frequently used include: Masterformat [19], Uniformat II [20], the Civil Engineering Standard Method of Measurement [21], CI/SfB [22], Uniclass [23], and Omniclass [24]. Schedule and Cost Control in Construction The Open Construction and Building Technology Journal, 2014, Volume 8 65 Fig. (1). Steps of the control process. In particular, the present model uses the Andalusia Construction Information Classification System (ACICS) [25]. Its most extended usage is for estimating cost in dwelling construction and it is mandatory in all public developments in Andalusia, Spain. This system divides work units into a hierarchical organization. The highest level is the construction site, L1 in Table 1. The next divisions are called chapters, L2 in Table 1, and each represents a construction process: Demolition, Earthwork, Foundations, Water disposal, Structures, Partitions, Roof, Installations, Insulations, Finishes, Carpentry, Glass and Polyester, Coating, Decoration, Urbanization, Safety, and Waste Management. The subsequent divisions are the sub-chapters. For example, the chapter called "Installations" is divided into the following sub-chapters: Air conditioning, Electricity, Water, Communications, Gas, Electro-mechanic appliances, Fire protection, and Illumination [26]. Since time and cost are closely related, it can be assumed that a close relationship exists between the cost estimate and schedule. However, it is not easy to numerically evaluate the extent to which these parameters affect each other, due to differences in their breakdown structures [27]. Even though ACICS is developed for the generation of construction budgets, it can also be used for the definition of the schedule. In order to double-check the work development, a second classification system is proposed to control quantities instead of cost. The classification, complex groups, is employed to control construction and demolition waste successfully [28, 29]. The complex groups are at the same level as sub-chapters in the ACICS hierarchy. The existence of complex groups allows the combination of similar work units within a chapter, which would not be possible if ACICS sub-chapters were used. For example, earthwork excavations, such as the foundation (02AVV00002), trenches (02ZMM00002), and pits (02PMM00002), have different unitary costs and belong to 66 The Open Construction and Building Technology Journal, 2014, Volume 8 Marrero et al. separate sub-chapter classifications in ACICS: 02A, 02Z, and O2P, respectively, but can still be grouped together into a "complex group", 02EX, since the main activity is common to all of these work units and they can take place during the same period of time, thereby rendering it possible for them to be controlled simultaneously. Table 1. Internal classification structure. Class Level Definitions L1. Construction site All the constructive elements that make up a construction site. L2. Chapter Element sets with a common characteristic. e.g.: 05. Structures. L3. Sub-chapter Chapter division into smaller sets with a common characteristic. e.g.: 05H. Concrete. L4. Section Sub-Chapter division into smaller sets with a common characteristic. e.g.: 05HH. Reinforced concrete. L5. Group Section division into smaller sets with a common characteristic. e.g.: 05HHJ. Reinforced concrete beam. L6. Work unit Group division into unitary elements. e.g.: 05HHJ00001 m3 Concrete HA-25 in.... The budget is introduced into two different scenarios, work quantities and cost, which is possible since the budget is the result of the addition of all work units that constitute the project, multiplied by their corresponding unitary price. For the first scenario, the sub-chapter cost is defined as, C s T =Q i i=1 l P i (1) where Cs T is the total cost of sub-chapter “s”, and Q i and P i are the quantity and price of the work unit “i”, which is part of the level “l” total in sub-chapter “s”. The second scenario is for the complex quantities, which are formed of similar work units within the same chapter of the ACICS, Qg T=Q i i=1 m  (2) where Qg T is the total quantity of complex group “g”, and “m” is the total number of divisions that are part of the complex group, which implies that all work units in a complex group have the same units of measurement, (cubic metres of concrete, tons of steel, etc). In order to be able to add and compare widely varying quantities, a coefficient is defined which transforms all quantities into a similar magnitude, Dg, D g =P g P max (3) where Pg is the average unitary price of the complex group "g" which is measured in monetary units, divided by the corresponding measurement unit, and where Pmax is the unitary price of the most expensive work unit in the project, divided by the corresponding measurement unit. For dwelling construction, the most expensive unitary price, Pmax, is normally the average price of all bathroom appliances. This approximation establishes the quantity importance in terms of its proportional unitary price with respect to the most expensive item in the project. The measurement units of Dg are therefore D g =u max u g      (4) where umax is the unit of the most expensive item, and ug is the group “g” measurement unit. Finally, all complex groups lie within the same complex measurement unit, Q g D g =u g u max u g     =u max (5) Several ASICS work units, which require similar materials, labour, and machinery, are grouped together and their cost is defined as an average. This average cost is subsequently divided by the most expensive work unit in the project, thereby obtaining Dg. For example, in a project where bathtubs are the most expensive work unit at 200 Euro/unit, then 30 m3 of refilling soil (at 5 Euro/m3) is transformed into 0.01*30 = 3 units. On the other hand, 30 m3 of reinforced concrete trenches (at 200 Euro/m3) is transformed into 30 units. Finally, the Dg factor is able to establish that 30 m3 of concrete is more important than 30 m3 of refilling soil due to its unit costs. The advantage of the alternative approach is that different cost estimation is obtained of the cost of the work performed on the construction site, rather than just the cost that is certified and ready for payment. The new quantity approach is independent of market conditions and cost fluctuations since the unitary costs are transformed into a proportion of the original market price, at the beginning of the project, with respect to the most expensive unitary price: the Dg factor. The quantity approach is easily implemented due to the hierarchical nature of the ASICS structure: the families of similar work belong to the same chapter and sometimes to the same sub-chapter. This is the case of reinforced concrete slab (03HAL10002) and reinforced concrete wall (03HAW10007), which belong to sub-chapter 03H and to "complex quantity" 03HA. 2.2. The Schedule Once the project budget of cost and quantities has been separately input following the ACICS hierarchy, the next step is to define the schedule, also in terms of ACICS. The schedule, which is additional data, is input in two scenarios, one for the cost and another for its corresponding quantities, in terms of ACICS and the complex quantity classification system, respectively. The work quantities and cost are then distributed over the duration of the construction, thereby creating reference scenarios for the actual labour/plant/material versus those planned. Schedule and Cost Control in Construction The Open Construction and Building Technology Journal, 2014, Volume 8 67 The work units are distributed in time periods, Tk, where the time distribution can be either in terms of manpower hours, days, weeks, machine hours or of any other characteristic that is proportional to the time necessary to perform the work [27], or in terms of program measurement packages [3]. The work amount in period “k” is called weight, Wsk and Wgk, where “s” or “g” is the subchapter or group of which the weight is a component, respectively. The entire project is planned at the beginning of the controlling process and the planning is revised when the actual work carried out deviates from the work planned. 2.3. The Images Once the project budget and schedule have been input, then the next step is to create the reference image scenario. In this section the term image refers to the representation of the scheduled costs and complex quantities in time. There is an image, planned cost per time period, for each sub-chapter which becomes the reference value. Another image is generated for the complex quantities per period [27]. Thus, if “n” is the total number of periods of the project, then the cost reference images are defined by: C sk P =C s T W sk W sk k=1 n  (6) where the planned cost Csk P of sub-chapter “s” in period “k” is obtained as the product of the total cost Cs T of sub-chapter “s” obtained from the budget multiplied by the relative proportion of the work to be carried out during such period and sub-chapter, divided by the sum of all the proportions. In a similar way, the second scenario is defined for the quantities, whereby the reference images are obtained from, Q gk P =Q g T W gk W gk k=1 n  (7) where Qgk P is the planned quantity of complex group “g” in period “k”, Qg T is the total amount of the complex group “g” obtained from the original budget, and Wgk is the proportion of the work of complex group “g” to be carried out in period “k”, and “n” is the total number of periods. 