Printer Friendly

Application of importance-satisfaction analysis and influence-relations map to evaluate design delay factors.

Introduction

The design of a facility construction project involves conceptual design, schematic design and detailed design. During the conceptual and schematic design phases, a prime designer (architect/engineer or A/E) seeks to consider information from a wide range of disciplines; represent candidate solutions, and generate new states of design from the current states to fulfil the requirements of the project client (or owner), including for example, the budget and spatial arrangements (Baldwin et al. 1999; Rivard, Fenves 2000). These two early phases ensure that the design deliverables satisfy the general needs of the project client. In the detailed design phase, the required design work is explicated; the design deliverables must be delivered in a timely fashion to prevent delays to future construction work.

When a design delay occurs, the completion time of the entire project may be postponed. Identifying design delay factors for improvement is critical to managing design duration (Yang, Wei 2010). However, many interrelated factors affect the design duration. These include, for example, the clarity of user needs, the timing of decisions made by the client, the design capabilities of A/E, and the management abilities of project management. When user needs are not clearly specified or client's decisions change frequently, the design deliverables are likely to be determined by trial and error. As time passes, the A/E begins to lose patience and spends less time on design. Consequently, the project client is unsatisfied with the design deliverables. Each factor influences each other factor. Finally, project participants (including facility users, decision makers, project management, and A/E) complain to each other, exacerbating delays. If key factors that drive design delays can be identified, and attention paid to them, then such delays can be eliminated or, at least, prevented from being increased. Nevertheless, identifying the cause-effect relationships among delay factors is difficult when a project involves numerous project participants and includes specialty knowledge (such as concerning a clean room or special mechanical/electrical/plumbing layout requirements).

This study proposes a new model to help identify the key factors that govern design delays. Correcting the key factors that most strongly affect delays is effective in preventing design delays. A real-world high-tech facility construction project in northern Taiwan is considered to demonstrate the advantages of the proposed model.

The rest of this paper is organized as follows. Section 1 reviews recent related studies. Section 2 presents details of the proposed model. Section 3 illustrates a case study of the proposed model. The final section draws conclusions and provides recommendations for future research.

1. Literature review

This section reviews past studies related to duration management design and delay factor identification for construction projects.

1.1. Management of design duration

The importance of efficient design management in ensuring the smooth running of a project is being increasingly appreciated (Chua et al. 2003; Senthilkumar et al. 2010). Much research has been undertaken to control design processes, and thereby increase the effectiveness of the management of design duration. For example, Sanvido and Norton (1994) proposed a building design process model that indicates the tasks on which a successful design depends. Bogus et al. (2005) developed a concurrent engineering approach that involved overlapping sequential design activities to reduce design duration.

Several researchers have directly dealt with design schedule management (Chua, Hossain 2011). For example, considering the uncertainty in the number of iterations of design activities, Luh et al. (1999) developed an optimization-based method to schedule the design process in a manufacturing project. Considering design iterations and information dependency, Austin et al. (2000) presented a model to schedule a building design project. Taking the uncertainty in the number of iterations and the number of design participants into account, and considering the possibility of conducting multiple design projects simultaneously, Wang et al. (2006) devised a simulation-based model to produce design schedules for a design firm.

1.2. Identification of delay factors

An increasing number of construction projects have experienced extensive delays (Odeh, Bataineh 2002). Causes of delays (i.e. delay factors) arise in all project phases. Research on delay factor identification has received considerable attention in the past several decades because timely identification is the basis for resolving delays (Baldwin, Manthei 1971; Yang, Wei 2010). Practitioners should foresee potential delay factors likely to confront their current and future projects and avoid or reduce these delays in a timely manner (Long et al. 2004; Sweis et al. 2008).

Past research studies may be broadly divided into two types: identifying delay factors in general and evaluating specific delay factors in detail:

1) In the first type of research, for instance, a survey conducted by Sweis et al. (2008) found that the financial difficulties faced by the contractor and an excess of change orders made by the project client are the two factors that most responsible for construction delays. Delay factors arising in international countries have been explored, such as Vietnam (Long et al. 2004), Nigeria (Aibinu, Odeyinka 2006), Malaysia (Sambasivan, Soon 2007), Jordan (Odeh, Bataineh 2002; Sweis et al. 2008), Saudi Arabia (Assaf, Al-Hejji 2006; Al-Kharashi, Skitmore 2009), Thailand (Ogunlana et al. 1996; Toor, Ogunlana 2008), Egypt (El-Razek et al. 2008), Zambia (Kaliba et al. 2009), Gaza Strip (Enshassi et al. 2009), Ghana (Fugar, Agyakwah-Baah 2010), Taiwan (Yang, Wei 2010), and India (Doloi et al. 2011).

2) In the second type of research, researchers assessed specific delay factors to find solutions to mitigate the delay impact. These specific delay factors include labor productivity (Alinaitwe et al. 2007), weather-related factor (Apipattanavis et al. 2010), and order variations (Enshassi et al. 2010). For example, Apipattanavis et al. (2010) identified the weather attributes (such as temperature and rain) that cause construction delays and developed a weather prediction model for mitigating disputes arising from weather-related delays.

Ramanathan et al. (2012) recently reviewed 113 construction delay factors from 41 studies around the world. Their analyses concluded several findings, including:

1) All 41 studies conducted semi- or full-structured expert interviews and performed questionnaire surveys to analyse the data obtained from the responses. Various indices such as importance index, frequency index and severity indices were then used to evaluate the factors.

2) Delay factors are specific to various countries (developed versus developing countries), project sizes (project contract price amount) and project types (such as housing, industrial and commercial projects) and that no root causes can be generalized. This finding is agreed by some researchers (Sweis et al. 2008).

3) None of these studies can be generalized and directly applicable "as is". This critical review presented a strong case against opinion surveys and suggested conducting statistical analyses of actual projects to generate meaningful results.

Existing studies are too numerous to describe in detail. Summaries are available in Toor and Ogunlana (2008) and Ramanathan et al. (2012). Table 1 compares some of recent studies and this work for indicating the significances of this study, including:

1) This study focuses on the design phase, while most past studies focused on the construction phase. Only very few studies discussed the delay factors occurring in the design phase (Wei 2005; Yang, Wei 2010; Yang et al. 2010). For instance, based on 95 valid responses from consultant engineers, Yang and Wei (2010) found that "changes in the client's requirements" are the main causes of both planning and design phase delays. Focusing on public construction projects under the build-operate-transfer contract model, Yang et al. (2010) indicated several design delay causes, such as change in client's needs, wrong client's needs recognized by the design team, inexperienced design team and improper plans and schedule developed by the design team. These design delay factors are applied for general projects.

2) Past studies identified delay factors for general projects. This study focuses on developing a model to identify factors and evaluate their cause-effect relationships. The identified delay factors are for a particular project (and are not intended to apply to general projects), although they may serve as references for other projects.

3) The proposed model here assesses the cause-effect relationships among factors to identify the key driving factors. Only a few past studies considered the correlations among factors and used different correlation analyses. For instance, several researchers used the Spearman rank correlation coefficient to analyse the agreement/disagreement between project parties or survey respondents (El-Razek et al. 2008; Al-Kharashi, Skitmore 2009). Their studies indicated that the clients and contractors often have opposing views. Doloi et al. (2011) used factor analysis to assess the correlations among delay attributes and grouped the highly correlated attributes into factors (such as lack of commitment and inefficient site management). They also developed a regression model to indicate that slow decisions from the owner and rework due to mistakes in construction had the greatest impact on construction delays in Indian construction projects.

In summary, this study has three major significances: it focuses on the design phase; it develops an evaluation model for a specific project and it evaluates causal relationships among factors for finding driving delay factors.

2. Proposed model

Delay factors may vary greatly with the characteristics of the project (including the participants of the project, management style, project complexity, and project type). Accordingly, a model that can help management find delay factors in a step by step manner is preferred. Figure 1 presents the steps in the proposed model for identifying the factors and sub-factors that govern design delays in a project. The steps in the model are as follows, and steps 1-3 are elucidated in greater detail in subsequent subsections.

Step 1: Defining the factors and sub-factors that influence the time taken to design a facility construction project. In this step, available studies can assist in generating a long list of possible factors. Expert interviews are conducted to assess the relevance of these factors to the project in question.

Step 2: Using the "importance-satisfaction analysis (ISA)" to evaluate the degree of importance and the degree of satisfaction of each identified factor. A factor that results in a delay is considered to be unsatisfactory.

Step 3: Applying the "decision making trial and evaluation laboratory technique (DEMATEL)" to construct a cause-effect influence-relations map or impact-relations map (IRM) among factors.

Step 4: Integrating the ISA and IRM. The ISA specifies factors that are important and highly unsatisfactory to design delays. Meanwhile, the IRM traces the key factors that govern design delays.

Steps 2-4 are repeated to find the problematic subfactors of the key factors. Management can attend to them to improve the design time.

2.1. Step 1: defining the factors and sub-factors

In defining the design delay factors and sub-factors, the following sub-steps are suggested:

1) Reviewing past studies (although most concern construction delays) to identify potential delay factors and sub-factors. Prior research showed that grouping delay factors helped to understand the factors systematically; but, no consensus existed on how factors should be grouped. For example, Odeh and Battaineh (2002) used seven groupings, such as client-, contractor-, consultant-, materials-, labour-, contract-, and contractual-related groups. Ramanathan et al. (2012) divided factors into 18 categories.

2) Conducting expert interviews to indentify suitable delay factors from the aforementioned potential delay factors.

2.2. Step 2: conducting ISA

The ISA method was based on the importance-performance analysis that was proposed by Martilla and James (1977), but "satisfaction" replaced "performance" (Tonge, Moore 2007). In the ISA, the input data (degree of satisfaction and degree of importance of each factor and sub-factor) collected from the questionnaires are normalized to a single measuring scale. Eqns (1) and (2) yield the initial degree of satisfaction (IDS) and standardized satisfaction value (SS):

IDS = SumDS/N; (1)

SS = (IDS - Average of IDS in all factors)/Standard deviation of IDS in all factors, (2)

where: N is the number of respondents; and SumDS is the sum of degrees of satisfaction from all respondents.

Eqns (3) and (4) calculate the initial degree of importance (IDI) and the standardized importance value (SI). The number of respondents in the case study was 36.

IDI = SumDI/N; (3)

SI = (IDI - Average of IDI in all factors)/Standard deviation of IDI in all factors, (4)

where SumDI is the sum of degrees of importance from all respondents.

The ISA evaluations are classified under the following four categories (SS, SI): (1) *(+,+): a factor with high satisfaction and high importance; (2) **(+,-): a factor with high satisfaction and low importance; (3) ***(-,-): a factor with low satisfaction and low importance, and (4) x(-,+): a factor with low satisfaction and high importance. The fourth category, x(-,+), should attract the most attention.

2.3. Step 3: performing DEMATEL-based IRM analysis

In the proposed model, the end product of the DEMATEL process is an IRM that is a visual representation of the interdependences of factors. The Battelle Memorial Institute of Geneva developed the DEMATEL method for a Science and Human Affairs Program to solve complex and interrelated problems (Gabus, Fontela

1973; Li 2009; Lin, Tzeng 2009). The DEMATEL method enables management to solve problems visually and divide the related factors (or variables) into cause and effect groups to improve understanding of causal relationships among factors (Li 2009; Wu, Tsai 2011; Hu et al. 2011; Wang et al. 2012). This method has been employed in numerous fields, including the development of global managers' competencies (Wu, Lee 2007), the assessment of value-creating industrial clusters in a science park (Lin, Tzeng 2009), and the capture of the causal relationships between strategic criteria in an organization (Jassbi et al. 2011).

The steps in the DEMATEL method are as follows

(Lin, Tzeng 2009; Wang et al. 2012): (1) Step D1--finding the average matrix; (2) Step D2--calculating the direct influence matrix; (3) Step D3--calculating the indirect influence matrix; (4) Step D4--deriving the total influence matrix; and (5) Step D5--obtaining the influence-relations map.

Step D1: Finding the average matrix

Suppose h experts are available to solve a complex problem and n factors are considered. The scores assigned by each expert yield an n x n non-negative answer matrix [X.sup.k], with 1 [less than or equal to] k [less than or equal to] h. Hence, [X.sup.1], [X.sup.2],..., [X.sup.h] are the answer matrices for each of the h experts, and each element of [X.sup.k] is an integer, denoted [X.sup.k.sub.ij]. The diagonal elements of each answer matrix X are all set to zero. The n x n average matrix A can then be computed by averaging the h experts' value (or score) matrices. The (i, j) element of the average matrix A is denoted [a.sub.ij] (average influence):

[a.sub.ij] = 1/h [h.summation over of (k=1)] [x.sup.k.sub.ij]. (5)

Step D2: Calculating the direct influence matrix

A direct influence matrix D is obtained by normalizing the average matrix A. That is:

D = s A, (6)

where s is a constant, which is calculated as follows (Li 2009):

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (7)

Notably, suppose the (i, j) element of matrix D (denoted [d.sub.ij]) is the direct influence of factor i on factor j. Then, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (Goodman 1988). Additionally, 0 < [n.summation over of (j=1)] [d.sub.ij], [n.summation over of (i=1)] [d.sub.ij] [less than or equal to] 1, and the sum of only one row or column equals one.

Step D3: Calculating the indirect influence matrix

The indirect influence of factor i on factor j declines as the power of the matrix increases, as in [D.sup.2], [D.sup.3],..., [d.sup.[infinity]] (Lin, Tzeng 2009). This fact guarantees convergent solutions to the matrix inversion, similar to an absorbing Markov chain matrix. The indirect influence matrix ID is obtained from the values in the direct influence matrix d. That is:

ID = [D.sup.2] + [D.sup.3] +... = [[infinity].sumatiob over of (i=2)] [D.sup.i] = [D.sup.2] [(I - D).sup.-1], (8)

where I is the identity matrix.

Step D4: Deriving the total influence matrix

The total influence matrix t is also an n x n matrix, and is given by (Li 2009):

T = D + ID = D + [D.sup.2] + [D.sup.3] +... = [[infinity].summation over of (i=1)][D.sup.i] = D[(I - D).sup.-1]. (9)

Let [t.sub.ij] be the (i, j) element of matrix t; the sum of the i-th rows and the sum of the j-th columns, [d.sub.i] and [r.sub.j], respectively, are obtained as follows:

[d.sub.i] = [n.summation over of (i=1)] [t.sub.ij] (i = 1, 2, 3..., n); (10)

[r.sub.j] = [n.summation over of (j=1)] [t.sub.ij] (j = 1, 2, 3..., n). (11)

Notably, [d.sub.i] represents the sum of the direct and indirect influences of factor i on the other factors, and [r.sub.j] denotes the sum of direct and indirect influences on factor j by the other factors. When j = i, [d.sub.i] + [r.sub.i] is an index of the strength of influences by and on a factor, and is a measure of the importance of that factor. The term [d.sub.i] - [r.sub.i] (also called "Relation") divides factors into the "cause group" and the "effect group". If dt - rt is positive, then factor i influences other factors more than it is influenced by, and so belongs to the cause group. Conversely, if [d.sub.i] - [r.sub.i] is negative, then factor i is influenced by more factors than it influences and it belongs to the effect group (Li 2009).

Step D5: Obtaining the influence-relations map

To visualize the complex causal relationships among factors using a visible structural model, an IRM is developed from the values of d + r and d - r, represented on the x axis and the y axis, respectively, of a graph (Lin, Tzeng 2009). Furthermore, the net influence matrix N is provided to evaluate the strength of the effect of one factor on another:

N = [Net.sub.ij] = [t.sub.ij] - [t.sub.ji]. (12)

3. Case study

This section presents the results of applying the proposed model to a case project. The project background is presented; factors and sub-factors are defined; input data are collected, the ISA and IRM methods are evaluated, and the results are discussed.

3.1. Project background

The case project is the construction of a high-tech facility of a national research centre located in northern Taiwan. The total floor area of the facility is approximately 53,000 [m.sup.2]. The case project comprises three main components: (1) civil and building construction (Civil); (2) mechanical, electrical and plumbing (MEP) works; and (3) special equipment construction (SPE). The Civil and MEP design is contracted out to an A/E. However, the SPE is designed by an internal team of the project client, because the SPE depends on particular domain knowledge, such as synchrotron accelerators and resistance against micro-vibrations. The planned design and construction durations of the Civil and MEP components of the facility are about one year and 2.5 years, respectively. The consultant fee for the A/E is around US $3.3 million, while the budget for construction (Civil and MEP) is approximately US $77.4 million.

The project client establishes a task force to manage the design of the Civil and MEP components, which represent the scope of this case study. The researchers of this study have worked closely with this task force. Four major groups of participants are involved in the design phase of this project (Fig. 2). They are: (1) facility users, who determine the requirements of the facility; (2) decision makers, who are the top management of the project client; (3) the project management team, which is the aforementioned task force; and (4) the A/E. The facility users, decision makers and project managers are all the colleagues of the project client, and they interact with the A/E both directly and indirectly. The facility users, from 11 departments of the research centre, have various specialties and needs.

After approximately 1.5 years, the design had not been completed, and the design was thus significantly delayed. The project client claimed that poor execution by the A/E as responsible for the delays, while the A/E claimed compensation for reworking and costs incurred by the late decisions and frequent change orders of the project client. Since several external factors (associated with governmental agencies) and internal factors affect the progress of the project, no one could convincingly present the causes of the delays. The relationship among the parties in the project was therefore very tense. In the midst of this situation, the research team began to implement the proposed model to help identify the design delay factors to resolve the dispute and hopefully prevent further delays. The research team members included practitioners with expertise in construction project management and researchers familiar with performing ISA and IRM.

3.2. Definitions of design delay factors and sub-factors

The research team first defined the factors and sub-factors that affect the design duration of the case project. Some studies were reviewed and factors were grouped based on project participants, because the team was asked to identify "who" caused the most delays. This study then identified four first-level delay factors, corresponding to the four groups of project participants (facility users, decision makers, project management, and A/E). Additionally, after referring to the work of Wei (2005) that focused on the design phase, a long list of possible factors was reserved for this case project. These factors included changes in the client's requirements, poor scope definition, client change orders, the client's complicated administration process, the client's financial problems and inadequate A/E experience.

Three project engineers and one project manager for the case project were then invited to join with the research team to further identify delay factors suitable for this specific project. During the factor indication process, each of the aforementioned possible factors was included, revised or deleted.

Table 2 summarises the identified 17 second-level sub-factors under the four first-level factors. These factors and sub-factors are described as follows: 1) User needs and specification requirements (US)--In this case project, the facility users needed high technical requirements on the resistance standards against micro-vibrations and several vertical/horizontal precisions of construction. Because this facility (i.e. accelerator) was only the second kind in Taiwan, none of the project participants (including the users) had sufficient experience in ascertaining a need (i.e. sub-factor US1) or clarifying a need (US2). Sometimes, governmental regulations (US4, such as building codes) and technical specifications (US5) prevented the full description of user needs. Moreover, many user needs were difficult to meet in design or in construction (US2). Each factor could cause a delay. For example, if the user needs were unclear (US2), then A/E would have problems in comprehensively translating these needs into design deliverables, leading to a design delay.

2) Organization's decision making and budget constraints (OB)--This OB factor was mainly involved the project client's decision-making and management abilities. Two top managers fully supervised the whole project and made most decisions, such as user needs, budget allocation, building architecture and layout. Consequently, whether or not their decisions were made timely (OB1), their supervising methods were appropriate (OB2) and their budget allocation was efficient (OB4) could impact design duration. Additionally, the project client was responsible for the effects of external factors, such as the project budget approved by governmental agencies (OB3). If part of budget was not available (OB3) or budget was allocated too much to certain components (such as the accelerator MEP systems, OB4), user needs might be altered and budget would be insufficient for other parts of the facility, leading to a redesign.

3) Project control and review management (PM)--This factor related to the effectiveness of project management, in terms of both communication and experience. The project client formed a project management team. Although most PM members had expertise in construction projects (PM2), they basically managed this high-tech facility construction project in a conventional way. For instance, no advanced management tools for scheduling were used (PM1) and no standard design review operating procedures were strictly implemented (PM4). Furthermore, the PM's communication with other participants was not effective (PM3). Overall, it was considered that the PM managed the project reactively.

4) Design execution and interface management (DM)--Although this A/E had plenty of experience in designing high-tech facilities, he tended to be overconfident and neglected the demands of this project. This factor concerned the A/E's abilities in drawing and design (DM1), estimating costs (DM2), and allocating design man-hours (DM4). Moreover, this project's MEP design involved numerous complicated systems developed by A/E subcontractors. A/E's ability to manage his subcontractors became crucial to timely design deliverables (DM3). Most of the design delay factors and sub-factors identified here are similar but more specific than those suggested by past studies. For example, prior research often categorized these factors into two broader groups (i.e. client and consultant), while this study further divided the client group into two groups (i.e. facility users and decision makers) and divided the consultant group into two groups (i.e. project management and A/E). A few factors here were unique because of the characteristics of the case project. For instance, sub-factors US3, US4 and US5 (related to facility users) occurred due to the high technical micro-vibration resistance and construction precision requirements for this case project. Sub-factor OB4 (decision makers' budget allocation management) and sub-factor DM3 (A/E's subcontractor management) might not be common to other projects.

3.3. Collection of input data

The required input data for the ISA and (DEMATEL-based) IRM are obtained using a set of questionnaires. Thirty-six experts (engineers, section managers or managers who were involved in this case project) were asked to fill out each questionnaire. Table 3 shows an example of the response from the questionnaire for executing ISA. In Table 3, the degree of importance and the degree of satisfaction range between ten (highest importance or satisfaction) and zero (lowest importance or satisfaction).

Table 4 presents an example of the response from the questionnaire for generating IRM. Each respondent was asked to evaluate the strength of the direct influence (effect) of a factor on each of the other factors using an integer scale (from zero to four). In Table 4, for example, suppose factor i (PM) has a weak direct influence on factor j (US): a score of "1" is given to represent this weak influence. Conversely, if the US factor has a strong direct influence on the PM factor, a score of "3" is assigned. A high score represents the belief that an improvement in the PM factor relies strongly on an improvement in the US factor.

The research team held several meetings to illustrate the details of the questionnaire to the experts in order to ensure that the questionnaires could respond effectively. Cronbach's [alpha] is utilized to test the reliability of data collected from the questionnaires in this case project. The test results reveal that these questionnaires used in the ISA and IRM analyses are reliable (as a exceeds 0.7) (Table 5).

3.4. Evaluation of ISA

Table 6 shows the evaluations made using ISA in the case study. As indicated earlier, the fourth category, x (-,+), should attract the most attention. In this case study, the OB factor (organization's decision making and budget constraints) falls under the fourth category: the OB factor is considered to be highly important, but with a low degree of satisfaction. Therefore, this OB factor should be improved immediately. Figure 3 graphically represents the results of the ISA evaluation. Factors (including the OB factor herein) that are in the second quadrant are easily identified for improvement.

3.5. Generation of IRM

The research team then carried out the steps of the DEMATEL method, as follows.

Step D1: Finding the average matrix--Table 7 presents an initial average matrix (average matrix A) of the factors in the case project. The value of h in Eqn (5) is 36 (36 respondents). For example, the initial average value of the effect of the US factor on the OB factor (US [right arrow] OB) is calculated to be 3.111, indicating a high direct influence (as 4.0 is the highest value).

Step D2: Calculating the direct influence matrix--In Table 7, the sum of the second row is the maximum value (8.973) of [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. Therefore, the value of s equals 1/8.973. Accordingly, based on Eqn (6) and the values of the average matrix A (Table 7), a direct influence matrix D is obtained and presented in Table 8. For instance, the value of direct influence US [right arrow] OB is calculated to be around 0.347 (= 3.111/8.973).

Step D3: Calculating the indirect influence matrix Table 9 shows the calculated indirect influence matrix id of factors in the case project, using the calculation functions in MATLAB (2009).

Step D4: Deriving the total influence matrix--Table 10 presents the total influence matrix obtained using Eqn (9). The table also provides the values of [d.sub.i] and [r.sub.j]. Table 11 displays the values of [d.sub.i] + [r.sub.i] and [d.sub.i] - [r.sub.i] for each factor. The results demonstrate that the values of [d.sub.i] - [r.sub.i] are positive for the US and OB factors, which therefore fall into the cause group. If management wishes to improve the effect group factors to exhibit satisfactory duration performance, they must take great care to control the cause group factors.

Step D5: Obtaining the influence-relations map (IRM) Figure 4 presents the IRM of the first-level factors in the case project. Using the map, management can visualize the difference between cause factors (OB and US) and effect factors (PM and DM). Table 12 shows the net influence matrix of factors using Eqn (12). For example, the net influence of the OB factor on the US factor is 0.137 (= 3.468-3.331; Table 10).

3.6. Integration of ISA and IRM

Table 13 presents the evaluation obtained using ISA and IRM. In the ISA analysis, only the OB factor (with positive importance; SI > 0) has unfavourable design duration performance (negative satisfaction; SS < 0). Therefore, the OB factor must be improved. Moreover, IRM analysis reveals that the US and OB factors are in the cause group, while the other two factors, PM and DM, are in the effect group. Overall, management strategy A (which requires no further improvement) can be applied to US, PM, and DM (SS > 0). Management strategy B (which requires direct improvements) should be applied to OB (SS < 0 and d - r > 0).

Figure 5 represents the evaluation in visually. The left of the figure presents the ISA analysis, whereas the right shows the IRM analysis. In the figure, the OB factor affects all other factors.

3.7. Identification of the key second-level sub-factors

Improving the OB factor depends on identifying the key sub-factors that are responsible for the delays. The input data collected from the aforementioned questionnaires and the same modelling steps as used in applying the ISA and IRM methods yield the following results:

--The performance of all four sub-factors (OB1~OB4) is not satisfactory;

--Sub-factors OB1 (DM's decision making), OB2 (DM's supervising ability), and OB3 (budget availability) are in the cause group, and they must be improved directly (strategy B; SS < 0 and d - r > 0);

--Sub-factor OB4 (DM's resource allocation) is in the effect group; it must be improved indirectly (strategy C; SS < 0 and d - r < 0). That is, sub-factor OB4 is improved by improving other sub-factors (OB1~OB3).

Table 14 shows the suggested strategies for improving the second-level delay sub-factors under the OB factor. Figure 6 represents the evaluation results of the ISA and IRM analyses of these sub-factors.

3.8. Discussion

The research team took around one month to collect the required questionnaires and conduct the analyses. The major evaluations in this case study are as follows:

--The ISA analysis reveals that the OB factor that is related to decision makers is the only unsatisfactory factor;

--The IRM analysis demonstrates that the OB and US factors are in the cause group (meaning that they influence other factors more than they are affected), whereas the PM and DM factors are in the effect group. The decision makers (related to the OB factors) and facility users (related to the US factors) should have higher responsibilities for the delays (if any) than the other two project participants (project management and A/E) in this case project;

--In the cause group, the US factor with the highest value of d + r is the most important factor (meaning that it most strongly influences, and is most strongly influenced by other factors) in preventing design delays, whereas the OB factor, with the highest value of d - r, has the greatest effect on the other factors;

--In the effect group, an improvement in the PM factor (with the lowest value of d - r) depends strongly on the improvements in other factors;

--The decision makers (related to the OB factor) should take more responsibility for the design delays than does the A/E (related to the DM factor);

--Improving the OB factor depends on the improvement of sub-factors OB1, OB2 and OB3.

The above evaluation results were presented to a top manager of the project client and some project management team members. Their main feedback was as follows:

--A visual diagram (Fig. 5) is useful for clearly communicating the cause-effect relationships among factors;

--The modeling results are useful in providing two types of information for managing design delays (1) factors or sub-factors to which delays are most sensitive, and (2) project participants who are most responsible for delays;

--Obtained management information should be provided as soon as possible to enable corrections to be made in a timely fashion. Regularly updating of the evaluation of the model is preferred;

--The modelling results are useful in supporting the design duration management. However, they are not useful as legal evidence in the resolution of disputes;

--The top manager appreciated the evaluation results even though the project' decision-makers were considered to have the most responsibility for the delays. However, he thought that the project client organization should take responsibility and that blame should not lie only with its top managers because many of their late decisions and inefficiencies were

caused by an ineffective organizational structure. He also indicated that some of the delay factors (related to client's needs) were caused by offering an excess of conflicting opinions by facility users and the scientific spirit of the organization, which always seeks perfect solutions to problems. Nevertheless, he would pay attention to improving the two driving delay factors, that is, the DM's decision making (OB1) and DM's supervising ability (OB2).

Conclusions

Factors that affect delays in design projects are complex and interrelated. Controlling key delay factors is an efficient means of managing the duration of design work. This study proposes an innovative model that helps identify the performance and cause-effect relationships among delay factors. The ISA method is adopted to evaluate the performance of each factor and sub-factor. A DEMATEL-based IRM analysis is employed to analyse the cause-effect interrelationships among factors. The ISA and IRM are then combined to trace the key factors that most strongly affect design delays. In the case study, of the four first-level delay factors and the 17 second-level delay sub-factors, the OB factor (organization's decision making and budget constraints) was found to be the key driver. The top manager should pay more attention to this factor to improve design duration.

Notably, specialist knowledge may be required to implement the proposed model successfully. For instance, in the case study, the research team members possessed expertise in construction project management, allowing them to search for relevant studies efficiently and communicate with project participants smoothly to identify appropriate design delay factors. Furthermore, the team members familiarized itself with the steps of performing the ISA and IRM, allowing them to not only clearly present the proposed model to the experts in order to gather the required questionnaires, but also to perform the modelling calculations and interpret the modelling results easily.

For facilitating the implementation of the proposed model, we recommend that future research computerize the proposed model to expedite the evaluation steps. Computerization also supports management to act appropriately and timely. Second, as implied by Ramanathan et al. (2012), analyses of the management solutions of delay factors for additional specific projects are needed. Third, comparing the practical meanings of delay causes in various countries is also valuable. Fourth, ISA and IRM schemes may be applied to solve different decisionmaking problems, which involve various interrelated factors.

Acknowledgements

The authors would like to thank the reviewers for their careful evaluation and thoughtful comments. The authors also thank the National Science Council of Taiwan (Contract No. NSC98-2221-E-009-169) for financially supporting this research. The valuable assistance and collaboration of respondents and experts in the case study are greatly appreciated.

References

Aibinu, A. A.; Odeyinka, H. A. 2006. Construction delays and their causative factors in Nigeria, Journal of Construction Engineering and Management 132(7): 667-677. http://dx.doi.org/10.1061/(ASCE)0733-9364(2006)132: 7(667)

Alinaitwe, H. M.; Mwakali, J. A.; Hansson, B. 2007. Factors affecting the productivity of building craftsmen--studies of Uganda, Journal of Civil Engineering and Management 8(3): 169-176.

Al-Kharashi, A.; Skitmore, M. 2009. Causes of delays in Saudi Arabian public sector construction projects, Construction Management and Economics 27(1): 3-23. http://dx.doi.org/10.1080/01446190802541457

Apipattanavis, S.; Sabol, K.; Molenaar, K. R.; Rajagopalan, B.; Xi, Y. P.; Blackard, B.; Patil, S. 2010. Integrated framework for quantifying and predicting weather-related highway construction delays, Journal of Construction Engineering and Management 136(11): 1160-1168. http://dx.doi.org/10.1061/(ASCE)CO.1943-7862.0000199

Assaf, S. A.; Al-Hejji, S. 2006. Causes of delays in large construction projects, International Journal of Project Management 24(4): 349-357. http://dx.doi.org/10.1016/j.ijproman.2005.11.010

Austin, S.; Baldwin, A.; Li, B.; Waskett, P. 2000. Analytical design planning technique (ADePT): a dependency structure matrix tool to schedule the building design process, Construction Management and Economics 18(2): 173-182. http://dx.doi.org/10.1080/014461900370807

Baldwin, A.; Austin, S.; Hassan, T.; Thorpe, A. 1999. Modeling information flow during the conceptual and schematic stages of building design, Construction Management and Economics 17(2): 155-167. http://dx.doi.org/10.1080/014461999371655

Baldwin, J. R.; Manthei, J. M. 1971. Causes of delay in the construction industry, Journal of Construction Division 97(2): 355-368.

Bogus, S. M.; Molenaar, K. R.; Diekmann, J. E. 2005. Concurrent engineering approach to reducing design delivery time, Journal of Construction Engineering and Management 131(11): 1179-1186. http://dx.doi.org/10.1061/(ASCE)0733-9364(2005)131: 11(1179)

Chua, D. K. H.; Hossain, A. 2011. A simulation model to study the impact of early information on design duration and redesign, International Journal of Project Management 29(3): 246-257. http://dx.doi.org/10.1016/j.ijproman.2010.02.012

Chua, D. K. H.; Tyagi, A.; Ling, S.; Bok, S. H. 2003. Process parameter-interface model for design management, Journal of Construction Engineering and Management 129(6): 653-663. http://dx.doi.org/10.1061/(ASCE)07339364(2003)129:6(653)

Doloi, H.; Sawhney, A.; Iyer, K. C.; Rentala, S. 2011. Analyzing factors affecting delays in Indian construction projects, International Journal of Project Management 30(4): 479-189. http://dx.doi.org/10.1016/j.ijproman.2011.10.004

El-Razek, M. E. A.; Bassioni, H. A.; Mobarak, A. M. 2008. Causes of delay in building construction projects in Egypt, Journal of Construction Engineering and Management 134(11): 831-841. http://dx.doi.org/10.1061/(ASCE)0733-9364(2008)134:11(831)

Enshassi, A.; Arain, F.; Al-Raee, S. 2010. Causes of variation orders in construction projects in the Gaza Strip, Journal of Civil Engineering and Management 16(4): 540-551. http://dx.doi.org/10.3846/jcem.2010.60

Enshassi, A.; Mohamed, S.; Abushaban, S. 2009. Factors affecting the performance of construction projects in the Gaza Strip, Journal of Civil Engineering and Management 15(3): 269-280. http://dx.doi.org/10.3846/1392-3730.2009.15.269-280

Fugar, F. D. K.; Agyakwah-Baah, A. B. 2010. Delays in building construction projects in Ghana, Australasian Journal of Construction Economics and Building 10(1/2): 103-116.

Gabus, A.; Fontela, E. 1973. Perceptions of the world problematique: communication procedure, communicating with those bearing collective responsibility. DEMATEL Report No. 1, Geneva, Switzerland, Battelle Geneva Research Center.

Goodman, R. 1988. Introduction to stochastic models. California, U.S.: Benjamin/Cummings Publishing Company Inc.

Hu, H. Y.; Chiu, S. I.; Cheng, C. C.; Yen, T. M. 2011. Applying the IPA and DEMATEL models to improve the orderwinner criteria: a case study of Taiwan's network communication equipment manufacturing industry, Expert Systems with Applications 38(8): 9674-9683. http://dx.doi.org/10.1016/j.eswa.2011.01.147

Jassbi, J.; Mohamadnejad, F.; Nasrollahzadeh, H. 2011. A fuzzy DEMATEL framework for modeling cause and effect relationships of strategy map, Expert Systems with Applications 38(5): 5967-5973. http://dx.doi.org/10.1016/j.eswa.2010.11.026

Kaliba, C.; Muya, M.; Kumba, K. 2009. Cost escalation ad schedule delays in road construction projects in Zambia, International Journal of Project Management 27(5): 522-531. http://dx.doi.org/10.1016/j.ijproman.2008.07.003

Li, C. W. 2009. A structure evaluation model for technology policies and programs: PhD Dissertation, Institute of Management of Technology, National Chiao Tung University, Taiwan.

Lin, C. L.; Tzeng, G. H. 2009. A value-created system of science (technology) park by using DEMATEL, Expert Systems with Applications 36(6): 9683-9697. http://dx.doi.org/10.1016/j.eswa.2008.11.040

Long, N. D.; Ogunlana, S.; Quang, T.; Lam, K. C. 2004. Large construction projects in developing countries: a case study from Vietnam, International Journal of Project Management 22(7): 553-561. http://dx.doi.org/10.1016/j.ijproman.2004.03.004

Luh, P. B.; Liu, F.; Moser, B. 1999. Scheduling of design projects with uncertain number of iterations, European Journal of Operational Research 113(3): 575-592. http://dx.doi.org/10.1016/S0377-2217(98)00027-7

Martilla, J. A.; James, J. C. 1977. Importance-performance analysis, Journal of Marketing 41(1): 77-79. http://dx.doi.org/10.2307/1250495

MATLAB, 2009. The Mathworks, Inc. [online], [cited 11 June 2009]. Available from Internet: http://www.mathworks.com/patents

Odeh, A. M.; Battaineh, H. T. 2002. Causes of construction delay: traditional contracts, International Journal of Project Management 20(1): 67-73. http://dx.doi.org/10.1016/S0263-7863(00)00037-5

Ogunlana, S. O.; Promkuntong, K.; Jearkjirm, V. 1996. Construction delays in a fast-growing economy: comparing Thailand with other economies, International Journal of Project Management 14(1): 37-45. http://dx.doi.org/10.1016/0263-7863(95)00052-6

Ramanathan, C.; Narayanan, S. P.; Idrus, A. B. 2012. Construction delays causing risks on time and cost--a critical review, Australasian Journal of Construction Economics and Building 12(1): 37-57.

Rivard, H.; Fenves, S. J. 2000. A representation for conceptual design of building, Journal of Computing in Civil Engineering 14(3): 151-159. http://dx.doi.org/10.1061/(ASCE)0887-3801(2000)14:3(151)

Sambasivan, M.; Soon, Y. W. 2007. Causes and effects of delays in Malaysian construction industry, International Journal of Project Management 25(5): 517-526. http://dx.doi.org/10.1016/j.ijproman.2006.11.007

Sanvido, V. E.; Norton, K. J. 1994. Integrated design-process model, Journal of Management in Engineering 10(5): 55-62. http://dx.doi.org/10.1061/(ASCE)9742-597X(1994) 10:5(55)

Senthilkumar, V.; Varghese, K.; Chandran, A. 2010. A web-based system for design interface management of construction projects, Automation in Construction 19(2): 197-212. http://dx.doi.org/10.1016/j.autcon.2009.10.007

Sweis, G.; Sweis, R.; Abu Hammad, A.; Shboul, A. 2008. Delays in construction projects: the case of Jordan, International Journal of Project Management 26(6): 665-674. http://dx.doi.org/10.1016/j.ijproman.2007.09.009

Tonge, J.; Moore, S. A. 2007. Importance-satisfaction analysis for marine-park hinterlands: a Western Australian case study, Tourism Management 28(3): 768-776. http://dx.doi.org/10.1016/j.tourman.2006.05.007

Toor, S. U. R.; Ogunlana, S. O. 2008. Problems causing delays in major construction projects in Thailand, Construction Management and Economics 26(4): 395-408. http://dx.doi.org/10.1080/01446190801905406

Wang, W. C.; Lin, Y. H.; Lin, C. L.; Chuang, C. H.; Lee, M. T. 2012. DEMATEL-based model to improve performance in a matrix organization, Expert Systems with Applications 39(5): 4978-4986. http://dx.doi.org/10.1016/j.eswa.2011.10.016

Wang, W. C.; Liu, J. J.; Liao, T. S. 2006. Modeling of design iterations through simulation, Automation in Construction 15(5): 589-603. http://dx.doi.org/10.1016/j.autcon.2005.07.010

Wei, P. R. 2005. A study of delay causes analysis in the planning and design phases of public construction projects: MS Thesis, Chunghua University, Taiwan.

Wu, H. H.; Tsai, Y. N. 2011. A DEMATEL method to evaluate the causal relations among the criteria in auto spare parts industry, Applied Mathematics and Computation 218(5): 2334-2342. http://dx.doi.org/10.1016/j.amc.2011.07.055

Wu, W. W.; Lee, Y. T. 2007. Developing global managers' competencies using the fuzzy DEMATEL method, Expert Systems with Applications 32(2): 499-507. http://dx.doi.org/10.1016/j.eswa.2005.12.005

Yang, J. B.; Wei, P. R. 2010. Causes of delay in the planning and design phases for construction projects, Journal of Architectural Engineering 16(2): 80-83. http://dx.doi.org/10.1061/(ASCE)10760431(2010)16:2(80)

Yang, J. B.; Yang, C. C.; Kao, C. K. 2010. Evaluating schedule delay causes for private participating public construction works under the build-operate-transfer model, International Journal of Project Management 28(6): 569-579. http://dx.doi.org/10.1016/j.ijproman.2009.10.005

Wei-Chih WANG (a), Chia-Li LIN (b), Shih-Hsu WANG (c), Jang-Jeng LIU (a), Ming-Tsung LEE (a)

(a) Department of Civil Engineering, National Chiao Tung University, Hsinchu, Taiwan

(b) Department of Recreation Management, Shih Chien University, 200 University Road, Neimen, 845 Kaohsiung, Taiwan

(c) Department of Civil Engineering, ROC Military Academy, Kaohsiung, Taiwan table 5-18

Received 05 Nov 2011; accepted 21 May 2012

Corresponding author: Wei-Chih Wang

E-mail: weichih@mail.nctu.edu.tw

Wei-Chih WANG is a Professor in the Department of Civil Engineering at National Chiao Tung University, Taiwan. He received his PhD degree from the University of California at Berkeley in 1996. He is a member of the Construction Management Association of Taiwan. His research interests include construction procurement, high-tech facility project management, and construction simulation.

Chia-Li LIN is an Assistant Professor in the Department of Recreation Management, Shih Chien University, Taiwan. He received his PhD degree from the Institute of Management of Technology at National Chiao Tung University. His research interests include technology policy and cluster spillover, product evaluation and service selection, service development strategies of 4C, electric commerce and service, information behavior and social networking service.

Shih-Hsu WANG is an Assistant Professor in the Department of Civil Engineering at ROC Military Academy, Taiwan. He received his PhD degree from the Department of Civil Engineering at National Chiao Tung University. His research interests include national competitiveness and construction consultant service.

Jang-Jeng LIU is a PhD candidate in the Department of Civil Engineering at National Chiao Tung University, Taiwan. Currently, he is also working as an engineering research at the National Synchrotron Radiation Research Center in Taiwan. His research interests include construction simulation and scheduling of design projects.

Ming-Tsung LEE is working as an Engineering Researcher at the National Synchrotron Radiation Research Center in Taiwan. He received his MS degree from the Department of Civil Engineering at National Chiao Tung University, Taiwan. His research interests include building information modelling and financial management.

Table 1. Comparisons of recent studies in identifying delay factors
for construction projects

                  Focused         Studied
Authors           project phase   countries      Significances

Toor and          Construction    Thailand       For general
  Ogunlana 2008                                    projects

Sweis et al.      Construction    Jordan         For general
2008                                               projects

El-Razek et       Construction    Egypt          For general
  al. 2008                                         projects

Al-Kharashi and   Construction    Saudi Arabia   For general
  Skitmore 2009                                    projects

Enshassi et       Construction    Gaza Strip     For general
  al. 2009                                         projects

Kaliba et         Construction    Zambia         For general
  al. 2009                                         projects

Fugar and         Construction    Ghana          For general
  Agyakwah-                                        projects
  Baah 2010

Yang et           All phases      Taiwan         For general
  al. 2010                                         BOT projects

Yang and Wei      Design          Taiwan         For general
2010                                               projects

Doloi et al.      Construction    India          For general
2011                                               projects

Ramanathan        Construction    All            For general
  et al. 2012                     the              projects;
                                  world            reviewing
                                                   41 past
                                                   studies

Current study     Design          Taiwan         For a specific
                                                   project;
                                                   building an
                                                   evaluation
                                                   model;
                                                   assessing
                                                   causal
                                                   relationships

Authors           Major delay factors

Toor and          (1) Lack of standardization in design; (2)
  Ogunlana 2008     Lack of contract's experience; (3)
                    Inadequate experience of consultant staff;
                    (4) confusing/ambiguous requirements.

Sweis et al.      (1) Financial difficulties faced by the
2008                contractor; (2) too many change orders by
                    the owner; (3) contractor's poor planning
                    and scheduling; (4) shortage of manpower.
El-Razek et         (1) Financing by contractor; (2) delays in
  al. 2008          payment by owner; (3) design changes by
                    owner.
Al-Kharashi and   (1) Shortage of experienced manpower; (2)
  Skitmore 2009     lack of finance by the client; (3)
                    contractor experience; (4) late in
                    reviewing / approving design documents by
                    consultant.
Enshassi et       (1) Delay because of closures leading to
  al. 2009          materials shortage; (2) unavailability of
                    resources; (3) delay in regular payments.
Kaliba et         (1) Delayed payments; (2) financial
  al. 2009          difficulties; (3) contract modification;
                    (4) materials procurement; (5) changes in
                    drawings.
Fugar and         (1) Delay in honoring payment
  Agyakwah-         certificates; (2) underestimation of
  Baah 2010         project cost; (3) underestimation of
                    project complexity; (4) difficulty in
                    assessing bank credit.
Yang et           (1) Change of client's needs; (2) wrong
  al. 2010          client's needs recognized by design team;
                    (3) inexperienced design team; (4)
                    improper plans and schedule developed by
                    design team.
Yang and Wei      (1) Changes in client's requirement; (2)
2010                client's complicated administration
                    process; (3) insufficient or
                    ill-integrated basic project data; (4)
                    inadequate integration on project
                    interfaces; (5) change orders caused by
                    deficiency design.
Doloi et al.      (1) Lack of commitment, inefficient site
2011                management, and poor site coordination are
                    the most critical factors of construction
                    delays; (2) Owner's slow decision and
                    rework due to errors in execution have the
                    greatest impact on delay duration.
Ramanathan        Among the 18 factor groups, the following
  et al. 2012       groups fall into the first five
                    categories: (1) owner; (2) contractor; (3)
                    design related and plant and equipment;
                    (4) labor; and (5) consultant and
                    contractual relationships.
Current study     (1) User needs and specification
                    requirements; (2) organization's decision
                    making and budget constraints; (3)
                    project control and review management; and
                    (4) design execution and interface
                    management.

Table 2. Factors and sub-factors that affect design delays

First-level factors              Description
/ second-level sub-factors

1. User needs and
  specification
  requirements (US)
  1.1 Uncertainty of             Whether or not to include a user
    user needs (US1)               need is uncertain.
  1.2 Clarity of                 User needs are not clearly defined.
    user needs (US2)
  1.3 Difficulty of              User requirements are too strict
    meeting user                   or difficult to meet.
    needs (US3)
  1.4 Limitations                Governmental regulations prevent
    imposed by                     user needs from being met.
    regulations (US4)
  1.5 Limitations                Technical specifications do not
    imposed by                     suffice to meet user needs.
    specifications (US5)
2. Organization's
  decision making and
  budget constraints (OB)
  2.1 DM's decision              Decisions made by decision makers
    making (OB1)                   are not timely or definite.
  2.2 DM's supervising           DM's supervision methods are
    ability (OB2)                  inefficient.
  2.3 Budget                     Project budgets are tight, so DM
    availability (OB3)             must change user needs.
  2.4 DM's resource              DM's budget allocation is inefficient.
    allocation (OB4)
3. Project control and
    review management (PM)
  3.1 PM's management            PM's project control methods are
    method (PM1)                   inefficient.
  3.2 PM's experience            PM's experience and professionalism
    (PM2)                          are insufficient.
  3.3 PM's communications        PM's communications with other
    (PM3)                          parties are inefficient.
  3.4 PM's reviews (PM4)         PM's reviews of design deliverables
                                   are inefficient.
4. Design execution and
    interface management (DM)
  4.1 Design ability (DM1)       A/E's ability in drawings
                                   and design is poor.
  4.2 Designer's cost            A/E's experience of estimating
    estimations (DM2)              costs is poor.
  4.3 Subcontractor              A/E manages his design
    management (DM3)               subcontractors ineffectively.
  4.4 A/E's resource             A/E allocates designers to design
    allocation (DM4)               jobs ineffectively.

Table 3. Example of the response from the ISA questionnaire

Factors and   Degree of    Degree of
sub-factors   importance   satisfaction

1. US
  1.1 US1     8            5
  1.2 US2     8            5
  1.3 US3     7            5
  1.4 US4     9            4
  1.5 US5     8            8
2. OB
  2.1 OB1     8            5
  2.2 OB2     7            5
  2.3 OB3     7            6
  2.4 OB4     7            6
3. PM
  3.1 PM1     9            7
  3.2 PM2     8            8
  3.3 PM3     8            8
  3.4 PM4     7            8
4. DM
  4.1 DM1     8            7
  4.2 DM2     8            7
  4.3 DM3     8            7
  4.4 DM4     7            6

Table 4. Example of the response from the IRM questionnaire

            Factor j   1. US      2. OB      3. PM      4. DM
Factor i
                                             [up
1. US       [right     [up        [right       arrow]
              arrow]     arrow]     arrow]
2. OB
3. PM       [right     [up
              arrow]     arrow]
4. DM

Note: 0: no influence; 1: weak direct influence; 2: moderate direct
influence; 3: strong direct influence; 4: very strong direct
influence.

Table 5. Reliability tests on questionnaires

Reliability tests                         Cronbach's   Results
                                          alpha

In ISA analysis:
  Data on degree of importance            0.899        High
  Data on degree of satisfaction          0.908        High
In IRM analysis:
  Data on first-level factors             0.849        High
  Data on sub-factors under each factor
    1. Data on second-level sub-factor    0.834        High
      under US factor
    2. Data on second-level sub-factor    0.858        High
      under OB factor
    3. Data on second-level sub-factor    0.896        High
      under PM factor
    4. Data on second-level sub-factor    0.881        High
      under DM factor

Table 6. Degrees of satisfaction and importance of factors

          Degree of satisfaction   Degree of importance
Factors   Initial value   SS       Initial value   SI       (SS, SI)
1. US     6.233           0.217    7.772           -1.209   * (+,-)
2. OB     5.583           -1.415   8.153           1.128    x (-,+)
3. PM     6.521           0.939    8.035           0.403    ** (+,+)
4. DM     6.250           0.259    7.917           -0.322   * (+,-)

Table 7. Initial average matrix A of factors

Factors   US      OB      PM      DM      Sum

1. US     0       3.111   2.972   2.861   8.944
2. OB     3.167   0       3.056   2.750   8.973
3. PM     2.500   2.361   0       2.861   7.722
4. DM     2.583   2.361   2.778   0       7.722
Sum       8.250   7.833   8.806   8.472

Table 8. Direct influence matrix D of factors

Factors   US      OB      PM      DM      Sum
1. US     0       0.347   0.331   0.319   0.997
2. OB     0.353   0       0.341   0.307   1.000
3. PM     0.279   0.263   0       0.319   0.861
4. DM     0.288   0.263   0.310   0       0.861
Sum       0.920   0.873   0.982   0.945

Table 9. Indirect influence matrix ID of factors

Factors   US       OB       PM       DM       Sum

1. US     3.196    2.984    3.283    3.195    12.658
2. OB     3.115    3.083    3.290    3.211    12.699
3. PM     2.783    2.68     2.995    2.836    11.294
4. DM     2.782    2.682    2.924    2.915    11.303
Sum       11.876   11.429   12.492   12.157

Table 10. Total influence matrix T of factors

Factors   US       OB       PM       DM       Sum (d)

1. US     3.196    3.331    3.614    3.514    13.654
2. OB     3.468    3.083    3.631    3.518    13.700
3. PM     3.062    2.943    2.995    3.155    12.154
4. DM     3.070    2.945    3.234    2.915    12.164
Sum (r)   12.795   12.302   13.474   13.101

Table 11. Degree of total influence of factors

Factors   d        r        d+r      d-r

1. US     13.654   12.795   26.449   0.859
2. OB     13.700   12.302   26.002   1.398
3. PM     12.154   13.474   25.627   -1.320
4. DM     12.164   13.101   25.266   -0.937

Table 12. Net influence matrix N of factors

Factors   US       OB       PM      DM

1. US     -
2. OB     0.137    -
3. PM     -0.552   -0.688
4. DM     -0.444   -0.573   0.080   -

Table 13. Suggested strategies for improving delay factors

          ISA                            IRM
Factors   SS       SI       (SS, SI)     d+r      d-r      Group

1. US     0.217    -1.209   * (+,-)      26.449   0.859    Cause
2. OB     -1.415   1.128    X (-,+)      26.002   1.398    Cause
3. PM     0.939    0.403    ** (+,+)     25.627   -1.320   Effect
4. DM     0.259    -0.322   * (+,-)      25.266   -0.937   Effect

Factors   Strategies

1. US     A
2. OB     B
3. PM     A
4. DM     A

Note: Strategy A: factor requires no further improvement (SS > 0);
Strategy B: factor must be improved directly (SS < 0 and d - r > 0).

Table 14. Strategies for improving second-level delay factors under
OB factor

                                ISA                            IRM
                       SS       SI       (SS, SI)     d+r      d-r

2.1 DM's decision      -2.245   1.967    X (-,+)      13.138   0.860
  makings (OB1)
2.2 DM's supervising   -1.383   1.093    X (-,+)      12.657   0.367
  ability (OB2)
2.3 Budget             -1.014   -0.154   *** (-,-)    12.133   0.037
  availability (OB3)
2.4 DM's resource      -0.398   -1.152   *** (-,-)    12.095   -1.264
  allocation (OB4)

                       Group    Strategies

2.1 DM's decision      Cause    B
  makings (OB1)
2.2 DM's supervising   Cause    B
  ability (OB2)
2.3 Budget             Cause    B
  availability (OB3)
2.4 DM's resource      Effect   C
  allocation (OB4)
COPYRIGHT 2014 Vilnius Gediminas Technical University
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Wang, Wei-Chih; Lin, Chia-Li; Wang, Shih-Hsu; Liu, Jang-Jeng; Lee, Ming-Tsung
Publication:Journal of Civil Engineering and Management
Article Type:Report
Geographic Code:1USA
Date:Aug 1, 2014
Words:9647
Previous Article:Diagnosis and repair of gypsum plaster coatings: statistical characterization and lessons learned from a field survey.
Next Article:Statistical survey of the pathology, diagnosis and rehabilitation of etics in walls.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters