Informal Decision Making for Construction Projects Inspired by Flocks of Birds S. Alireza ABBASIAN HOSSEINI1 Min LIU2 Simon M. HSIANG3 Abstract: The formal decision making (DM ) in construction industry (such as PERT and CPM ) is based on the assumption that the participant in the center of the activity takes the responsibility of decision and then pass the results to the next particip ant. However, in practice, there is an informal DM system for when and how to conduct construction activities simply by establishing a frame of reference based on the Social Network (SN) of specialty trades. Formal DM usually comes with clear objectives for efficiency and effect iveness. On the other hand, the informal DM is usually done through benchmarking and copying the best performer in the SN. In this paper, we illustrate how the informal DM is done in construction projects and show how it can provide easy and reliable steps for impro ving performance. A fast-track residential construction project ($2.5M , duration: 6 month) from Raleigh area, NC, involving eleven major specialty trades was studied. We use two complementary approaches to explain the informal DM in the construction projects: Data Envelopment Analysis (DEA) approach to focus on the optimum directions and trajectories and the SN approach to emphasize the relative positions and relationships within a group. The findings will tell us how the informal DM is done in a construction project and whom should be copied by an underperforming specialty trade to reach better performance.
Keywords: social network, data envelopment analysis, interdependency, decision making. 1. INTRODUCTION
The highly coordinated movements of flocks of birds (or schools of fish) are among the most fascinating phenomena found in nature. While some hypothesize that mysterious communication is involved, in reality such formation is coordinated by the moment to-moment decisions of individual birds. Simply speaking, each bird prefers to instantly copy any changes in speed or direction of his (successful) neighbor birds rather than investigating the best flying direction or velocity by itself (Thompson et al. 1974). It is amazing how complex behavior can be generated by the application of this simple decision making rules, and we speculate we have similar situations in the construction industry. Although the construction industry has existed for thousands of years, the underlying rules or mechanism governing those interactions and decision making have not been well understood or explained. We commonly understand and accept a formal decision making (DM ) procedure (such as PERT and CPM ), which specifies the responsibility and risk of each participant and draws boundaries for him/her using contract language. Although the formal DM usually comes with clear objectives for efficiency and effectiveness, without suitable coordination and collaboration, many management problems may be occurred in a project (Wambeke et al., 2012) Indeed, in practice, there is an informal DM system for when and how to conduct construction activities simply by establishing a frame of reference based on the Social Network (SN) of specialty trades. Informal DM , like finding direction by a bird in a flock, is usually done through benchmarking and copying the best performer in the SN without 1 Ph.D. Candidate, Department of Civil, Construction and Environmental Engineering, North Carolina State University, Raleigh, NC 27695; PH (919) 917-4547; Email:
[email protected] 2 Assis tant Professor, Department of Civil, Construction and Environmental Engineering, North Carolina State University, Raleigh, NC 27695; PH (919) 513-7920; Email:
[email protected] 3 Derr Professor, Industrial Engineering, Texas Tech University, Lubbock, TX 79409; PH (806) 742-3543; Email:
[email protected]
having the required knowledge of being efficient. This informal SN based decision making process has not been documented in industry handbooks or academic research work. How do specialty trades in construction engineering act as a flock of birds? Inspired by flocks of birds and construction specialty trades, the purpose of study is to illustrate how the informal DM is done in construction projects and show how it can provide easy and reliable steps for improving performance. In doing so, we use two complementary approaches to present specialty trades’ organization (the flock of birds’ movement): 1) Data Envelopment Analysis (DEA), which focuses on the optimum directions and trajectories and 2) SN approach, which emphasizes the relative positions within a group.
2. LITERATURE REVIEW Data Envelopment Analysis (DEA) and Social Network Analysis (SNA) have been applied in various researches as two powerful decision-making tools. Useful managerial information provided by DEA can improve the productivity of service business (Zhu 2009); and the interdependency embedded in construction processes leads to the motivation to use social network analysis to study the interaction between the various trades. Although we can find some researches on implementing DEA or SN in construction projects, there is no specific research on implementing them together. In our research, DEA and SN play a complementary rule to illustrate the informal DM in construction industry (construction projects). In two following sub sections, we briefly discuss the application of DEA and SN.
2.1. Data Envelopment Analysis (DEA) DEA measures the productivity by the ratio between a weighted sum of inputs and a weighted sum of outputs. It uses operations research techniques to automatically calculate the weights assigned to the inputs and outputs of the production units being assessed (Kabnurkar, 2001).
2.2. Social Network (SN) Analysis SN analysis, introduced by M oreno (1960), has been known as a methodology to determine the conditions of social structures by investigating the interactions, relations and interrelationships of a set of actors (nodes) (De Nooy et al. 2005, Park et al. 2011). A SN refers to a pattern of ties that exist among different entities (nodes) such as countries, states, organizations, etc. (Wambeke et al., 2011a). While classic SN research has concentrated on sociological networks, it has been applied to many research fields (such as aerospace equipment, automotive bodies, and computer and office equipment) with the goal of investigating various relationships among organizations and individuals (Park et al. 2011). SNA has become important within the engineering and construction field recently due to significant attention to some concepts such as trust and communication between project participants (Chinowsky et al. 2008). Wambeke et al. (2011a) believed that an underlying social network of trades exists in a construction project and its recognition can contribute to project success. However, there are just a few research studies using social networks in the construction industry. Chinowsky et al. (2008), us ing SNA, modeled the information passed through the team members to reduce the uncertainty during construction. Wong et al. (2010) investigated the differences between global and domestic projects through robust project network designs. In another research endeavor, SNA is suggested for use w ith learning dynamics and building information modeling (BIM ) (Taylor et al. 2009). Wambeke et al. (2011a) outlined the procedures to identify a social network of construction trades and determine its key members. Wambeke et al. (2011b) also reduced variation by creating a decision support system through quantitative variation analysis and identification of the social network of trades.
Technically speaking, DEA gives an efficiency score to each of DM Us and establishes a common region (Empirical Frontier) based on the ratios between inputs and outputs (Figure 1-a). At the frontier, DM Us are the most optimized for the variables (inputs and outputs) and they get the maximum efficiency score (1.00) from which a relative classification of the DM Us becomes possible. Therefore, from DEA perspective, DM Us are classified into “efficient” and “inefficient”. As shown in Figure 1-a, in the flock of birds, the birds with the better performance are depicted at the empirical frontier. In other words, those birds (DM Us) have the better ratio of Outp uts/Inputs (O/I), so they attain the best performance compared to the other birds in the flock. In a construction project, inputs can refer to labor, resources, material, schedule and etc., and outputs variables are those variables measure the performance such as time or cost. DEA also provides “excellent” values of the variables for the DM U to be able to move from inefficient to efficient (Charnes et al. 1978, Banker et al. 1984, Azadeh and Ghaderi 2007, El-M ashaleh et al. 2010). When the flock of birds (DM Us) is organiz ed bas ed on SN, we investigate how many connections each bird (DM U) has and how densely the birds (DM Us) are interconnected. Each connection is established based on different reasons. In a construction project, relationships among specialty trades can be established bas ed on different kind of reasons such as sharing space, sharing resources and being in the sequence of a process. As shown in Figure 1-b, in our flock of birds example, a bird usually prefers to have relations with those of her neighbors have the better efficiency (better O/I ratio), so that she can emulate their behavior to enhance her performance. We believe that we have similar situation in construction project. The informal DM is done through benchmarking and learning from th e best performers among the specialty trades. A specialty trade has many relations Empirical Frontier
Output
In the past two decades, DEA has been widely used in large amount of areas (For example electric utilities evaluation (Fare et al. 1985), software development (Banker and Kemerer, 1984), highway maintenance efficiency (Cook et al. 1990), steel industry productivity (Ray et al. 1998), health care (Chilingerian and Sherman, 2004)). Although DEA is well deployed in other industries, its use in construction is still in the early stages. DEA has been explored mostly as the optimum direction for performance assessment. EI-M ashaleh (2003) used DEA to evaluate the firm-level performance of construction contractors. Pilateris and M cCabe (2003) developed DEA model for financial benchmarks of the construction industry. M cCabe et al. (2005) combined their improved DEA with contractor prequalification benchmarks for the “practical frontier”. Vitner et al. (2006) compared project efficiency in a multi-project environment based on DEA. El-M ashaleh et al. (2006) made use of DEA to quantify the impact of information technology on contractors’ performance. EI-M ashaleh et al. (2007, 2010) advanced the DEA trade-off analysis based on specialty trades performance metrics and reallocated resources according to overall performance.
Input
a. Flock of birds organized by DEA
3. DMUS’ ORGANIZATION VIA DEA/SN APPROACH We depict the movements of birds in a flock (or specialty trades in a project) using two complement approaches, DEA and SN, to illustrate how the informal DM can occurred in construction projects. At the operational level, the DEA evaluates the efficiency of Decision M aking Units (DMUs), while the SN analysis focuses on relationships among the DM Us (Figure 1).
b. Flock of birds organized by SN
Figure 1: DEA-S N approaches for flock of birds
with the other specialty trades of project based on different reasons. Thus, through these relations, she can improve herself (at least to some extent) by learning from the best specialty trades (benchmarks). This (informal DM ) is like a quick and effective improving shortcut, which is more reachable than following the clear objectives of becoming efficient.
Bertelsen et al., 2007). They are: weather condition, material delivery, equipment availability, labor capacity, design and working method clarification, prerequisite work readiness, and space sharing adequacy. We tracked the levels of precondition readiness before execution of each task and used those (in percentage) as indicators of the coordination and planning quality. Outputs. Unreliable work plan (difference between what a trade plans to do and what actually gets done) are the root causes of productivity loss (Howell et al., 1993; Ballard et al., 2005; Wambeke et al., 2011a; Wambeke et al., 2011b; Wambeke et al., 2012; Wang et al., 2012; Hajifathalian et al., 2012). Therefore, we use the following four types of outputs as indicators of project performance: 1) Percent of tasks starting on time (the number of planned tasks are started at the planned time / the number of total planned tasks), 2) Percent of tasks finishing on time within planned duration (the number of planned tasks are finished within the planned duration / the number of total planned tasks.), 3) Cost performance ratio (total planned cost / total actual cost), and 4) Schedule performance ratio (total planned work hours / total actual work hours). Summary of the raw data are shown in the Table 1.
4. RESEARCH METHODOLOGY We propose a procedure to illustrate how the informal DM is done in a construction project. This research uses a case study involving a general contractor to investigate the informal DM during the construction of a fast-track residential construction project ($2.5M , duration: 6 month) from Raleigh area, NC. The project includes Twelve major specialty trades (see Table 1, CO: Concrete, WP: Wood & Plastic, FI: Finishing, PL: Plumbing, M E: M echanical, TM : Thermal & M oisture Protection, DW: Doors & Windows, MA: M asonry, EL: Electrical, M L: M etal, SP: Special Work, and LA: Landscaping). Figure 2 illustrates the main six parts of proposed mechanism which explains in following paragraphs.
4.2. Principal Components of Inputs & Outputs 4.1. Data of DEA Inputs/Outputs The first step in DEA requires specification of the efficiencies for each specialty trade. Efficiency is defined as a weighted sum of outputs divided by a weighted sum of inputs. In its simplest form, we can use the linear summation of inputs and outputs separately (raw inputs vs. raw outputs). The other good way to obtain these weights is to perform Principal Component Analysis (PCA) separately on the inputs and outputs, and use the Eigen vectors associated with the top Eigen values thus accounting for the majority of variation in the data. We use PCA, because it is a variable reduction procedure and useful, since we believe some of our variables are correlated to the one another, possibly because they are measuring the same criteria.
Inputs. In construction planning, 7 types of preconditions need be established for a task to be executed smoothly (Koskela et al. 2002,
Data envelopment benchmarks 7
5
Adjacency matrix of space/time/sequence sharing
6
Social network
4.3. Data Envelopment Benchmarks
Informal decision making
We will us e the principal component of Inputs and principal component of Outputs to perform DEA (using D EAfrontier© Software). Besides finding the efficiencies for our specialty trades, the analysis also generates the Empirical frontier representing a
Figure 2: Research methodology flowcharts
Table 1: S ummary of the Raw Data Data envelopment analysis (DEA) Input variables Output variables
Labor
Material
Pre. work
Space
Design Info
Percent of tasks starting on time
Percent of tasks finishing on time
Cost performance ratio
Schedule performance ratio
CO WP FI PL ME TM DW MA EL ML SP LA
Equipment
1 2 3 4 5 6 7 8 9 10 11 12
Weather
Specialty trades
Social network (SN) Various reasons
0.61 0.81 0.87 0.93 0.83 0.86 0.71 0.86 0.76 0.89 0.88 0.74
0.75 0.75 0.83 0.81 0.77 0.78 0.79 0.82 0.83 0.83 0.79 0.65
0.80 0.78 0.79 0.82 0.80 0.77 0.71 0.85 0.80 0.81 0.64 0.80
0.81 0.84 0.77 0.80 0.73 0.77 0.82 0.91 0.84 0.75 0.58 0.93
0.70 0.76 0.66 0.77 0.76 0.67 0.72 0.78 0.77 0.65 0.74 0.82
0.85 0.81 0.81 0.85 0.86 0.81 0.85 0.84 0.85 0.91 0.78 0.66
0.88 0.83 0.87 0.83 0.91 0.74 0.85 0.87 0.86 0.89 0.95 0.69
0.81 0.66 0.72 0.68 0.86 0.70 0.59 0.69 0.82 0.74 0.68 0.77
0.75 0.68 0.80 0.64 0.79 0.72 0.68 0.74 0.80 0.76 0.72 0.80
0.78 0.92 0.77 0.86 0.82 0.71 0.79 0.99 0.96 0.78 0.73 0.75
0.83 1.05 0.80 0.91 1.12 0.76 0.74 0.97 1.07 0.79 0.73 0.79
Logical sequence
Principal components of outputs (O)
Data of social network inputs
Sharing time
3
Principal components of inputs (I)
4
Adjacency Matrix for SN (A SN) (See Table 2)
2
Data of DEA inputs/outputs
Sharing area
1
4.4. Data of Social Network Inputs In this research, the specialty trades are defined as the nodes of SN and the ties between nodes can be derived from three kinds of relationships among them: sharing space, time and sequence. We define a tie (relationship) between two nodes (specialty trades), if they have any relations based on one of aforementioned reasons. The weakness and strength of the tie are also determined regarding to the type and frequency of the relation.
4.5. Adjacency Matrix of Space/Time/Sequence Sharing In this part, we establish Adjacency M atrices based on SN inputs (A SN) representing the level of interdependent intensity. The ASN is generated from the data, where the element “aij” represents the number of times (frequency), a particular specialty trade “i" communicates with specialty trade “j” based on any of aforementioned reasons (sharing space, time and sequence). Table 2 shows the ASN of the specialty trades of the case study.
LA are the benchmarks among the 12 specialty trades in the project. On the other approach, we obtained the SN of specialty trades based on the relationships existed between each of two specialty trades (Figure 3-b). Figure 3-b not only shows the relationships Table 3: Efficiency of S pecialty Trades Efficiency Benchmarks and % of inputs need Efficiency score to copy to be efficient 1.00 Efficient 1.00 Efficient 1.00 Efficient 1.00 Efficient 1.00 Efficient 1.01 Inefficient 4% (WP) + 4% (M E) 1.03 Inefficient 89% (WP) 1.04 Inefficient 80% (WP)+16% (EL)+9% (LA) 1.12 Inefficient 100% (WP) 1.27 Inefficient 45% (WP)+31% (M E)+28 (EL) 1.29 Inefficient 14% (CO)+47% (WP)+46% (EL) 1.31 Inefficient 86% (WP)+16% (M E)
Trades CO WP ME EL LA SP DW MA PL FI ML TM
Output
benchmark of performance that the specialty trades not on the frontier should try to achieve. It also provides the direction of improvement for inefficient specialty trades to become efficient.
LA WP ME
4.6. Social Network Based on the ASN constructed in previous section, we establish SN of specialty trades using PajekTM. Completely explanation about the details of constructing SN via the software is out of scope of the paper. The SN of specialty trades is shown in the Figure 3-b.
MA
EL
CO
4.7. Informal Decision Making
SP DW
ML PL
The final part of the research is to explain how the informal DM is done based on the DEA and SN results of the data of the case study.
Efficient
FI
5. APPLICATION AND RESULTS
Not Efficient
TM
Using the methodology presented in Section 4 yields the following results: 1) A ranking of specialty trades based on the efficiency score, 2) recognition of the benchmarks amongst specialty trades 3) identification of the relevant benchmarks for each inefficient specialty trades based on his SN. By running DEA, an efficiency score for each of specialty trades plus the direction of improvement for inefficient specialty trades were obtained. Those plus ranking of specialty trades are shown in Table 3. DEA empirical frontier was also established. As can be seen in the Table 3 and Figure 3-a, CO, WP, M E, EL and Table 2: AS N of the S pecialty Trades Trade CO WP FI PL M E TM DW M A EL M L CO 0 3 0 1 0 1 0 3 0 1 WP 3 0 0 3 0 0 1 2 0 0 FI 0 0 0 0 1 0 1 0 4 0 PL 1 3 0 0 0 1 0 3 0 0 ME 0 0 1 0 0 1 2 0 1 0 TM 1 0 0 1 1 0 0 2 0 0 DW 0 1 1 0 2 0 0 0 1 0 MA 3 2 0 3 0 2 0 0 0 1 EL 0 0 4 0 1 0 1 0 0 0 ML 1 0 0 0 0 0 0 1 0 0 SP 0 0 2 0 0 0 0 0 1 0 LA 0 0 1 0 0 0 2 0 0 0
Input
a. Specialty trades organized by DEA LA WP ME
MA
EL
CO
SP LA 0 0 0 0 2 1 0 0 0 0 0 0 0 2 0 0 1 0 0 0 0 1 1 0
SP DW
ML PL
FI Tight relations Regular relations
TM
b. Specialty trades organized by SN
Figure 3 : DEA-SN approaches for specialty trades
existed amongst the specialty trades, it tells us that how tight the relations between two trades are. The thickness of the lines represents the tightness of the relations. For instance although PL communicates with four other specialty trades during the project, he has more communications with WP and M A. Informal DM for a specialty trade, like finding direction by a bird in a flock, is done through benchmarking and learning from the best performers in his own SN. First, he needs to know who are the benchmarks in the projects and then based on his own SN, he focuses on his relationships with better performer. Simply speaking, all he needs is to learn from the efficient specialty trade(s) in his SN and improve himself to reach to the frontier. The trades should be copied by an inefficient specialty trades were determined and can be seen in Table 3. For example, we explain the process for one of specialty trades, M L. M L is inefficient based on the DEA analysis (see Figure 3-a). According to Table 3, we find that he needs to learn from CO, WP and EL to improve his performance and become efficient. On the other hand, from the SN relations of M L (see Figure 3-b). We can find that M L has only weak relationships with CO and MA among all 11 possible relations. In other words, M L only communicates effectively with one of her three benchmarks. Therefore, What M L is done as an informal DM is to focus on his relations with the CO (efficient trade) and try to copy what they do instead of investigating the true goals and objectives. Note that changing in the performance of the specialty trades may lead to changing in the results of DEA, thus after each step of improvement for PL, he may need to find the new improvement direction.
6. CONCLUSION In this paper, inspired by movement of flock of birds, we tried to illustrate how the informal DM is done in construction projects and show how it can provide easy and reliable steps for improving performance. Using a case study ($2.5M fast -track residential construction project), two complement approaches were explained to show the procedure: D EA approach to find the benchmarks and directions of improvement and SN approach to investigate the relationships existed among the specialty trades of the project. Four specialty trades out of 12 trades of the projects were determined as the benchmarks based on running the DEA on their Inputs and Outputs. Other trades were categorized to the three groups according to their efficiency score. Then, through the underlying SN of the project, it was shown that how an inefficient trade can learn from the better performer and improve herself by copying his behavior. This research is significant in three primary regards. First, the research is unique as it is the first use of SN and DEA together to improve trades’ performance. Secondly, it looks at construction relationships from an exclusive perspective. And third, t he analytical process is presented step-by-step, thus, it is repeatable for others to adopt and apply to their own research case within the construction industry.
References Azadeh, A., and Ghaderi, S. F. (2007). “An integrated PCA DEA framework for assessment and ranking of manufacturing systems bases on equipment performance.” Computer-Aided Eng. Software, 24(4), 347-372. Banker, R. D. and Kemerer, C. F. (1984). “Scale economies in new software development.” IEEE Transactions on Software Eng., 15(10), 1199-1205. Banker, R. D., Charnes, A. and Cooper, W. W. (1984). “Some models for estimating technical and scale inefficiencies in data envelopment analysis.” Management Science, 30(9), 1078-1092. Bertelsen, S., Henrich, G., Koskela, L., and Rooke, J. (2007). “Construction physics.” Proceedings IGLC-15, July 2007, Michigan, U.S.
Charnes, A. A., Cooper, W. W., and Rhodes, E. (1978). “Measuring the efficiency of decision making units.” Eur. J. Oper. Res., 2(6), 429-444. Chilingerian, J. A. and Sherman, H. D. (2004). Health care applications: from hospitals to physicians, from productive efficiency to quality frontiers, Handbook on Data Envelopment Analysis, Kluwer Academic Publishers, Boston. Chinowsky, P., Diekmann, J. and Galotti, V. (2008). “Social network model of construction.” J. Constr. Eng. Manage., 134(10), 804-812. Cook, W. D., Roll, Y., and Kazakov, A. (1990). “A DEA model for measuring the relative efficiency of highway maintenance patrols.” INFOR, 28(2), 113-124. De Nooy, W., Mrvar, A. and Batagelj, V. (2005). Exploratory network analysis with Pajek®, Cambridge University Press, New York, U. S.. EI-Mashaleh, M. S., Rababeh, S. M. and Hyari, K. H. (2010). “Utilizing data envelopment analysis to benchmark safety performance of construction contractors.” J. Project Manage., 28(1), 61-67. El-Mashaleh, M. S. (2003). “Firm performance and information technology utilization in the construction industry: an empirical study.” Ph.D. Diss., University of Florida at Gainesville. El-Mashaleh, M. S., O’Brien, W. and Minchin, R. (2006). “Firm performance and information technology utilization in the construction industry.” J. Constr. Eng. Manage., 132(5), 499-507. El-Meshaleh, M. S., Minchin, R. E. Jr. and O’Brien, W. J. (2007). “Management of construction firm performance using benchmarking.” J. Manage. Eng., 23(1), 10-17. Fare, R., Grosskopf, J. L. and Lovell, C. A. K. (1985). The measurement of efficiency of production, Kluwer Academic Publishers, Boston/Dordrecht/London, UK. Howell, G., Laufer, A. and Ballard, G. (1993). “Uncertainty and Project Objectives.” Project Appraisal Journal, 8(1), 37-43. Hajifathalian, K., Wambeke, B., Liu, M., and Hsiang, S. (2012). “Effects of Working Strategy and Duration Variance on Productivity and Work in Progress.” Accepted by ASCE J. Constr. Eng. Manage. in 2011, in production. Kabnurkar, A. (2001). “Mathematical modeling for data envelopment analysis with fuzzy restrictions on weights,” thesis, presented to Virginia Polytechnic Institute and State University at Falls Church, Virginia, in partial fulfillment of the requirement for the degree of Master of Science. Koskela, L., Howell, G., Ballard, G. and Tommelein, I. (2002). The foundations of lean construction (Chapter-14), In Design and Construction: Building in Value, Oxford, UK. McCabe, B., Tran, V. and Ramani, J. (2005). “Construction prequalification using data envelopment analysis.” Can. J. Civ. Eng., 32(1), 183-193. Moreno, J. L. (1960). The sociometry reader, The Free Press, Glencoe, UK. Park, H., Han, S. H., Rojas, E. M., Son, J. and Jung, W. (2011). “ Social Network Analysis of Collaborative Ventures for Overseas Construction Project.” J. Constr. Eng. Manage., 137(5), 344-355. Pilateris, P. and McCabe, B. (2003). “ Contractor financial evaluation model.” Can. J. Civ. Eng., 30(3), 487-499. Ray, S., Seiford, L. M. and Zhu, J. (1998). “Market entity behavior of Chinese state-owned enterprises.” OMEGA, 26(2), 263-278. Taylor, J.E., Levitt, R. and Villarroel, J.A. (2009). “Simulating learning dynamics in project networks.” J. Constr. Eng. Manage., 135(10), 1009-1015. Thompson, W. A., Vertinsky, I. and Krebs, J. (1974), “The Survival Value of Flocking in Birds: A Simulation Model.” Journal of Animal Ecology, 43(3), 785-820. Vitner, G., Rozenes, S. and Spraggett, S. (2006). “Using data envelope analysis to compare project efficiency in a multi-project environment.” Intl. J. Project Manage., 24(4), 323-329. Wambeke, B., Hsiang, S. and Liu, M. (2011a). “Causes of variation in construction project task starting times and duration”, ASCE J. Constr. Eng. Manage., 137(9), 663-677. Wambeke, B., Liu, M. and Hsiang, S. (2011b). “Trade variation and the Social Network of Construction Trades.” Submitted to ASCE J. Constr. Eng. Manage. in April 2011, under revision. Wambeke, B., Liu, M. and Hsiang, S. (2012). “Using Pajek and centrality analysis to identify a social network of construction trades.” accepted by ASCE J. Constr. Eng. Manage. in 2011, in production. Wang, C., Liu, M., Hsiang, S. and Leming, M. (2012). “ Causes and penalties of variation-a case study of a precast concrete slab production facility.” accepted by ASCE J. Constr. Eng. Manage. in 2011, in production. Wong, K., Unsal, H., Taylor, J. E. and Levitt, R. (2010). “Global dimension of robust project network design.” J. Constr. Eng. Manage., 136(4), 442-451. Zhu, J. (2009). Quantitative Models for Performance Evaluation and Benchmarking, Second edition, Springer, New York, U. S..