2.4. Survey Once the quantity and cost scenarios are established, then the actual work carried out is noted in a survey during each construction period, which can be defined in terms of days, weeks, or months, depending on the level of time control. The cost survey consists of information about the completed work on the construction site for which the subcontractors are to be paid. The survey is organized in the WBS budget, which is familiar to the construction managers; this aspect facilitates its implementation. The information is normally checked on the construction site in order to control cost as established in the Spanish law on Public Sector Contracts [30]. 2.5. The Comparison Once the survey data is input, the differences between the images for planned quantities and cost and their actual values are determined. Positive differences show overruns and negative differences show work carried out under budget, and constitute the first control level for the construction site manager. In the succeeding control level, deviation indicators are defined, Vsk =Csk RCsk P Csk R k=1 n  s=1 p  100 (8) V gk =Q gk R Q gk P () D g Q gk P D g k=1 n  g=1 q  100 (9) where Vsk is the cost deviation indicator of sub-chapter “s” during period “k” for “p” total sub-chapters and “n” total periods. The indicator is the percentage of cost deviation with respect to the total project budget, and Csk R and Csk P are the actual and the planned costs, respectively. In the second scenario, Vgk is the quantity deviation indicator, Qgk Rand Qgk P are the actual and the planned quantities, respectively, of group “g” during period “k” for “q” total groups and “n” total periods. The indicator Vgk is the quantity deviation as a percentage with respect to the total project budget. The total number of groups, g, and the total number of sub-chapters, p, that are part of a chapter, are not fixed since not all dwelling projects are made up of the same number of sub-chapters or groups. The indicators Vsk and Vgk are positive values when the process is more expensive than planned or ahead of schedule, respectively, and negative otherwise. Only one indicator is reported per chapter, and is determined as an average value of all the sub-chapter indicators. This assumption allows a simple calculation of the total chapter indicator. For chapter “c” during period “k”, the deviation indicators are defined as: V ck C = V sk s=1 p  p (10) V ck Q = V gk g=1 q  q (11) for cost and quantity respectively, where “p” is the total number of sub-chapters and “q” is the total number of complex groups that are part of chapter “c”, during period “k”. The project deviations, during period “k”, are calculated as: VPk C= Vck CCsk p s=1 p  c=1 r  Csk p s=1 p  c=1 r  (12) 68 The Open Construction and Building Technology Journal, 2014, Volume 8 Marrero et al. VPk Q= Vck QQgk pDg g=1 q  c=1 r  Qgk pDg g=1 q  c=1 r  (13) where “r” is the total number of chapters that make up the project. 2.6. Evaluation and Report The results need to be reliable, facilitate a rapid knowledge of the situation, and allow an in-depth analysis of the causes. To this end, two analyses are carried out when the survey is compared against the reference image. First a simple report is made and then a control chart is generated. The control charts are the representation of the indicators of each chapter for each time period, equations (10) and (11), divided by the corresponding standard deviation for the same period. The standard deviation is determined for the population made up by the individual indicators of the sub-chapters. There is a standard deviation determined per time period. First, the mean of the indicators of the complex quantities and sub-chapter costs are calculated, respectively, V k C = V sk s=1 r  r ; V k Q = V gk g=1 r  r (14) where “r” is the total number of sub-chapters or total of complex quantities during period "k". The standard deviation during the period "k" is then calculated, Sk Q= Vsk Vk Q () 2 g=1 r  r1 ; Sk C= Vsk Vk C () 2 s=1 r  r1 (15) The control chart centre is set at 0, which indicates that there is no difference between planned and actual values. The upper and lower control limits are established at ±1.5 S k C,±1.5Sk Q, for cost and quantity control, respectively. The zone inside these control limits represents approximately 86% of the population in a normal distribution. In SPC, three periods that show an increasing trend towards the upper limit or a decreasing trend towards the lower limit, even though they may remain within the control limits, indicate that the processes need to be corrected [8]. 3. CASE STUDY The previously explained model is used to control the construction of 6,553 m2 for 48 dwellings in social housing in Cadiz, Spain. The buildings are of six storeys each, with an underground parking area, and a ground level plus another 4 storeys. Each dwelling has a lounge, kitchen, hallway, 3 bedrooms, and one bathroom. The building foundation is a reinforced concrete pad and has reinforced concrete walls for the underground parking area. The rest of the structure consists of concrete footings and one-directional beams. The façade is made of a double brick wall with an insulated cavity, coated with waterproof mortar on the outside, plaster on the inside, and both sides are painted. The finishes are typical for social housing: terrazzo floor, aluminium-framed windows, and wooden interior doors. The roof is flat and transmittable, except for the stairway case which has a sloped roof with Arabic tiles. For the project duration, there is no material nor manpower scarcity. The construction location has no accessibility problems. In Tables 2 and 3, the total amount of each sub-chapter is distributed throughout the various time periods. The "code" column refers to the sub-chapter code and the "description" is its corresponding short definition as defined in ASICS. Since the work schedule can be revised several times during the project development, the work already completed during old schedules is entered in the third column, named “base”, and the remaining work to be distributed in the new schedule is entered in column 4, named “work”. In this way the manager is kept up to date with the amount of remaining work. The column denoted “total” is the sum of all weights in the sub-chapter, Totals=W sk k=1 n  (16) Finally, columns "T1, T2, ..., Tn" represent the weights corresponding to the work distributed in “n” time periods. Tables 2 and 3 represent the schedule in terms of cost of the first four months of the project. The sub-chapter costs include manpower, materials, and machinery costs. However, the model allows the control level to be focused on any specific aspect, for example, to control only material cost. This is possible since ASICS defines each type of cost, (manpower, material, and machinery), separately within a unitary price. The decision depends on the critical aspect which needs to be controlled more closely. In Tables 2 and 3, the work distribution of chapters 02 and 03 (except for sub-chapter 03H) takes place almost completely within the first 4 months, in time periods T1 to T4 in the project, and hence the four periods add up to 100 as shown in column "Total". Work of other sub-chapters is not 100 % completed during those four periods and only a partial planning is represented, which adds up to less than 100. In the case study, Wsk represents the percentage of work which takes place during period "k". However, other weighting values can be used, such as concrete cubic metres consumed in the foundation during each period; in fact, during each period, any characteristic that describes the work intensity during the project execution can be used. Finally, the total amount of the characteristic is added up in the "Total" column, and the percentage corresponding to each period is determined and is used to define the reference values or images in terms of Euros. Similar tables are generated for the schedule in terms of complex quantity classification in Tables 4 and 5. The main difference between the two tables is the demolition chapter, which does not exist as a "complex quantity" owing to the difficulties in developing new combinations from the cost classification system. In Tables 4 and 5, the first and second columns, “code” and “description”, are the code from ASICS and a short description, respectively. The "description" contains the measurement unit of the quantities. Additionally, the new Schedule and Cost Control in Construction The Open Construction and Building Technology Journal, 2014, Volume 8 69 Table 2. The work schedule in terms of cost for the first three chapters and first four months (I). Code Description Base Work Total T1 T 2 T 3 T 4 01A Bricks 0 01C Foundations 0 01E Enclosures 0 01L Installations 0 01K Carpentry and safety 0 01Q Roof 0 01R Finishes 0 01S Water disposal 0 01T Preparation work 631.06 100 100 01X Structure 0 01W Others 0 01 DEMOLITION 631.06 02A Open excavations 2 027.03 100 100 02P Pads 39 547.92 100 57 25 18 02R Refilled and compacting 454.24 100 57 25 18 02T Transport 74 127.52 100 57 25 18 02W Others 100 57 25 18 02Z Trenches 105.20 100 57 25 18 02 EARTHWORK 116 261.91 03A Reinforced steel 0 03C Special foundations 0 03E Formwork 0 03H Concrete 74 127.52 100 20 35 03W Others 0 03 FOUNDATIONS 74 127.52 04C Hanging pipelines 3 319.37 100 04E Underground pipelines 9 995.13 100 04V Vertical pipelines 854.54 100 04W Others 2 126.23 100 04 WATER DISPOSAL 16 295.26 05A Steel 0 05F Concrete slab 234 691.39 100 05H Reinforced concrete 388.34 100 05M Wood 0 05W Others 6 520.47 100 05 STRUCTURES 241 600.20 06A Arches and vaults 0 70 The Open Construction and Building Technology Journal, 2014, Volume 8 Marrero et al. (Table 2) contd…. Code Description Base Work Total T1 T 2 T 3 T 4 06B Concrete blocks 0 06C Stone blocks 0 06D Brick partitions 22 949.49 100 06L Brick walls 111 698.21 100 06P Prefabricated 0 06W Others 1 031.33 100 06 PARTITIONS 135 679.03 07H Horizontal roof 19 275.85 100 07I Sloping roof 21 817.44 100 07W Others 3 335.64 100 07 ROOFS 44 428.93 Table 3. The work schedule in terms of cost for the first three chapters and first four months (II). Code Description Base Work Total T1 T 2 T 3 T 4 08C Air conditioning 0 08E Electricity 1 970.07 100 08F Water 1 351.57 100 08L Gas 0 08M Electro-mechanic appliances 20 306.64 100 08P Fire protection 327.73 100 08V Audiovisual 17 388.15 100 08S Hygiene 0 08W Others 0 08 INSTALLATIONS 41 344.14 09A Acoustic 0 09I Waterproofing 454.24 100 8 09T Thermal 0 09W Others 0 09 INSULATION 454.24 10A Cladding 26 124.89 100 10C Continuous 49 651.67 100 10L Lightweight 100 10P Stairs 5 399.16 100 10S Floor 70 081.36 100 10T Ceiling 100 10W Others 102 894.01 100 Schedule and Cost Control in Construction The Open Construction and Building Technology Journal, 2014, Volume 8 71 (Table 3) contd…. Code Description Base Work Total T1 T 2 T 3 T 4 10 FINISHES 254 151.08 11A Steel 598.31 100 11L Alloys 3 503.23 100 11M Wood 21 817.13 100 11S Security and safety 24 770.80 100 11W Others 0 11 CARPENTRY 50 689.47 12L Glass: large format 0 12V Glass: small format 100 12W Others 0 12 GLASS AND PLASTICER 13E Exterior 3 060.51 100 13I Interior 11 603.86 100 13W Others 0 13 COATING 14 664.36 14M Furniture 163.24 100 14W Others 0 14 DECORATION 163.24 15A Sewers 0 15C Safety signalling 0 15E Electricity 0 15G Gas 0 15J Gardening 0 15M Earthmoving 0 15P Pavement 0 15S Water supply 0 15T Telephone 0 15V Furniture 0 15W Others 0 15 URBANIZATION coefficient Dg is defined for the current project in the column with the same name. In the project, each sub-chapter average price is divided by the most expensive item in the project: 03HAZ00030 trenches of reinforced concrete, their cost being 168.66 Euro/m3. For example, the 02EX group has an average cost of 5.06 Euro/m3 and its corresponding Dg is 0.03. Once the schedule is defined, the next step involves the collection and organization of the actual data at the construction site. After the data is collected it is input into Tables 6 and 7, column "CR" being the real values paid up to that period. A comparison is then performed between the planned scenario (CP) and the actual scenario (CR). The CP column is determined by means of equations (6) and (7), for example the 03H group during period T4 is the result of the accumulation of previous periods, CP = 74 127.52*(0.0+0.0+0.35+0.20) = 40 770.00 Euros (17) 78 The Open Construction and Building Technology Journal, 2014, Volume 8 Marrero et al. C03: Foundation has been significantly off schedule since month 2 (T2). The chapter needs to be checked and rescheduled. The same also happens with Chapter C04, Water disposal, during month 4 (T4). Fig. (2). Statistical process control of project cost. On the other hand, the control chart for quantities (Fig. 3) shows slightly different behaviour since market cost fluctuations are not taken into consideration and only the quantities of work completed are considered. Moreover, the indicators are determined with respect to the total corrected quantities (by means of the Dg correcting factor) in the project and not with respect to the total project cost. Chapter 03 Foundation is off target during month 2 because the work started earlier than planned, but this deviation is not significant in month 3 (T3) and 4 (T4), since Chapter 06 is the important quantity in the project, even though it is not an important cost in (Fig. 2). Fig. (3). Statistical process control of project quantities. The tool improves the control of cost and schedule, especially at manager level, since a simple general graph indicates the overall construction site behaviour or that of several construction sites at the same time. The most important contribution of the control chart, as in any application of SPC, is that it provides a simple visual support to managers; this is especially significant under a multi-project management environment, since the representation remains in an identical form for all the projects. The same WBS is used, the indicators are dimensionless and independent of the project size, and the control limits can be set to identical values in all project graphs. The SPC general rules apply, as in industrial processes where SPC is commonly used: when three points are under the control limit but with increasing deviations or a major fluctuation takes place, then the process is considered out of control and this can be visually detected. CONCLUSION To guarantee the effectiveness of the model, the following objectives are met: 1. The model implementation is simple and quick; a commonly used spread sheet is the only software needed. 2. The information required in order to use the model is easy to obtain and causes no obstacles to the daily workload of the data collector. 3. The results are reliable since they are based on the cost data, which is a contractual commitment, and are represented by sensitive indicators that are easy to understand: mean and standard deviation. 4. The structure of the results facilitates rapid understanding of the general situation and also allows an in-depth analysis of the causes. 5. The cost of implementing the system is significantly lower than the savings provided by its use. Future work involves the development of the classification system so that it includes the control of construction projects other than that of dwellings, such as libraries, gymnasiums, and school and college buildings. Furthermore, another alternative to this approach could involve the organization of the budget and activities through a process classification system instead of the present cost classification system. CONFLICT OF INTEREST The authors confirm that this article content has no conflict of interest. ACKNOWLEDGEMENTS Declared none. REFERENCES [1] L.J. Isidore and W.E. Back, “Probabilistic optimal cost schedule”, J. Constr. Eng. Manage., vol. 127, pp. 431-437, June 2001. [2] Z.I. Sakka and S.M. El-Sayegh, “Float consumption impact on cost and schedule in the construction industry”, J. Constr. Eng. Manage., vol. 133, pp. 124-130, Feb. 2007. [3] Y. Jung and S. Kang “Knowledge-based standard progress measurement for integrated cost and schedule performance control”, J. Constr. Eng. Manage., vol. 133, pp. 10-21, Jan. 2007. [4] Y.A. Olawale and M. Sun, “Cost and time control of construction projects: inhibiting factors and mitigating measures in practice”, Constr. Manage Econ., vol. 28, pp. 509-526, 2010. [5] H.L. Gantt, Organizing for Work. Howe: New York, 1919. Harcourt, Brace. [6] J.J. Moder and C.R. Phillips, Project Management with CPM and PERT. Van Nostrand Reinhold: New York, 1970. [7] M.A. Cox, “Simple normal approximation to the completion time distribution for a PERT network”, Int. J. Proj. Manage., vol. 13, pp. 265-270, Apr. 1995. [8] W.A. Shewhart, “Statistical method from the viewpoint of quality control”. The Graduate School of the Department of Agriculture: Washington DC, 1939. Schedule and Cost Control in Construction The Open Construction and Building Technology Journal, 2014, Volume 8 79 [9] G.E.P. Box, A. Luceño, M.C. Paniagua-Quiñones, Statistical control by monitoring and adjustment. Hoboken, NJ: 2nd ed. Wiley Series in Probability and Statistics. John Wiley & Sons, Inc., 2009. [10] J.F. MacGregor and T. Kourti, “Statistical process control of multivariate processes”, Control Eng. Pract., vol. 3, pp. 403-414, Mar. 1995. [11] R.L. Mason and J.C. Young, Multivariate statistical process control with industrial applications. SIAM: Philadelphia, 2002. [12] S. Bersimis, S. Psarakis and J. Panaretos, “Multivariate Statistical Process Control Charts: an Overview”, J. Qual. Reliab. Eng. Int., vol. 23, pp. 517-543, May 2007. [13] G.M. Smith, Statistical process control and quality improvement. Pearson-Prentice Hall, Upper Saddle River, New Jersey, 2004. [14] K.M. Nassar, W. M. Nassar and M.Y. Hegab, “Evaluating cost overruns of asphalt paving project using statistical process control methods”, J. Constr. Eng. Manage., vol. 131, pp. 1173-1178, Nov. 2005. [15] S.S. Leu, and Y.C. Lin, “Project performance evaluation based on statistical process control techniques”, J. Constr. Eng. Manage., vol. 134, pp. 813-819, Oct. 2008. [16] Electronic Industries Alliance (EIA). Earned value management systems. Publication no. ANSI/EIA-748. EIA: Arlington, VA, 1998. [17] L.S. Kang and B.C. Paulson, “Information management to integrate cost and schedule for civil engineering projects”, J. Constr. Eng. Manage., ASCE, vol. 124, pp. 381-389, May 1998. [18] N.N. Eldin, “Management of engineering/design phase.” J. Constr. Eng. Manage., vol. 117, pp. 163-175, Jan. 1991. [19] Construction Specifications Institute, Masterformat manual of practice (MP2-1). Construction Specifications Canada (CSI/CSC): Alexandria, VA, 1983. [20] Construction Specifications Institute, UniFormatTM: A Uniform classification of construction systems and assemblies. Construction Specifications Institute: Alexandria, VA, 1998. [21] CESMS. Civil engineering standard method of measurement. London: 3rd ed., Thomas Telford, 1991. [22] A.R. Jones, CI/SfB construction indexing manual. Royal Institute of British Architects (RIBA) Publications Ltd.: London, UK, 1987. [23] Uniclass-unified classification for the construction industry. National Building Specification Service Ltd.: London, UK, 1996. [24] International Organization for Standardization (ISO) TC 59 SC/WG2, Classification of information in the construction industry. ISO Tech. Report 14177, Oslo Norway, 1994. [25] Consejería de Vivienda y Ordenación del Territorio de la Junta de Andalucía. Base de Costes de la Construcción de Andalucía 2013 (Andalusian Construction Costs Database 2013, Seville, Spain). <http://www.juntadeandalucia.es/viviendayordenaciondelterritorio/ www/jsp/estatica.jsp?pma=0&pmsa=0&e=planificacion/publicacio nes/banco_precios_construccion/bcca08rev2b/bcca08rev2b.html> [Accessed Jan 10, 2014]. [26] M. Marrero and A. Ramirez-de-Arellano, “The building cost system in Andalusia: application to construction and demolition waste management”, Constr. Manage. Econ., vol. 28, pp. 495–507, 2010. [27] H.S. Lee and K.J. Yi, “Application of mathematical matrix to integrate project schedule and cost”, J. Constr. Eng. Manage., vol. 125, pp. 151-742, May 1999. [28] J. Solis-Guzman, M. Marrero, M.V. Montes-Delgado and A. Ramirez-de-Arellano, "A Spanish model for quantification and management of construction waste", Waste Manage., vol. 29, pp. 2542- 2548, 2009. [29] M. Marrero, J. Solis-Guzman, B. Molero-Alonso, M. Osuna- Rodriguez and A. Ramirez-de-Arellano, Demolition waste management in Spanish legislation. Open Construct. Build Tech. J., vol. 5 (Supl 2-M7), pp. 162-173, December 2011 [Online] Available; http://www.benthamscience.com/open/tobctj/articles/V005/ SI0123TOBCTJ/162TOBCTJ.pdf [Accessed Jan 21, 2013]. [30] Spanish Government – 2007 General Courts, Ley 30/2007, de 30 de octubre, de Contratos del Sector Público. (Law 30/2007, October 30, Contracts of the Public Sector). Edited by Official Bulletin, Spanish Government, Madrid, Spain. Received: January 30, 2014 Revised: May 06, 2014 Accepted: May 08, 2014 © Marrero et al.; Licensee Bentham Open. This is an open access article licensed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/- licenses/by-nc/3.0/) which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited.