Integrating exploratory factor analysis and fuzzy AHP models for assessing the factors affecting the performance of building construction projects: The case of Ethiopia

Abstract

Project performance is a critical issue for the construction industry. The construction industry is complex in its nature because it contains a large number of parties involved in the planning, execution, and monitoring of all types of civil works. Moreover, construction industries are always unique and heterogeneous; they will always encounter many unforeseen events that will negatively affect the construction everywhere, and the construction projects suffer from many problems and complex issues in the performance of building projects. Therefore, the study was focused on assessing the factors affecting the performance of building projects. The study would employ an integrated methodology of factor analysis and fuzzy AHP methods. First, the factor analysis is used to classify and reduce the input variables and their variable coefficients are determined. Second, fuzzy AHP is used to determine the preference weights of the input variables, and for ranking. To achieve this objective, data were collected from primary and secondary sources. This study used an integrated methodology as a data analysis method, and SPSS software version 23 and Microsoft Excel were used as analysis tools. The study pointed out that project location (0.083), changes in design (0.068), market fluctuation (0.067), financial difficulties (0.066), and size of the project (0.064) are identified as the most significant factors that affect the performance of building construction projects. Finally, it can be recommended that the practitioners take appropriate actions in improving the performance of building projects, and the methodologies can be implemented to demonstrate its practicality.

Keywords:

• building construction
• factor analysis
• fuzzy AHP
• performance

PUBLIC INTEREST STATEMENT

The current study focuses on Integrating Exploratory Factor Analysis and Fuzzy AHP Models for assessing the factors affecting the performance of building construction projects. To achieve the objectives, a literature review was conducted to identify the input variables and prioritize the impact level of the identified factors on the performance of building construction projects. Furthermore, one of the fundamental aims of this study was to identify the most significant factors that affect the performance of building construction projects. In this study, the participants were selected based on their prior knowledge, skill, and competencies. The findings of this study are thought to be helpful to provide a reference guide for the building construction company, academicians, and researchers, and for further exploration on the topic.

1. Introduction

The construction industry is one of the most dynamic and responsive sectors at both the global and national levels (Nyangwara & Datche, 2015). The construction industry is complex in its nature because it contains a large number of parties as clients, contractors, consultants, stakeholders, shareholders, and regulators involved in the planning, execution, and evaluation (monitoring) of all types of civil works, such as buildings, roads, railways, water works, dams, and hydropower construction (Bekr, 2017). However, as construction projects are always unique and heterogeneous, they will always encounter many unforeseen events that will negatively affect the construction process everywhere (Keng, 2015). In developing countries, the industry can be affected by different challenges, and it is difficult to achieve the intended objective which is to complete the construction works on time, as per the estimated budget and desired quality (Ofori, 2000).

The success of any project is obtained through proper management from the beginning to the end of the project undergoes through the triple constraints of time, money, and quality the balancing of which results in a satisfactory project output (Salikuma et al., 2016).

Efficient management of projects in construction industries is becoming a challenge with time. It can be a poor performance in monitoring and control of the project and a lack of organizational capability. Accurate performance indicators such as cost and schedule performances are critical to support decision-making and predict project results (Al-Zwainy et al., 2020; Pookot, 2017).

The most important gap enforced for conducting the study was building construction projects in Ethiopia face a variety of problems and complex issues in performance because the projects are full of activities, tasks, and constraints. The complexity and other factors affect the performance of the project significantly in terms of cost, time, and quality (G. M. Ayalew et al., 2023).

In doing this study, the geographical location in which the study was conducted, the number of population, statistical analysis, and research methodologies were identified as the most significant gaps, and it makes the study unique from other study to assess the problem in today’s Ethiopia construction.

Moreover, the academicians did not find any study that is the same as the current study in the literature while searching in international libraries and discrete periodicals such as Scopus, Springer, Taylor France, and others, aimed at integrating the factor analysis and Fuzzy AHP methodologies for assessing the factors affecting the performance of building construction projects. Thus, conducting a study on the applied methodologies for identifying the most significant factors that affect the performance of building construction projects is a hot research topic and the best area of exploration to be studied to enhance the performance of building construction projects.

The evaluation of qualitative data by the decision expert is always subjective, imprecise, and vague (Meharie, Gariy, Ngumbau, et al., 2019). In such a situation, fuzzy set theory is a suitable and powerful tool in dealing with the uncertain environment with vagueness, imprecision, and ambiguity (G. G. Ayalew et al., 2022)

Thus, this study integrated a factor analysis and a fuzzy analytical hierarchy process (fuzzy AHP) approach and provides to identify rational and systematic input variables (factors) that affect the performance of building construction projects while filling the knowledge gaps mentioned above.

Therefore, based on the identified problem assessing the factors affecting the performance of building construction projects in Ethiopia was identified as the most important problem to be studied for improving the performance of building projects.

Thus, the objectives of the study focused on integrating exploratory factor analysis and fuzzy AHP models for assessing the factors affecting the performance of building construction projects in Ethiopia. Conducting this study is almost unique of its kind for assessing the most significant factors affecting the performance of building construction projects. To achieve this study, there is a need to explore the factors that affect the performance of building construction projects in the recent literature. Examining the significant factors in the building construction projects at early stages can enable us to undertake a remedial measure on the identified factors, and make adjustments to the project performance regarding the project cost and project time.

The study provides a novel and a significant contribution to previous studies on the development of integrated models which are exploratory factor analysis and fuzzy AHP models on the building project performance factor assessment. In addition, these methodologies provide as to conduct an assessment based on their relative impotence, subjectivity, and uncertainties are taken into account through the fuzzy set theory in a fuzzy environment increase the validity and reliability of the study.

The importance of this study is as a reference guide for the company in general to know the most significant factors that affect the performance of building construction projects. This can serve us to reduce the nonperformance of the building construction projects. This study even provides a secondary source of data for future researchers to conduct a study through using the applied methodologies.

Moreover, ranking of the factors will enable the practitioners in the building construction projects in Ethiopia to take appropriate actions in improving the performance of the construction projects.

The remainder of this paper is organized as follows. Section 1 Construction project management, and factors affecting the performance of building construction projects, and Section 2 presents how the integrated methodology of factor analysis and fuzzy AHP can be adopted. Section 3 shows numerical analysis and results of factor analysis and fuzzy AHP results along with some discussions relating to the factors affecting the performance of building construction projects. Section 4 presents discussions of the findings, managerial implications, and limitations. And in section 5, conclusions and remarks are discussed.

2. Literature review

The construction industry is vital for the development of any nation. In many ways, the pace of the economic growth of any nation can be measured by the development of physical infrastructures, such as buildings, roads, and bridges (Oke et al., 2016).

Albtoush et al. (2022) stated that the construction industry plays an important role in the economy, and the activities of the industry are also vital to the achievement of national socio-economic development goals of providing shelter, infrastructure, and employment. As Gituro and Mwawasi stated, many projects in developing countries encounter considerable time and cost overruns, fail to realize their intended benefit, or are even totally terminated and abandoned before or after their completion.

Oke et al. (2016) also noted that management practices in Ethiopia are even far behind those of poor-performing developing countries in Africa. The level of construction project management practice in terms of adapting general project management procedures, project management functions, tools, and techniques is unsatisfactory.

2.1. Construction project performance

Project performance is a critical issue for the construction industry. Project deliverables, such as timely completion and client satisfaction, are often used as yardsticks to determine success (Takim & Akintoye, 2002).

Performance management is a continuous process of identifying, measuring, and developing performance in organizations by linking each individual’s performance and objectives to the organization’s overall mission and goals (Zewdie, 2016).

Project performance gives a better understanding of the overall project goals. It is a more organized way to manage the project. It also helps in more efficient use of project resources, faster project completion, and lower project costs (Vyas & Birajdar, 2016).

2.2. Factors affecting the performance of building construction projects

Most construction projects suffer from cost and time overruns due to a multiplicity of factors (Bhosekar & Vyas, 2012). Cost overrun and delay in projects are the foremost challenges associated with nearly all projects in the construction industry. The inclination of construction projects toward overruns is due to the risk and uncertainties (Sharma & Gupta, 2021). The following factors were identified in the literature as the factors affecting the performance of building construction projects.

2.2.1. Construction parties related factors

In any construction project, there are three main parties involved in the construction projects. Each of the construction parties has a significant role in the construction project the building contractor plans and coordinates construction activities and must complete the project within the established time and budget (Zewdie, 2016).

2.2.1.1. Method of construction

According to Abay et al. (2019), the method of construction projects depends on the amount of work means that when the amount of work executed is high the building projects be executed at a higher construction cost. On the contrary, when the task to be executed is small, it needs lower resources, and due to these reasons, method of construction is a factor in the project performance.

2.2.1.2. Lack of supervision

Sahat et al. (2017) stated that supervision (controlling) is a standard arrangement, such as sales quotas, quality standards, or level production; an examination to assess actual work performance compared to predefined standards; and taking necessary corrective action.

2.2.1.3. Shortage of technical personnel

Memon et al. (2012) stated that very much relevant factors of time and cost performance were lack of communication between parties, delay in obtaining permits from governmental agencies, lack of coordination between parties, delay in progress payment by the owner, and financial difficulties of the owner.

2.2.1.4. Inaccurate quantity takes off

While carrying out the construction projects, accurate quantity takeoff helps to provide a better estimation of the construction cost. The quantity of the construction can be determined based on detailed design and clear specifications to achieve the final products of the project (Jasper et al., 2017).

2.2.1.5. Delay in payment

A delay in payment of completed works from the owner significantly affects the cash flow of the contractor and causes a delay in the procurement of resources (Memon et al., 2012).

2.2.1.6. Management condition

According to Huang and Ho (2013), management conditions are an integral part of the project management strategy consisting of items and resources needed for project completion. It is provided by the contracting parties to facilitate a completed project contribute to execute the adequate compensation is necessary for poor management conditions (Molla et al., 2020).

2.2.1.7. Contractor’s cash flow

According to Lendo-Siwicka et al. (2016), construction cash flow is the balance of the amount of cash being received and paid by a business during a defined period, sometimes tied to a specific project.

2.2.1.8. Poor site management

According to Zain et al. (2021), ineffective site management has created many problems. Problems such as project delays, contractual problems, and financial problems occur and affect the development of a project and may lead to the abandonment of many abandoned projects (Xie et al., 2019).

2.2.2. Project-related factors

Osuizugbo and Okuntade (2020) stated that project-related factors considered in the building projects that affect the performance of the project were client-related, contractor-related, consultant-related, material-related, and labor- and equipment-related.

2.2.2.1 Conflict among Project Participants

According to Osuizugbo and Okuntade (2020), conflict management is a process of communication for changing the negative emotions in conflict to a state of emotions that allows for working out a solution to the conflict.

2.2.2.2. Owners interference

According to Lendo-Siwicka et al. (2016), client interference relates to all those acts or omissions made by the client which adversely affect the project and hamper the project’s performance and can be defined as “a word or statement that expresses denial, disagreement, or refusal.”

2.2.2.3. Size of project

According to Mahamid (2014), the size of a project or program will impact the degree of difficulty in achieving its objectives, but large projects are not necessarily complicated or complex. According to Lendo-Siwicka et al. (2016), the size of a construction project (contract amount) influences the rate of cost overrun and the project completion date.

2.2.2.4. Inadequate planning and scheduling

According to Acebes et al. (2015), a construction project plan is prepared which is defined as a management summary document that describes the essentials of a project in terms of its objectives, justification, and how the construction cost and construction time objectives are to be achieved.

2.2.2.5. Change in Scope and Design

Tochaiwat (2016) stated that changes to scope can impact deadlines, budgets, and project quality. A scope change process should be outlined in the project plan so that all stakeholders are aware of the protocol for efficiently managing any potential change requests with minimal project disruption (Huang & Ho, 2013).

2.2.2.6. Complexity of the project

According to Lendo-Siwicka et al. (2016), the concept of complexity can relate to any subject and therefore there is a wealth of information about it. Project success in terms of cost, time, and quality has historically been poor in the construction.

2.2.2.7. Number of floors

According to Tochaiwat (2016), the number of floors is one of the important parameters for developers in selecting products suitable to their projects. He also found that the number of floors also affects the construction performances, in terms of duration, costs, and quality of the public users.

2.2.2.8. Delays in decision-making process

According to Labor (2019), a decision made will be carried out only after a certain amount of time elapses due to regulatory reasons at that instant it affects time. Sahat et al. (2017) found that slow decision-making delays the project completion as the contractor waits for confirmation to carry out certain works.

2.2.3. Resource-management-related factors

Building construction project resource management is the practice of planning, scheduling, and allocating people, money, and technology to a project or program. Thus, to achieve the performance of the projects, the process of allocating resources to achieve the greatest organizational value, the right resources being available at the right time for the right work (Lendo-Siwicka et al., 2016).

2.2.3.1. Equipment Failure at the Work Site

(Bourassa et al) stated that accidental events in manufacturing industries can be caused by many factors, including work methods, lack of training, equipment design, maintenance, and reliability.

2.2.3.2. Hiring of equipment

Purchasing the construction equipment in a construction project is a tricky decision for construction projects. This is due to methods of assessing the construction project having a significant effect on the performance of building construction projects (Abdellatif & Alshibani, 2019).

2.2.3.3. Efficiency of equipment

The equipment efficiency evaluates how well a machine performs its designed task in terms of quantity and quality of performance to achieve the performance. The measurement of equipment’s efficiency is an important evaluation because underutilization of increased production costs can have an effect on the project performance (Lendo-Siwicka et al., 2016).

2.2.3.4. Shortage of skilled labor

According to Sahat et al. (2017), poor workmanship occurs due to a lack of care by the contractor while installing the material, resulting in poor finishing and the product being non-functional, and these affect the performance of the projects.

2.2.3.5. Productivity of manpower

Azim Eirgash (2019) highlighted that manpower productivity reflects how efficiently the manpower is combined with other factors of production, how many of these other inputs are available per worker, and how rapidly embodied and disembodied technical change proceeds.

2.2.3.6. Late delivery of materials on site

According to Abdellatif and Alshibani (2019), the main cause of project cost overruns was identified to be related to the shortage of materials and the deficiency in the financing, and a recommendation to overcome that was to reduce the mistake of handling human resources and materials.

2.2.3.7. Materials shortage

Sahat et al. (2017) found that not all materials imported from overseas manufacturers are produced by them. The raw materials or components of certain materials are sometimes ordered and collected from other countries.

2.2.3.8. Unfavorable weather conditions

In construction projects, unfavorable weather conditions are potentially harmful weather conditions to reduce risk, and health and safety problems (Abdellatif & Alshibani, 2019).

2.2.4. Project condition related factors

The project condition is a visual representation of how the project is progressing. It is a reportable variable determined by the relationship between the planned, projected, and estimated dates of the project (Richardson et al., 2016).

2.2.4.1. Lack of experience

In construction projects, the capability and capacity of manpower are also considered in the construction project, and it can be identified as the most significant factor in the project. This mostly focuses on the talent and knowledge of the personnel in construction projects (Omran).

2.2.4.2. Project location

According to Zewdie (2016), constructing a facility in a locality is very different from constructing one in other areas. The differences are in labor costs, the availability of materials, equipment, delivery logistics, and climate conditions.

2.2.4.3. Rework

According to Sachdev and Agrawal (2017), activities in construction projects that have to be done more than once in the project site, or activities that remove work previously installed as part of the project regardless of source, where no change order has been issued and no change of scope has been identified by the owner.

2.2.5. Economic condition related factors

Economic conditions refer to the state of an economy that determines the scale of production and consumption activities that relate to determining how resources are allocated to the overall economy of a country or geographical region through various business and economic cycles.

2.2.5.1. Financial difficulties

Akali and Sakaja stated that the financial problem leads the contractor to order variations to suspend the projects due to economic crises to reduce the contract amount to make the budget match with his financial capability and to make the project feasible.

2.2.5.2. Market fluctuation

According to Sachdev and Agrawal (2017), escalation of material prices has been the critical factor that leads to project cost overrun and affects the cost performance of the project. Memon (2014) stated that fluctuation in the price of the material, shortage of site workers, lack of communication between parties, and frequent design changes are most factors construction performance.

2.2.5.3. Interest Rate

Azim Eirgash (2019) stated that in terms of economic theory, interest is the price paid for inducing those with money to save it rather than spend it, and to invest in long-term assets rather than hold cash. The rate of interest affects the project cost.

2.2.5.4. Exchange rate fluctuation

Journal and May (2020) reported the profit rate of project, material, equipment, labor costs, and escalation of material prices. The exchange rate indicates an unanticipated increase in the foreign currency price of the domestic currency (Kandil, 2009).

2.2.5.5. Cost of variation orders

Akomah et al. (2018) stated that variation of cost is an almost inevitable situation in construction projects. Variations which occur on a given project are unique and can be linked to the extent of time and money made available for planning and play an important role in determining the closing cost and time of projects (Hassan & Ali, 2020).

2.2.5.6. Political instability

Faremi and Ogunsanmi (2019) stated that institutionalization could be measured by the level of “adaptability, complexity, autonomy, and coherence” of these organizations and procedures, and political instability is a function of the decay of political institutions, and there is significant effect to be considered in the construction project performance.

2.2.6. Design and documentation related factors

Design and documentation is a collection of documents and resources that covers all aspects of your product design. Documentation should include information about users, product features, and project deadlines; all essential implementation details; and design decisions that your team and stakeholders have agreed on.

2.2.6.1. Changes in design

Aslam et al. (2019) found that the phenomenon of cost overrun due to design changes is universal. Thus, almost every country is experiencing the unfavorable effect of design changes on the cost and time performances.

2.2.6.2. Unclear specification

The construction specification that is used in the construction project should be properly written, clear, and concise. The specification provides detailed information on the construction time and even detailed information on the quality of the materials used in the construction project (Shenoy, 2015).

2.2.6.3. Delays in design work

A study in Shenoy (2015) reported that delay in design is one of the most significant risks to project delays. He also stated that the majority of structural failures and associated damage costs are due to errors in planning, design, construction, and utilization, rather than variability in construction material, strengths, and structural loads.

2.3. Summary of literature review

Some of the studies identified the most important factors in several countries and various project types, while other studies discussed the influence of factors on the success of projects, and proposed ways to improve construction performance (Journal & May, 2020).

Several studies have been conducted on the factors that affect the performance of construction projects in a number of ways, some of which are presented and summarized in Table 1. Similarly, the authors discovered that project location, changes in design, market fluctuation, financial difficulties, size, and complexity of the project were found to be the most significant factors that affect the performance of building construction projects. The authors added that planning for the project should be considered when constructing the building projects, failing to do this may escalate the cost of the project and result in time overrun, hence it affects the performance of the project. Moreover, the authors conducted his study based on the previous study; thus, the study has come up with six major project factors, and 36 sub-factors for their similarities with the Ethiopia environment and its social, economic, and political conditions.

Table 1. Summary of the previous studies related to construction project performance factors in different countries

Authors	Methods used	Project	Country	Project performance factors (findings)
Nyangwara and Datche (2015)	Descriptive Statistics	Any	Kenya	Average delay in claim approval and payment approval, Availability of resources, Leadership skills, and Escalation of material prices.
Salikuma et al. (2016)	Case Study	Any	Different Countries	Specifications of works, Quantity of materials, Cost of labors, and Location of site of work to track the project in terms of work, time and money.
Albtoush et al. (2022)	Factor analysis	Any	Jordan	Quality-related factors, cost-related factors, time related factors, and contract-related factors
Takim and Akintoye (2002)	Case Study	Any	UK	Construction cost, construction time, cost predictability, time predictability, and client satisfaction.
Vyas and Birajdar (2016)	Case Study	Public	Different Countries	Risk, project team in decision making, and firm commitment
Bhosekar and Vyas (2012)	Case Study	Public	Different Countries	Labor, materials and equipment are used to track time and costs for the project.
Sharma and Gupta (2021)	Descriptive Statistics	Any	Different Countries	Delay in obtaining permission from authorities, Poor supervision and site management, unforeseen ground conditions, and unrealistic time schedule.
Sahat et al. (2017)	Case Study	Private	Brunei Darussalam	Availability of construction materials, materials, inventory management system, and type of materials.
Memon et al. (2012)	Descriptive Statistics	Public	Malaysia	Design and documentation issues, financial resource management and project management.
Molla et al. (2020)	Descriptive Statistics	Public	Ethiopia	Liquidity of the organization, the quality of equipment and raw materials
Memon (2014)	Descriptive Statistics	Public	Malaysia	Shortage of site workers, lack of communication between parties, incorrect planning, and scheduling
Boateng and Abaye (2019)	Case Study	Private	Different Countries	Material shortages, unavailability of resources, and a low level of project leadership skills
Aidan et al. (2020)	Descriptive Statistics	Public	Different Countries	Financial difficulties faced by the contractor, manpower shortages (of skilled, semiskilled, or unskilled labor).

3. Methodology

The main purpose of this study was to examine the factors that affect the performance of building construction projects by integrating combined methodologies of exploratory factor analysis, and fuzzy AHP. Exploratory factor analysis can be used to reduce the number of items in a questionnaire by identifying the underlying structure of items in a data set and determining whether the data can be explained by a smaller number of domains (Vahidnia et al., 2008). Similarly, in the literature, the FAHP is a predominant approach in constructing an evaluation model.

This study encompasses both quantitative and qualitative research to obtain information for the objective of the study. Quantitative approaches were conducted by questionnaire survey to obtain information for the level of factors by using Likert scale, whereas qualitative research was conducted for fuzzy AHP multi-criteria decision-making (MCDM) problem to prioritize values of input variables by using Saaty’s scale. Fuzzy AHP is often subjective, and it requires the broad domain knowledge of the practitioners and industry experts. The Fuzzy AHP model provides for the conversion of this subjective expert knowledge to an objective input variable index, whereas exploratory factor analysis is a quantitative multivariate method in which the interrelationships between a set of continuously measured variables are defined through a variety of underlying linearly independent reference variables (Albtoush et al., 2022).

3.1. Questionnaire design

The study employs exploratory factor analysis, and fuzzy AHP approach to examine the factors affecting the performance of building construction projects. The questionnaire has three sections. The first section focused on the demographic information of respondents; the second questionnaire was developed to collect data for factor analysis. The questionnaire survey was distributed to 68 professionals and academicians, who were chosen as the target populations, who have had sufficient work experience in the building construction industry in Ethiopia, via e-mail and physical presence.

Respondents of the survey were asked to answer “rate the factors affecting the performance of building construction projects?” on each of the 36 variables prepared based on a 5-point Likert scale (5—very high, 4—high, 3—moderate, 2—low, and 1—very low) for analyzing the data by using SPSS Software version 23 for reduction of unfavorable variables, and then after the third questionnaire can build as a pairwise comparison matrices analysis with fuzzy scales.

The relative weights of each factor were developed using a 9-point scale comparison matrix as recommended by Saaty’s (Saghafian & Hejazi, 2005). The scale ranges from 1 = equally important up to 9 = absolute important, as presented in Table 2. Similarly, the number of decision experts in Fuzzy AHP methods was selected based on their subject knowledge, experience, skill, and personal contacts.

Table 2. Membership function of linguistic scale (Kil et al., 2016)

Fuzzy number	Linguistic variables	Scale of fuzzy number
1	Equally frequent	(1, 1, 1)
2	Equal to moderate frequent	(1, 2, 3)
3	Moderate frequent	(2, 3, 4)
4	Moderate to strong frequent	(3, 4, 5)
5	Strong frequent	(4, 5, 6)
6	Strong to very strong frequent	(5, 6, 7)
7	Very strong frequent	(6, 7, 8)
8	Very strong to absolute frequent	(7, 8, 9)
9	Absolute frequent	(9, 9, 9)

3.2. Sample size and sampling technique

In this study, the target populations were purposively conducted on only professional officials because the researcher believes that the respondents have better theoretical and technical knowledge and information concerning the study’s issues. The numbers of decision experts were selected based on their subject knowledge and industry experience (5 years or more), and educational background (masters and above) to collect the response or opinion. In this study, a total response of 68 experts were received through emails and face-to-face interviews from project participants for a preliminary questionnaire set for carrying out the factor analysis. Likewise, the sample sizes used for FAHP model were 10 decision experts, because the experts in multi-criteria decision-making problem may depend on the nature of the problem and the availability of experts.

According to Fu et al. (2020), FAHP model is applicable to the responses of small groups through the upper limit number of experts should not exceed 10. Since responses from more than 10 experts lead to a high degree of inconsistency, it makes your results unreliable. Thus, in this study, 10 representative decision expert panels were chosen, and a series of face-to-face interviews were conducted for each decision experts in a group out of 68 project participants to rank the importance levels of the variables.

In this study, a non-probability sampling technique and a purposive approach were adopted because it allows the selection of respondents based on available information, and whose experience permits an understanding of the factors affecting the performance of building construction projects. The adoption of this technique also helps us to eliminate respondents who did not fit the requirements. In the case of multi-criteria decision-making problems, the sampling technique used for conducting the study was consensus-based MCDM methods. We call it “Experts Panel” in MCDM methods (G. G. Ayalew et al., 2022). Hence, the decision experts were selected based on their profession, commitment to participate in the study, and personal contacts.

3.3. Study variable

3.3.1. Independent variable

Independent variables cause a change in the independent variables (Wani, 2017). In this study, factors that affect the performance of building construction projects such as construction parties, project-related factors, resource-management-related factors, project condition, economic condition, and design- and documentation-related factors are major independent variables.

3.3.2. Dependent variable

Dependent variable is the response variable or output of the character that changes due to variations in the independent variables (Wani, 2017). In this study, the dependent variable is the performance of building construction projects. It is considered dependent because its value depends upon the value of the independent variables. Hence, the performance of building construction project is a factor that is measured depending on the effectiveness of the independent variable’s effect.

First, the data collection method employed for conducting this study to examine the most influential factors that affect the performance of building construction projects was done through a questionnaire survey as the primary source of data collection method. In the second part, the method of data collection was extracted according to what was mentioned in the previous studies that are suitable for carrying out the current study was defined in the literature to answer the research question.

3.4. Data processing and analysis methods

In this study, exploratory factor analysis and Fuzzy AHP methodologies were used as data analysis methods, and Statistical Package for Social Science and Microsoft Excel were used as a statistical tool.

3.4.1. Exploratory factor analysis

Exploratory factor analysis is a statistical method employed to increase the reliability of the scale by identifying inappropriate items that can be removed and eliminating the inappropriate factor should be considered because it may decrease the standard of correlation among the rest of the items (Somiah et al., 2015)

According to Yu and Richardson (2015), exploratory factor analysis technique has three key uses: (a) to understand the structure of a set of variables; (b) to construct a questionnaire to measure an underlying variable; and (c) to reduce a data set to a more manageable size without much loss of the original information. In factor analysis, individual questionnaire items are grouped into homogeneous domains that represent common characteristics.

Principal component analysis and common factor analysis are the two main factor analytical techniques for carrying out factor analysis. In this study, the principal component analytical technique was applied. To apply exploratory factor analysis based on principal component analysis, the following steps and procedures followed (Jugessur, 2022). The SPSS version 23 was employed to perform the analysis:

(i) Assessing correlation matrices: Correlation matrices are symmetrical, with values below the diagonal mirroring those above as it is observed in Table 9. The diagonal of the matrix consists of initial communality values, equal to 1.0.

Table 3. Random consistency index (RI) (Vahidnia et al., 2008)

n	1	2	3	4	5	6	7	8	9
RI	0	0	0.6	0.9	1.1	1.2	1.3	1.4	1.5

Table 4. The triangular fuzzy number of positive/negative linguistic scale (Wang et al., 2019)

Linguistic variable	Fuzzy scale (positive linguistic variable)	Triangular fuzzy number	Fuzzy scale (negative linguistic variable)	Triangular fuzzy number
Equally frequent	$\tilde{1}$	(1, 1, 1)	${\tilde{1}}^{-1}$	$\left(\frac{1}{1},\frac{1}{1},\frac{1}{1}\right)$
Equal to moderate frequent	$\tilde{2}$	(1, 2, 3)	${\tilde{2}}^{-1}$	$\left(\frac{1}{3},\frac{1}{2},\frac{1}{1}\right)$
Moderate frequent	$\tilde{3}$	(2, 3, 4)	${\tilde{3}}^{-1}$	$\left(\frac{1}{4},\frac{1}{3},\frac{1}{2}\right)$
Moderate to strong frequent	$\tilde{4}$	(3, 4, 5)	$4^{-1}$	$\left(\frac{1}{5},\frac{1}{4},\frac{1}{3}\right)$
Strong frequent	$\tilde{5}$	(4, 5, 6)	${\tilde{5}}^{-1}$	$\left(\frac{1}{6},\frac{1}{5},\frac{1}{4}\right)$
Strong to very strong frequent	$\tilde{6}$	(5, 6, 7)	${\tilde{6}}^{-1}$	$\left(\frac{1}{7},\frac{1}{6},\frac{1}{5}\right)$
Very strong frequent	$\tilde{7}$	(6, 7, 8)	${\tilde{7}}^{-1}$	$\left(\frac{1}{8},\frac{1}{7},\frac{1}{6}\right)$
Very strong to absolute frequent	$\tilde{8}$	(7, 8, 9)	${\tilde{8}}^{-1}$	$\left(\frac{1}{9},\frac{1}{8},\frac{1}{7}\right)$
Absolute frequent	$\tilde{9}$	(9, 9, 9)	${\tilde{9}}^{-1}$	$\left(\frac{1}{9},\frac{1}{9},\frac{1}{9}\right)$

Table 5. Respondents demographics

Parameter	Frequency	Percentage (%)	Parameter	Frequency	Percentage (%)
Job Position			Gender
Construction Engineer	13	19.12%	Male	52	76.47%
Project Manager	11	16.18%	Female	16	23.53%
Office Engineer	9	13.24%	Field of Specialization
Site Engineer	17	25.00%	Civil Engineering	38	55.88%
Supervisor	7	1.29%	Structural Engineering	12	17.65%
Resident Engineer	5	7.35%	Construction Engineering	18	26.47%
Senior Engineer	6	8.82%	Work experience of respondents
Educational level			6–10 years	31	45.59%
Msc. Degree	61	89.70%	11–15 years	24	35.29%
PhD.	7	1.29%	≥15years	13	19.12%

Table 6. Descriptive statistics of each element of the factors affecting the performance of building construction projects

Descriptive Statistics
	N	Minimum	Maximum	Mean	Std. Deviation	Skewness		Kurtosis
	Statistic	Statistic	Statistic	Statistic	Statistic	Statistic	Std. Error	Statistic	Std. Error
Construction Parties	68	1.00	4.75	2.926	.80925	.082	.306	−.675	.604
Project Related	68	1.00	5.00	3.049	1.00892	−.321	.306	−.442	.604
Resource Management Related	68	1.00	4.67	2.918	1.00491	−.337	.306	−.707	.604
Project Condition	68	1.00	4.40	2.908	.91839	−.362	.306	−.663	.604
Economic Condition	68	1.00	5.00	2.879	.94287	−.245	.306	−.374	.604
Design and Documentation	68	1.00	4.60	2.685	1.08410	−.151	.306	−1.096	.604
Valid N (listwise)	68

Table 8. Eigenvalues, total variances explained for the final six-factor structure

Component	Initial eigenvalues			Extraction sums of squared loadings			Rotation sums of squared loadings
	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %
1	6.071	26.396	26.396	6.071	26.396	26.396	4.219	18.345	18.345
2	3.836	16.677	43.073	3.836	16.677	43.073	3.450	15.001	33.347
3	2.762	12.007	55.080	2.762	12.007	55.080	2.735	11.889	45.236
4	2.160	9.390	64.470	2.160	9.390	64.470	2.680	11.652	56.888
5	1.887	8.205	72.675	1.887	8.205	72.675	2.656	11.548	68.436
6	1.554	6.756	79.430	1.554	6.756	79.430	2.529	10.994	79.430

Table 9. Factor correlation matrix

Factor		1	2	3	4	5	6
1		1.000
2		.078	1.000
3		.006	.258	1.000
4		.156	.274	.082	1.000
5		.337	.286	.307	.128	1.000
6		.103	.343	.275	.150	.214	1.000

Note: Extraction Method: Principal Axis Factoring.

Rotation Method: Promax with Kaiser Normalization.

Methods to assess the correlation matrices were carried out through the following methods. The first method directly assesses the correlation matrix value above or below the diagonal (greater than 0.80) suggesting that these items may be redundant and could be excluded, whereas weakly correlated values (less than 0.30) indicate a lack of shared variance, a requirement for factor analysis. The second method assesses the initial communalities within a correlation matrix. Values greater than 0.60 are desirable, indicating that the shared variance is enough to suggest that the matrix is factorable. The third method, Bartlett’s Test of Sphericity, should reach a statistical significance of less than .05 to conduct an EFA.

The fourth method for accessing correlation matrices was the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy test used to verify the sampling adequacy for the analysis. It helped us to evaluate how strongly an item is correlated with other items in the EFA correlation matrix. KMO values above 0.90 suggest that an item is ”marvelous” at contributing to the interrelationship with other items, between 0.80 and 0.90 is “meritorious,” between 0.70 and 0.80 is “middling,” and less than 0.60 is “mediocre,” miserable,”or “unacceptable” (Wahed, 2018).

(ii) Extraction of principal components or factors with eigenvalues greater than 1, known as Kaiser’s criteria (eigenvalue >1 rule), and assessing the factor loading matrix. The next step in the factor analytical process is to determine the extent to which items load on a given factor as it was observed in Figure 1. The factor loading matrix contains factor loadings and correlations of every item with each of the retained factors.

Figure 1. Steps and procedures in the factor analytical process, and fuzzy AHP.

(iii) Factor rotation and principal components rotate about the original variable’s axis. The goal of factor rotation is to assist with factor interpretation by achieving a structurally simple matrix compared with the original factor loading matrix. In doing so, varimax with the Kaiser normalization method of orthogonal rotation is applied to preserve the principal components due to its simplicity in the interpretation of the factors, and also a new transformation matrix is formed.

(iv) Factor interpretation, the final step of factor analysis, is the assignment of meaning and name to each factor according to the domain each represents.

The overall framework and steps of the newly proposed methodology were carried out by integration of factor analysis, and fuzzy AHP is shown in Figure 1.

3.4.2. Fuzzy analytical hierarchy process

Analytic hierarchical process (AHP) is a powerful method to solve complex decision problems. Any complex problem can be decomposed into several sub-problems using AHP in terms of hierarchical levels where each level represents a set of variables or attributes relative to each sub-problem (G. G. Ayalew et al., 2022). The fuzzy AHP method is based on the assessment of relative weight of input variables, and group variables by comparing them in order to determine their ratio and hierarchical ranking dependent on the importance of each factor.

The fuzzy AHP method is an aggregated technique for multi-criteria problems that was developed based on Saaty’s (Basahel & Taylan, 2016a). The analytical hierarchy process is primarily applied to decision problems in uncertain situations with multiple criteria. Hummel et al. (2014) highlighted that in the analytical hierarchy process, fuzzy set theory and fuzzy operation were introduced to overcome the subjectivity, inaccuracy, and fuzziness produced when making decisions. Fuzzy set theory is one of the most common methods used in multi-criteria decision-making studies.

The AHP can be stated as a decision-making and estimation method which gives the percentage distribution of decision points according to factors affecting the decision that is used if there is a defined decision hierarchy (Stief et al., 2018). AHP that uses both the linguistic assessments and numerical values for an alternative selection problem having a multi-level hierarchical structure will be represented. It is impossible to achieve accurate results or assessments if the data is immeasurable. In other words, crisp data or a single numerical value is insufficient to model real-life problems. This is due to the fact that human evaluations are usually vague and cannot be expressed with an exact value called fuzzy (G. G. Ayalew et al., 2022).

A fuzzy set can be defined mathematically by a membership function, which assigns each element x in the universe of discourse X a real number in the interval [0, 1] (Russo & Camanho, 2015). Fuzzy truth represents membership in vaguely defined sets, and variables over these sets are called fuzzy variables. A triangular fuzzy number is a special class of fuzzy numbers whose membership is defined by three real numbers, expressed as (l, m, u). The triangular fuzzy numbers are represented as follows in Figure 2. According to Huang and Ho (2013), the fuzzy set can be defined as follows:

(1) $\mathit{\mu }\mathit{A}\left(\mathit{x}\right)=\mathit{x}-\mathit{l}/\mathit{m}-\mathit{l},\mathit{l}\le \mathit{x}\le \mathit{mu}-\mathit{x}/\mathit{u}-\mathit{m},\mathit{m}\le \mathit{x}\le \mathit{u}\text{}0,\mathit{Otherwise}\text{ .}$(1)

Figure 2. Fuzzy triangular number $\tilde{\mathbf{A}}$.

A linguistic variable is a variable that applies words or sentences in artificial language to describe the degree of value. The concept of linguistic variable is useful in dealing with situations which are too complicated to be stated in quantitative values. In this study, triangular fuzzy numbers are used to represent the value (Basahel & Taylan, 2016a).

where l and $\mathrm{u}$ mean the lower and upper bounds of the fuzzy number $\tilde{\mathrm{A}}$, and m is the middle value for ${\tilde{\mathrm{A}}}_1$, the TFN can be denoted by $\tilde{\mathrm{A}}=\left(\mathrm{l},\mathrm{m},\mathrm{u}\right)$.

The operational laws of TFN ${\tilde{\mathrm{A}}}_1$ = (l₁, m₁, u₁) and ${\tilde{\mathrm{A}}}_2$ = (l₂, m₂, u₂) are presented as following equations (Sun, 2010).

Addition of the fuzzy number ⊕.

(2) ${\tilde{\mathrm{A}}}_1\oplus {\tilde{\mathrm{A}}}_2=({\mathrm{l}}_1+{\mathrm{l}}_2,{\mathrm{m}}_1+{\mathrm{m}}_2,{\mathrm{u}}_1+{\mathrm{u}}_2)$(2)

Multiplication of the fuzzy number ⊗

(3) ${\tilde{\text{A}}}_1\otimes {\tilde{\text{A}}}_2=({\text{l}}_{\text{\hspace{0.05em}1}}\text{*}{\text{l}}_{\text{\hspace{0.05em}2}},{\text{m}}_{\text{\hspace{0.05em}1}}\text{*}{\text{m}}_{\text{\hspace{0.05em}2},}{\text{u}}_{\text{\hspace{0.05em}1}}\text{*}{\text{u}}_{\text{\hspace{0.05em}2}}),\text{for}{\text{l}}_{\text{\hspace{0.05em}1}}{\mathit{ }}_{,}{\text{l}}_{\text{\hspace{0.05em}2}}\text{0};{\text{m}}_{\text{\hspace{0.05em}1}},{\text{m}}_{\text{\hspace{0.05em}2}}\text{ 0};{\text{u}}_{\text{\hspace{0.05em}1}},{\text{u}}_{\text{\hspace{0.05em}2}}\text{ 0}$(3)

Subtraction of the fuzzy number $\mathit{\theta }$

(4) ${\tilde{\text{A}}}_1\mathit{\theta }{\tilde{\text{A}}}_2=\left({\text{l}}_{\text{\hspace{0.05em}1}}{\mathit{ }}_{\text{\hspace{0.05em}--}}{\text{l}}_{\text{\hspace{0.05em}2}},{\text{m}}_{\text{\hspace{0.05em}1}}{\mathit{ }}_{\text{\hspace{0.05em}-}}{\text{m}}_{\text{\hspace{0.05em}2}},{\text{u}}_{\text{\hspace{0.05em}1}}{\mathit{ }}_{\text{\hspace{0.05em}-}}{\text{u}}_{\text{\hspace{0.05em}2}}\right)$(4)

Division of a fuzzy number $\mathrm{\varnothing }$

(5) ${\tilde{\text{A}}}_1\varnothing {\tilde{\text{A}}}_2=\left({\text{l}}_{\text{\hspace{0.05em}1}}{\mathit{ }}_{\text{\hspace{0.05em}/}}{\text{m}}_{\text{\hspace{0.05em}2}},{\text{m}}_{\text{\hspace{0.05em}1}}{\mathit{ }}_{\text{\hspace{0.05em}/}}{\text{m}}_{\text{\hspace{0.05em}2}},{\text{u}}_{\text{\hspace{0.05em}1}}{\mathit{ }}_{\text{\hspace{0.05em}/}}{\text{l}}_{\text{\hspace{0.05em}2}}\right),\text{for}{\text{l}}_{\text{\hspace{0.05em}1}}{\mathit{ }}_{,}{\text{l}}_{\text{\hspace{0.05em}2}}\text{0};{\text{m}}_{\text{\hspace{0.05em}1}},{\text{m}}_{\text{\hspace{0.05em}2}}\text{0};{\text{u}}_{\text{\hspace{0.05em}1}},{\text{u}}_{\text{\hspace{0.05em}2}}\text{0}$(5)

Reciprocal of the fuzzy number

(6) ${\tilde{\text{A}}}^{-1}=\left({\text{l}}_{\text{1}},{\text{m}}_{\text{1}},{\text{u}}_{\text{1}}\right){)}^{\text{-1}}\text{=}({\text{1/u}}_{\text{1}}{\mathit{ }}_{,}{\text{1/m}}_{\text{1}},{\text{1/l}}_{\text{1}}),\text{for}{\text{l}}_{\text{1}}{\mathit{ }}_{,}{\text{l}}_{\text{2}}\text{0};{\text{m}}_{\text{1}},{\text{m}}_{\text{2}}\text{0};{\text{u}}_{\text{1}},{\text{u}}_{\text{2}}\text{0}$(6)

The analytic hierarchy process (AHP) is a multi-criteria method of analysis which is a powerful method to solve complex decision problems. Any complex problem can be decomposed into several sub-problems using AHP in terms of hierarchical levels where each level represents a set of criteria or attributes relative to each sub-problem (Sun, 2010). The following steps were used to implement the fuzzy analytic hierarchical process technique (Kumar et al., 2013).

3.4.2.1. Break down decision problems

In this step, a decision hierarchy is constructed by breaking a general problem into individual criteria. The factors based on the feedback from the questionnaire are shown in Figure 3 in the form of a hierarchical diagram. The structures of the analytic hierarchy process were established by identifying 6 variable groups and 23 input variables (Tahvili et al., 2015).

Figure 3. Hierarchical structure of fuzzy AHP model.

3.4.2.2. Construct a hierarchical Model

The hierarchical structure of the factors affecting the performance of building construction project includes three levels. The top of the hierarchy is the overall objective or goal, the middle nodes are the relevant attributes (variable group), and the last level is the decision input variables or sub-factors of the decision problem.

The methodology of the current study was carried out according to the research aim, and the relationships among main variables and input variables are determined and reflected in the hierarchical model as shown in Figure 3 as follows.

3.4.2.3. Questionnaire survey and building pairwise comparison matrices with fuzzy scales

Based on the established hierarchical structure, a questionnaire was designed to sort out the importance levels of the factors that were identified, and the input variables were converted into a triangular fuzzy scale to reflect the expert’s rating of importance. To compare the importance of each input variable, it was categorized into nine linguistic terms as shown in Table 2. The linguistic scale in the fuzzy AHP method was used to categorize the variables (V) based on their importance.

The use of ratings enables the decision experts to analyze each variable with respect to other variables for their subsequent ranking relative to each other (Annals, 2017). By using triangular fuzzy numbers, via pair-wise comparison, the fuzzy judgment matrix Ă is constructed as Equation 7(7) $\frac{⌣}{\text{A}}=\left\{{\tilde{a}}_{\mathit{ij}}\right\}=a_{12}.\dots \dots \dots .\dots \mathrm{..}{a}_{1\mathit{n}}{a}_{21}1\dots .\dots .\dots a_{2\mathit{n}}{a}_{31}{a}_{32}1\dots \mathrm{..}{a}_{3\mathit{n}}\mathrm{..}\dots .\dots .\dots .\dots \mathrm{..}\dots .\dots .\dots \dots a_{\mathit{n}1}{a}_{\mathit{n}2}\dots \mathrm{..}\dots .\dots .1\frac{⌣}{\text{A}}\text{=}\left\{{\tilde{a}}_{\mathit{ij}}\right\}=a_{12}\dots \mathrm{.....}{a}_{1\mathit{n}}/a_{12}-a_{12}1\dots \mathrm{....}{a}_{2\mathit{n}}11a_{13}-0.7\mathit{ex}{a}_{13}1/1a_{23}-a_{23}\dots a_{3\mathit{n}}\mathrm{........}\dots \mathrm{........}\dots /1a_{1\mathit{n}}-a_{1\mathit{n}}1/1a_{2\mathit{n}}-a_{2\mathit{n}}\dots \mathrm{......1}$(7) . The pair-wise comparisons are members of the set: {1/9, 1/8, 1/7, 1/6, 1/5, 1/4, 1/3,1/2, 1,2,3,4,5,6,7,8,9}. The pair-wise comparison judgments are represented by fuzzy triangular numbers denoted by ã_ij = (l_ij, m_ij, u_ij) and n(n-1)/2 judgments are required for each comparison group for a level to construct a positive fuzzy reciprocal comparison matrix Ă = {ã_ij} (Basahel & Taylan, 2016b). The matrix is expressed as follows.

(7) $\frac{⌣}{\text{A}}=\left\{{\tilde{a}}_{\mathit{ij}}\right\}=a_{12}.\dots \dots \dots .\dots \mathrm{..}{a}_{1\mathit{n}}{a}_{21}1\dots .\dots .\dots a_{2\mathit{n}}{a}_{31}{a}_{32}1\dots \mathrm{..}{a}_{3\mathit{n}}\mathrm{..}\dots .\dots .\dots .\dots \mathrm{..}\dots .\dots .\dots \dots a_{\mathit{n}1}{a}_{\mathit{n}2}\dots \mathrm{..}\dots .\dots .1\frac{⌣}{\text{A}}\text{=}\left\{{\tilde{a}}_{\mathit{ij}}\right\}=a_{12}\dots \mathrm{.....}{a}_{1\mathit{n}}/a_{12}-a_{12}1\dots \mathrm{....}{a}_{2\mathit{n}}11a_{13}-0.7\mathit{ex}{a}_{13}1/1a_{23}-a_{23}\dots a_{3\mathit{n}}\mathrm{........}\dots \mathrm{........}\dots /1a_{1\mathit{n}}-a_{1\mathit{n}}1/1a_{2\mathit{n}}-a_{2\mathit{n}}\dots \mathrm{......1}$(7)

Where,

${\tilde{a}}_{\mathit{ij}}=\left\{\begin{array}{c}\tilde{1},\tilde{2},\tilde{3},\tilde{4},\tilde{5},\tilde{6},\tilde{7},\tilde{8},\tilde{9}\mathrm{Variable}\mathrm{i}\mathrm{is}\mathrm{relative}\mathrm{importance}\mathrm{to}\mathrm{variable}\mathrm{j}\\ 1\mathrm{\forall }\mathrm{i}=\mathrm{j}\mathrm{variable}\mathrm{i}\mathrm{is}\mathrm{equal}\mathrm{importance}\mathrm{to}\mathrm{variable}\mathrm{j}\\ {\tilde{1}}^{-1},{\tilde{2}}^{-1},{\tilde{3}}^{-1},{\tilde{4}}^{-1},{\tilde{5}}^{-1},{\tilde{6}}^{-1},{\tilde{7}}^{-1},{\tilde{8}}^{-1},{\tilde{9}}^{-1}\mathrm{Variable}\mathrm{i}\mathrm{is}\mathrm{relative}\mathrm{less}\mathrm{importance}\mathrm{to}\mathrm{variable}\mathrm{j}\end{array}\right.$

3.4.2.4. The consistency test

The consistency ratio (CR) is used to directly estimate the consistency of pairwise comparison to assess the reliability and credibility of the questionnaire (Huang & Ho, 2013). When the value of CR ≤10% indicates a good level of consistency in the comparative judgments represented in that matrix, and it is acceptable. CR value greater than 10% has resulted in the inconsistency of judgments within that matrix (See Figure 4). This suggests that the evaluation process needs to be reviewed and repeated to improve the consistency of the questionnaire. As proposed by Abay et al. (2019), the consistency ratio is calculated as per the following steps.

1. Calculate the eigenvector and largest eigenvalue (λmax) for each matrix of order n. Principal eigenvalue (λmax) is obtained from the summation of products between each element of the eigenvector and the sum of columns of the reciprocal matrix.

   Figure 4. The membership values of linguistic variables.

   

2. Compute the consistency index and consistency ratio for each matrix of order “n” by the formula (Vahidnia et al., 2008).

   (8) $\mathrm{CI}=\frac{\mathit{\lambda }\max -\mathrm{n}}{\left(\mathrm{n}-1\right)}\mathit{\hspace{0.17em}}$(8)
   (9) $\mathrm{CR}=\frac{\mathrm{CI}}{\mathrm{RI}}\mathit{\hspace{0.17em}}$(9)

where RI is a random consistency index obtained from a large number of simulation runs and varies depending upon the number of variables as shown in Table 3.

n= number of variables

3.4.2.5. Transforming fuzzy scales into triangular fuzzy numbers

The fuzzy nine-scale linguistic scaling to weigh the significance of each variable was transformed into triangular fuzzy numbers, as shown in Table 4.

3.4.2.6. The importance determination

 

i) The experts’ decision

To aggregate the decision of experts, the fuzzy geometric mean was the most common operation used in fuzzy multiple criteria decision-making. The following formula can be used to transfer expert opinions into triangular fuzzy numbers (Fu et al., 2020).

(10) ${\mathrm{a}}_{\mathrm{ij}}=({{\mathrm{a}}_{\mathrm{ij}}}^1\otimes {{\mathrm{a}}_{\mathrm{ij}}}^2\otimes {{\mathrm{a}}_{\mathrm{ij}}}^3\dots \dots .\otimes {\mathrm{a}}_{\mathrm{ij}}{{\mathit{ }}^{\mathrm{N}})}^{1/\mathrm{n}}$(10)

where $a_{\mathit{ij}}$ is the integrated triangle fuzzy number by N experts.

$a_{\mathit{ij}}$^k is the i-th to the j-th factor pair comparison by expert k.

⊗ is the symbol of matrix multiplication.

$a_{\mathit{ij}}$ = (a_ij¹⊗ a_ij ²⊗ a_ij ³ … … … … … .⊗ a_ij¹⁰) ^1/10

$a_{\mathit{ij}}$ = (a_ij¹⊗ a_ij ²⊗ a_ij ³ … … … … … .⊗ a_ij¹⁰) ^1/10

ii) Calculating fuzzy geometric mean and fuzzy weight of variables

To find the results of the fuzzy geometric mean and fuzzy weight, the fuzzy analytical operation can be done by using MS Excel for the calculation of importance coefficient is calculated using formula presented below.

(11) ${\mathrm{r}}_{\mathrm{i}}=({{\mathrm{a}}_{\mathrm{i}1}}^1\otimes {{\mathrm{a}}_{\mathrm{i}2}}^2\otimes {{\mathrm{a}}_{\mathrm{i}3}}^3\otimes \dots .\otimes {{\mathrm{a}}_{\mathrm{in})}}^{1/\mathrm{n}}$(11)

(12) $\mathrm{wi}={\mathrm{r}}_{\mathrm{i}}\otimes ({\mathrm{r}}_1\oplus {\mathrm{r}}_2\oplus {\mathrm{r}}_3\oplus \dots \dots \dots \oplus {{\mathrm{r}}_{\mathrm{n})}}^{-1}$(12)

where a_in is the fuzzy comparison value of factors i to factors n.

r_i is the geometric mean of fuzzy comparison value of factors i to each factor.

w_i is the fuzzy weight of the i-^th variables and can be indicated by a triangular fuzzy number.

⊕ is the symbol of a matrix plus

Hence, w_i = (Lwi, Mwi, Uwi). Lwi, Mwi, and Uwi stand for the lower, middle, and upper values of the fuzzy weight of the i-^th variables, respectively.

3.4.2.7. Weight priority by defuzzification

Defuzzification is used to renew the fuzzy number into a single crisp value (Fu et al., 2020). The center of the area converts a fuzzy weight into a non-fuzzy value and has been extensively applied in defuzzification by the following equation (Wang et al., 2019).

(13) $\mathrm{BN}{\mathrm{P}}_{\mathrm{i}=}\left[\left(\mathrm{U}{\mathrm{w}}_{\mathrm{i}}-\mathrm{L}{\mathrm{w}}_{\mathrm{i}}\right)+\left(\mathrm{M}{\mathrm{w}}_{\mathrm{i}}-\mathrm{L}{\mathrm{w}}_{\mathrm{i}}\right)\right]/3+\mathrm{L}{\mathrm{w}}_{\mathrm{i}}$(13)

i) Calculate the normalized local weights of variables.

After calculating the non-fuzzy value, it can perform normalization to obtain the local weights for input variables. The sum of local weights of the group variables and input variables in the same hierarchy was 1.00.

(14) $\mathrm{BN}{\mathrm{P}}_{\mathrm{w}1}=\mathrm{BN}{\mathrm{P}}_1/(\mathrm{BN}{\mathrm{P}}_1+\mathrm{BN}{\mathrm{P}}_2\dots .+\mathrm{BN}{\mathrm{P}}_{\mathrm{n}})$(14)

ii) Calculate the global weights of all input variables.

According to Kil et al. (2016), the value of the global weight equaled the value of the local weight within each variable group multiplied by the value of the local weight within each input variable. The sum of global weights was also 1.00. The ranking was arranged according to the order of the global weight.

(15) $\mathit{Global}\mathit{weight}=\mathit{value}\mathit{of}\mathit{the}\mathit{local}\mathit{weight}\mathit{within}\mathit{each}\mathit{variable}\mathit{group}\times \mathit{value}\mathit{of}\mathit{local}\mathit{weight}\mathit{within}\mathit{each}\mathit{input}\mathit{variable}$(15)

4. Case study: Numerical analysis

4.1. Demographic analysis

The characteristics of the respondents participating in the survey are summarized in Table 5. As shown in Table 5, 19.12% were construction engineer, 16.18% were project manager, 13.24% were office engineer, 25.00% were site engineer, 10.29% were supervisor, 7.35% were resident engineer, and 8.82% were senior engineer. The educational level of the respondents indicates that 89.70% had MSc Degree and 10.29% had PhD. The results of the study presented in the table show that 76.47% of the respondents were male and 23.53% were female. Hence, the respondents were not gender biased, and the sampling technique ensured inclusion of all members of the population being sampled for the study. Regarding the working experience, 45.59% of the respondents have between 5 and 10 years of working experience, 35.29% of the respondents have between 11 and 15 years of working experience, and 19.12% of the respondents have greater than 15 years of working experience.

4.2. Factor analysis results

At this stage, the input variables affecting the performance of building construction projects are identified and defined through an intensive literature review and expert interview for the required numerical analysis. In carrying out the exploratory factor analysis technique the results of the factor structure of the scale were presented in descriptive statistics as shown in Table 6. The result of factor analysis and the reliability analysis on pilot items was executed to test the reliability of the preliminary questionnaire set.

4.2.1. Descriptive statistics

The descriptive statistics, including the means, standard deviations, minimums, and maximums of the six proposed factors that affect the performance of the construction projects, are presented in Table 6. The results revealed that project-related factor had a significant factor in the building construction projects (M = 3.049), construction parties (M = 2.926), resource management related (M = 2.918), project condition (M = 2.908), economic condition (M = 2.879), and design and documentation (M = 2.685).

The minimum value was the same in all six factors which is one, and the maximum values were five for project condition, design, and documentation. In addition, the results supported the variables as normally distributed based on the degrees of skewness and kurtosis because both were less than the absolute value of one.

4.2.2. Factor analysis for validity

To test the reliability of the data set, Cronbach’s α coefficient method was used. Cronbach’s α coefficient for the data set is more than 0.7, which is a recommended threshold value. Therefore, the data set is considered reliable. In addition, KMO and Bartlett’s test were used as preliminary analyses to check the suitability of the data set collected through the questionnaire surveys for running the factor analysis (Jugessur, 2022). Consequently, the KMO measure verified the sampling adequacy for the analysis; KMO is 0.676, which is well above the acceptable limit of 0.5. According to Yu and Richardson (2015), Bartlett’s test of sphericity should be significant (the value of Sig. should be less than 0.05). For these data, Bartlett’s test is highly significant since the significance (p) is 0.000, indicating that there exist some relationships between the variables. Therefore, the results of KMO and Bartlett’s test revealed that factor analysis is appropriate (See Table 7).

Table 7. The results of KMO and Bartlett’s test

KMO and Bartlett’s Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy.		.676
Bartlett’s Test of Sphericity	Approx. Chi-Square	46.441
	Df	15
	Sig.	.000

An exploratory factor analysis was conducted on the 36 items with a Varimax rotation using SPSS. In this study, the six group variables were used to determine the pattern of the structure in the 36 items of factors that affect building construction projects performance that creates a scree plot.

4.3. Preliminary six-factor structure

An initial analysis was run to obtain eigenvalues for each factor in the data. KMO measure verified the sampling adequacy for the analysis, KMO = .676 which is above Kaiser’s recommended threshold of 0.6. Bartlett’s test of sphericity p < 0.000 indicated that correlations between items were sufficiently large for EFA. Six factors had eigenvalues greater than one, as the scree plot clearly illustrates in Figure 5. The initial 36-item structure explained 79.892% of the variance in the pattern of relationships among the items.

Figure 5. Scree plot of factor analysis.

The scree plot graph shows the eigenvalue against the factor number. The elbow shape appears well defined, indicating that some impacting factors have been extracted. The eigenvalue was set to the value of six factors that were extracted having an impact. The table of total variance explained as shown in Table 8 displays the % variance of the factors. As can be seen in Figure 5, if a line is drawn at the set eigenvalue of 2, it can be deduced that below the line indicates weak impacting factors. Above the line are strongly impacting factors.

4.4. Final six-factor structure

Based on the results of the initial exploratory factor analysis, the final six-factor structure in this study was composed of 23 items by deleting 13 items that were cross-loaded on two or more factors. As presented in Table 10, the results of the rotated component matrix for each component matrix were five items for component 1, five items for component 2, three items for component 3, four items for component 4, three items for component 5, and three items for component 6.

Table 10. The items and final six-factor structure of the factors affecting the performance of building projects after factor reduction procedures

Rotated component matrix^a
Factors	Component
	1	2	3	4	5	6
Method of Construction		.782
Inaccurate Quantity take off		.859
Delay in Payment		.827
Management condition		.820
Poor Site Management		.751
Size of Project				.695
Change in Scope, and design				.824
Complexity of Project				.843
Number of Floors				.745
Equipment Failure at the Work Site	.837
Hiring of equipment	.876
Efficiency of equipment	.835
Shortage of Skilled Labor	.931
Productivity of Manpower	.907
Lack of Experience					.963
Project Location					.964
Rework					.805
Financial Difficulties			.938
Market Fluctuation			.784
Interest Rate			.938
Changes in Design						.725
Unclear Specification						.943
Delays in Design Work						.928

Note: Extraction Method: Principal Component Analysis.

Rotation Method: Varimax with Kaiser Normalization.

a. Rotation converged in 6 iterations.

Finally, this 23-item structure was found to explain 79.430% of the variance in the pattern of relationships among the items as shown in Table 8. The percentages explained by each factor were 26.396% (construction parties), 16.677% (project related), 12.007% (resource management related), 9.390% (project condition), 8.205% (economic condition), and 6.756% (design and documentation), respectively. In this study, the three factors were highly correlated to each other as it is observed in Table 9. The factor correlation between economic conditions and construction parties was 0.337; the correlation between economic conditions and resource management was 0.307; the correlation between design and documentation and project-related factors was 0.343.

4.4.1. Item analysis for reliability

Reliability tests were conducted at the beginning of the analysis of the results to check the reliability of the data before they were analyzed (Heale & Twycross, 2015). To measure the reliability of the questionnaire, it can use Cronbach’s alpha. Cronbach’s alpha shows a degree of internal consistency. The reliability coefficient normally ranges between 0 and 1. According to (Rooshdi et al., 2018) when Cronbach’s Alpha is more than 0.7 there is high internal consistency for the data set. The closer the alpha (α) is to 1, the greater the internal consistency of items in the instrument is assumed. The formula that determines alpha is fairly simple and makes use of the items (variables), k, in the scale and the average of the inter-item correlations, r. The conceptual formula of Cronbach’s Alpha is defined by (Wahed, 2018).

(16) $\mathit{\alpha }=\mathit{\hspace{0.17em}}\frac{\mathrm{Kr}}{1+\left(\mathrm{K}-1\right)\mathrm{r}}$(16)

An item analysis was conducted to test the reliability of each factor of the factors affecting the performance of building construction projects. According to Heale and Twycross (2015), satisfactory internal consistency ranges from 0.7 to 0.9. All six factors on this scale had a high rating for reliability. The value of Cronbach’s alpha is 0.866 for the entire questionnaire. The Cronbach’s alpha for construction parties, project-related, resource management-related, project condition, economic condition, and design and documentation were 0.875, 0.822, 0.938, 0.916, 0.931, and 0.882, respectively.

4.5. Fuzzy analytic hierarchy process results

The variable groups and input variables resulting from the factor analysis (see Table 10) are converted to the hierarchical structure to transform these grouped input variables as the schematic structure depicted (see Figure 3). The hierarchical structure of the decision problem is shown in Figure 3, and its ultimate goal is to identify the most significant factor that affects the performance of building projects based on the relative importance weight of the input variables. Thus, to define and compute the importance weights of variable groups and individual input variables under each variable group, the fuzzy AHP method was applied. The method of calculating priority weights of the different decision variables is discussed furthe, and this finding can be justified as it is in line with the findings of previous studies (G. G. Ayalew et al., 2022).

1. First, the decision-making team consisted of 10 decision experts that fill the judgment matrix.

2. The relative importance of the input variables, then the pairwise comparison matrices of variables will be obtained. It can apply the fuzzy numbers defined in Table 2.

3. Calculating the consistency index and consistency ratio.

When the value of CR ≤ 10% indicates a good level of consistency in the comparative judgments represented in that matrix, it is acceptable. CR value greater than 10% has resulted in the inconsistency of judgments within that matrix.

The consistency index (C.I.) and the consistency ratio (C.R.) for a comparison matrix are calculated using the following equations (Vahidnia et al., 2008).

(17) $\mathrm{CI}=\frac{\mathit{\lambda }\max -\mathrm{n}}{\left(\mathrm{n}-1\right)}\mathrm{CR}=\frac{\mathrm{CI}}{\mathrm{RI}}$(17)

As an example, a fuzzy pairwise comparison matrix of the variable group.

$\mathit{\lambda }\max$ = 6.349, n = 6, RI _(n = 6) = 1.24. Therefore, the CI can be determined by using Equation 9(9) $\mathrm{CR}=\frac{\mathrm{CI}}{\mathrm{RI}}\mathit{\hspace{0.17em}}$(9) , and the CR by using Equation (9)(9) $\mathrm{CR}=\frac{\mathrm{CI}}{\mathrm{RI}}\mathit{\hspace{0.17em}}$(9) , of the pairwise judgmental matrix can be calculated as follows: CI = 0.069.

(18) $\mathit{CR}=\frac{0.069}{1.24}\mathit{\hspace{0.17em}}$(18)

Thus, the judgmental matrix is acceptable. The consistency ratios of all other matrices are less than 10%. Thus, all the judgments are consistent.

d) Computing the elements of aggregated pairwise comparison matrix of the variables in the hierarchy were computed through combining the collected data from all experts by using a geometric mean method, that is,

a_ij = (a_ij¹⊗ a_ij²⊗ a_ij³ … … … … .⊗ a_ij^N)^1/n

a_ij = (a_ij¹⊗ a_ij²⊗ a_ij³ … … … … .⊗ a_ij^N)^1/10

As a sample calculation, the aggregated fuzzy pairwise comparison values for the variable group are shown in the following matrix Ă.

Table

	V1	V2	V3	V4	V5	V6
V1	(1.000,1.000,1.000)	(0.867,1.021,1.173)	(0.920, 1.025, 1.139)	(0.951, 1.037, 1.139)	(1.136, 1.227, 1.307)	(1.113, 1.198,1.273)
V2	(0.852,0.980,1.153)	(1.000,1.000,1.000)	(0.945, 1.050, 1.163)	(1.000, 1.096, 1.193)	(1.099, 1.192, 1.280)	(1.063, 1.149,1.234)
V3	(0.887,0.982,1.091	(0.855,0.940,1.047)	(1.000, 1.000, 1.000)	(0.906, 1.050, 1.219)	(1.145, 1.256, 1.368)	(1.109, 1.219,1.336)
V4	(0.878,0.964,1.052)	(0.832,0.903,0.984)	(0.820,0.953, 1.104)	(1.000, 1.000, 1.000)	(1.093, 1.237, 1.390)	(1.089, 1.233,1.386)
V5	(0.765,0.815,0.880)	(0.781,0.839,0.910)	(0.731, 0.796, 0.873)	(0.717, 0.805, 0.909)	(1.000, 1.000, 1.000)	(1.056, 1.188,1.326)
V6	(0.786,0.835,0.899)	(0.810,0.870,0.941)	(0.749, 0.821, 0.902)	(0.722, 0.811, 0.919)	(0.754, 0.842, 0.947)	(1.000, 1.000,1.000)

4.5.1. The Importance determination

i. In order to determine the fuzzy geometric mean and fuzzy weights of group variable, the fuzzy geometric mean technique can be used in this paper by using the following equation (See Table 11).

r₁ = (a₁₁⊗ a₁₂ ⊗ a₁₃ ⊗ a₁₄ ⊗ a₁₅⊗ a₁₆)^1/6

r₁= (1 ⊗ 0.451 ⊗ 1.401⊗1.306 ⊗ 2.08) ^1/5, (1 ⊗ 0.689⊗ 1.922 ⊗1.73⊗2.552) ^1/5, (1 ⊗1.089 ⊗ 2.448 ⊗ 2.191⊗ 3.042) ^1/5)

r₁= (1 ⊗ 0.867⊗ 0.920 ⊗ 0.951⊗ 1.136 ⊗ 1.113) ^1/6, (1 ⊗ 1.021⊗ 1.025 ⊗ 1.037⊗ 1.227 ⊗ 1.198) ^1/6, (1 ⊗ 1.173⊗ 1.139 ⊗ 1.139 ⊗ 1.307⊗ 1.273) ^1/6

r₁ = (0.976, 1.049, 1.121)

Likewise, the remaining r_i values are obtained as follows:

r₂ = (0.980, 1.051, 1.127)

r₃ = (0.961, 1.033, 1.113)

r₄ = (0.932, 1.004, 1.080)

r₅ = (0.824, 0.871, 0.927)

r₆ = (0.799, 0.861, 0.934)

To determine the fuzzy weight of the group variable, Equation (12)(12) $\mathrm{wi}={\mathrm{r}}_{\mathrm{i}}\otimes ({\mathrm{r}}_1\oplus {\mathrm{r}}_2\oplus {\mathrm{r}}_3\oplus \dots \dots \dots \oplus {{\mathrm{r}}_{\mathrm{n})}}^{-1}$(12) is applied.

w₁= r₁⊗ (r₁⊕ r₂⊕ r₃⊕ r₄⊕ r₅⊕ r₆)⁻¹

w₁= (0.976, 1.049, 1.121) ⊗ (1/(1.121 + … . + 0.934), 1/(1.049 + … + 0.861), 1/(0.976+ … . + 0.799))

w₁= (0.155, 0.179, 0.205)

Similarly, the remaining fuzzy weights w_i values are:

W₂ = 0.155, 0.179, 0.206

W₃ = 0.152, 0.176, 0.204

W₄ = 0.148, 0.171, 0.197

W₅ = 0.131, 0.148, 0.169

W₆ = 0.127, 0.147, 0.171

ii. To apply the COA method to compute the BNP and BNP_w values of the fuzzy weights of the group variable: To take the BNP value of the weight of group variables as an example, the calculation process is as follows:

BNP_i = [(Uw_i- Lw_i) + (Mw_i- Lw_i)]/3 +Lw_i

BNP_i = [(0.205–0.155) + (0.179–0.155)]/3 + 0.155

BNP_i = 0.179

BNP_w1 = BNP₁/(BNP₁ + BNP₂ … . … . + BNP_n)

BNP_w1 = 0.179/(0.179 + 0.180 + 0.177 + 0.172 + 0.150 + 0.148) = 0.209

Similarly, the BNP value of the remaining group variables and input variables can be obtained in a similar computational procedure (see Tables 12–18).

Table 11. Reliability of factors affecting the project performance

Group variables	Cronbach’s alpha	Number of items
Construction parties	0.875	5
Project related	0.822	4
Resource management Related	0.938	5
Project Condition	0.916	3
Economic condition	0.931	3
Design and Documentation	0.882	3
Total Reliability Statistics of Group Variables	0.866	23

Table 12. The fuzzy weights of the variable groups with respect to the goal

Variable group (Vi)	Fuzzy weights (wi)	BNP	Normalized local weights (BNPw)	Rank
Construction parties	(0.155, 0.179, 0.205)	0.179	0.209	2
Project related	(0.155, 0.179, 0.206)	0.180	0.210	1
Resource management Related	(0.152, 0.176, 0.204)	0.177	0.207	3
Project Condition	(0.148, 0.171, 0.197)	0.172	0.200	4
Economic condition	(0.131, 0.148, 0.169)	0.150	0.174	5
Design and Documentation	(0.127, 0.147, 0.171)	0.148	0.172	6

Table 13. The fuzzy weights of the input variables with respect to construction parties

Input variable (Vij)	Fuzzy weights (wi)	BNP	Normalized local weights (BNPw)	Rank
Method of Construction	(0.189, 0.222, 0.258)	0.223	0.221	2
Inaccurate Quantity take off	(0.209, 0.247, 0.288)	0.248	0.246	1
Delay in Payment	(0.154, 0.180, 0.213)	0.182	0.181	4
Management condition	(0.163, 0.192, 0.225)	0.193	0.192	3
Poor Site Management	(0.139, 0.160, 0.186)	0.162	0.160	5

Table 14. The fuzzy weights of the input variables with respect to project related

Input variable (Vij)	Fuzzy weights (wi)	BNP	Normalized local weights (BNPw)	Rank
Size of Project	(0.262, 0.302, 0.347)	0.303	0.301	1
Complexity of Project	(0.257, 0.295, 0.336)	0.296	0.294	2
Change in Scope, and design	(0.167, 0.191, 0.221)	0.193	0.192	4
Number of Floors	(0.185, 0.212, 0.245)	0.214	0.213	3

Table 15. The fuzzy weights of the input variables with respect to resource management related

Input variable (Vij)	Fuzzy weights (wi)	BNP	Normalized local weights (BNPw)	Rank
Equipment Failure at the Work Site	(0.184, 0.216, 0.255)	0.218	0.216	1
Hiring of equipment	(0.167, 0.199, 0.237)	0.201	0.199	3
Efficiency of equipment	(0.181, 0.217, 0.257)	0.218	0.216	2
Shortage of Skilled Labor	(0.151, 0.177, 0.211)	0.180	0.178	5
Productivity of Manpower	(0.162, 0.191, 0.224)	0.192	0.190	4

Table 16. The fuzzy weights of the input variables with respect to project condition

Input variable (Vij)	Fuzzy weights (wi)	BNP	Normalized local weights (BNPw)	Rank
Project Location	(0.381, 0.413, 0.447)	0.414	0.413	1
Lack of Experience	(0.261, 0.283, 0.310)	0.285	0.284	3
Rework	(0.279, 0.303, 0.329)	0.304	0.303	2

Table 17. The fuzzy weights of the input variables with respect to economic condition

Input variable (Vij)	Fuzzy weights (wi)	BNP	Normalized local weights (BNPw)	Rank
Financial Difficulties	(0.334, 0.377, 0.428)	0.38	0.378	2
Market Fluctuation	(0.337, 0.383, 0.431)	0.384	0.382	1
Interest Rate	(0.217, 0.240, 0.268)	0.242	0.241	3

Table 18. The Fuzzy weights of the input variables with respect to Design and Documentation

Input variable (Vij)	Fuzzy weights(wi)	BNP	Normalized local weights (BNPw)	Rank
Changes in Design	(0.362, 0.394, 0.427)	0.394	0.393	1
Unclear Specification	(0.261, 0.286, 0.315)	0.287	0.287	3
Delays in Design Work	(0.291, 0.320, 0.352)	0.321	0.32	2

The normalized weights of the input variable placed at the third level in the hierarchy structure can be presented in Table 18.

5. Discussions, managerial implications, and limitations

5.1. Discussion of findings

This study explored the appropriate set of input variables to conduct factor analysis and fuzzy AHP methods. The hybrid methodology of factor analysis and fuzzy AHP is used to supplement the overall significant factors that affect building construction projects. In this study, a set of 23 factors were considered as factors which affect the performance of building construction projects. The results of exploratory factor analysis revealed that the six-factor structure of the factors affecting the performance of building construction projects were 79.430% of the variance in the pattern of relationships among the items. All six factors had high reliabilities with the value of Cronbach’s Alpha greater than 0.7. Twenty-three items remained in the final questionnaire after deleting 13 which cross-loaded on multiple factors. As a result, the six-factor structure of the factors affecting the performance of building construction projects has been confirmed through this study. Thus, after conducting the validity and reliability of the data, the results of the fuzzy analytic hierarchy process are discussed as follows. The decision experts weighted the value of the group variables and input variables as it is presented in Table 19.

Table 19. Weighted values and rankings considered by decision experts

Variable group	Local weight	Input variable (Vij)	Local weights	Global weights	Ranking by category	Overall ranking
Construction parties	0.209	Method of Construction	0.223	0.047	2	12
		Inaccurate Quantity take off	0.248	0.052	1	10
		Delay in Payment	0.182	0.038	4	21
		Management condition	0.193	0.040	3	19
		Poor Site Management	0.162	0.034	5	23
Project related	0.210	Size of Project	0.303	0.064	1	5
		Complexity of Project	0.296	0.062	2	6
		Change in Scope, and design	0.193	0.041	4	18
		Number of Floors	0.214	0.045	3	15
Resource management Related	0.207	Equipment Failure at the Work Site	0.218	0.045	1	13
		Hiring of equipment	0.201	0.042	3	17
		Efficiency of equipment	0.218	0.045	1	13
		Shortage of Skilled Labor	0.180	0.037	5	22
		Productivity of Manpower	0.192	0.040	4	20
Project Condition	0.200	Project Location	0.414	0.083	1	1
		Lack of Experience	0.285	0.057	3	8
		Rework	0.304	0.061	2	7
Economic condition	0.174	Financial Difficulties	0.380	0.066	2	4
		Market Fluctuation	0.384	0.067	1	3
		Interest Rate	0.242	0.042	3	16
Design and Documentation	0.172	Changes in Design	0.394	0.068	1	2
		Unclear Specification	0.287	0.049	3	11
		Delays in Design Work	0.321	0.055	2	9

As it is observed in Table 19, the decision experts compared local weights in each group variable and ranked project-related factors (0.210), and construction parties (0.209) are identified as the first and second most significant factors that affect the performance of building construction projects. Thus, the decision experts believe that project-related, and construction parties should be viewed as the most significant factors which affect the performance of building construction projects. Osuizugbo and Okuntade (2020) found that project-related factor has significant factors that affect building construction projects also conducted another study that factors incorporated project-related factor were client-related, contractor-related, consultant-related, material-related, labor- and equipment-related, contract management-related, and external-related.

Moreover, Memon et al. (2012) also stated that the construction parties have a significant role in the construction project, that the project participants plan and coordinate construction activities, and must complete the project within the established time and budget. Resource management-related factor (0.207) is the next identified factor by the decision experts that affects the performance of building construction projects followed by project condition (0.200), and economic condition (0.174). Azim Eirgash (2019) proved that to achieve the performance of the projects’ project resource management, condition of the project, and economic condition are significant for achieving the greatest organizational value. Thus, the right resources are available at the right time for the right work.

Contrariwise, Design, and documentation (0.172) was the least significant factor that affects the performance of building construction projects. This result indicates that the perceptions of the decision experts are influenced by the priority weights of project-related factors and construction parties overshadow the level of effect in the building construction project. This does not indicate that design and documentation are not affecting the project performance, but have a relatively low significant effect as compared to the other factors.

Individually, further examining each input variable under their group variable, the highest and the most significant factor among the input variables for construction parties category factors was inaccurate quantity takeoff (0.248) followed by the method of construction (0.223), management condition (0.193), and delay in payment (0.182). On the contrary, poor site management (0.162) is the least important criterion. The greatest weighted value under project-related factors was the size of the project (0.303) followed by the complexity of the project (0.296), and the number of floors (0.214). On the contrary, changes in scope and design are the least significant factors in the perception of the decision experts.

Next, the equipment failure at the work site (0.218), the efficiency of equipment (0.218), and the hiring of equipment (0.201) were identified as the most significant factors affecting the building project’s performance by the decision experts. Contrariwise, productivity of manpower (0.192) and shortage of skilled labor (0.180) were identified as the least are the least factors.

Furthermore, the most significant factor of input variables under the project condition-related factor was project location (0.414), and rework (0.304). This finding can be supported by (Al-Zwainy et al., 2020), that the project location has a significant influence on the costs of the project. The differences are in labor costs, the availability of materials and equipment, delivery logistics, local regulations, and climate conditions. Contrariwise, lack of experience (0.285) is the least significant factor that affects the performance of public building construction projects identified by the decision experts.

Moreover, the most significant factors of input variables under the economic condition-related factor were market fluctuation (0.384) and financial difficulties (0.380). Contrariwise, the interest rate (0.242) is the least factor identified by the decision experts.

Finally, the results show that the value for changes in design (0.394) and delays in design work (0.321) were the most significant factors affecting the performance of building construction projects. Contrariwise, unclear specification (0.287) is the least factor identified by the decision experts. Aslam et al. (2019) supported that the phenomenon of cost overrun due to design changes is universal. Thus, almost every country is experiencing the unfavorable effect of design changes on the cost and time performances of projects.

According to these results, project location, changes in design, and market fluctuations were ranked as the first, the second, and the third factors affecting the project performance. This factor is important for the stakeholders because the location of project, design changes, and market fluctuation will be more costly to the stakeholders.

Overall, the input variable with the highest-ranked final weights among global weights were project location (0.083), changes in design (0.068), market fluctuation (0.067), financial difficulties (0.066), size of the project (0.064), and complexity of the project (0.062). This finding can be supported by Al-Zwainy et al. (2020) that constructing a facility in a locality is very different from constructing one in other areas, and project location has a significant influence on the costs of the project. Meharie, Gariy, Ndisya Mutuku, et al. (2019) also proved that the project size and complexity of the project have a highly significant effect on the performance of the project. He also proved that project size and project complexity were found to have the most significant effect on the performance of building construction projects. Regarding project size or complexity, it can be generalized that, the larger the project, the more expensive it will be.

On the contrary, the lowest ranked sub-factor was management condition (0.040), productivity of manpower (0.040), delay in payment (0.038), shortage of skilled labor (0.037), and poor site management (0.034). This does not mean that the identified factors are not significant factors yet even this was significant in the project performance, but it is relatively low as compared to the factors mentioned in above. This shows that management condition, productivity of manpower, delay in payment, shortage of skilled labor, and poor site management did not play a major role in affecting the project performance management by the perception of respondents.

This was mainly because the professionals in the building construction project absolved themselves of any responsibility for the management condition, manpower productivity, method of payment, and site management. This meant that the project participants to conduct timely approvals of payments, management conditions, and manage productivity and site management within the time they stipulated.

5.2. Managerial implications

The findings of this study provide some significant implications for identifying the factors affecting the performance of building construction projects. This study provides a way to examine the most significant factors using factor analysis and fuzzy AHP methods. It would also help us to understand how academicians conduct a study by integrating the applied methodologies.

This study provides us some significant implications for the policy-makers, project managers, and practitioners. The policy-makers in the industry need to have a clear mission and vision to formulate, implement, and evaluate their performance. The construction management professionals and customers should also strive to properly take market fluctuations into account when developing the quantity, size of the project, location of the project, and changes in design. Governments can use this result to revise and establish construction policies.

Regarding the project practitioners, the study provides knowledge on the significant factors that affect the performance of building construction projects, and also provides us for the practitioners to create plans, and specification accordingly. Additionally, this paper helps us to carry out an integrated methodology of factor analysis, and Fuzzy AHP methods. Moreover, carrying out this study helps us to conduct a study on the remedial solution for the identified factors as a future area of study. Furthermore, this study helps in identifying the significant factors that affect the performance of building construction projects. Hence, it helps us to create an appropriate solution for the identified factors in a future area of study and to carry out the construction projects according to plans and specifications.

5.3. Limitations

In the course of conducting this research, some knowledge was gained, which reveals the limitations of the techniques or methodologies in this research article. With this insight, it is worthwhile to outline additional methods that could be pursued in the future to extend the frontier of knowledge concerning factor analysis and fuzzy methods.

The first limitation of this study is related to the use of exploratory factor analysis as an analysis method. It is an advantageous statistical method used to examine the construct validity data. However, because EFA is not a sufficient tool to test the theoretical foundations of the instrument, a confirmatory factor analysis (CFA) should be conducted to further the knowledge in this area by integrating several types of fuzzy set theories to check the consistency of the results.

Second, during the factor identification process, in most cases, a large number of input variables were considered for identifying the factors, which requires high computational requirements. This makes the integrating model of factor analysis and fuzzy AHP more complex and time-consuming as far as the formulation of the historical database is concerned. This paper looks forward to investigating the factors affecting the performance of building construction projects.

6. Conclusions and remarks

The aim of this research is to construct an exploratory factor analysis and fuzzy AHP models to explore the factors that affect the performance of building projects. Project performance is a global concern measured in all projects. Numerous factors (i.e. construction parties, project related, resource management related, project condition, economic condition, and design and documentation) affect the performance of building projects. Neglecting these factors leads to a significant effect on project performance. Recognizing these factors is critical to the ability of the building project performance. In this study, factor analysis and fuzzy AHP were employed to analyze the results using SPSS software version 23 and Microsoft Excel, respectively. Factor analysis helps to eliminate inappropriate factors that decrease the standard of correlation, and the most important factors that affect the project performance were determined via fuzzy AHP model.

In light of the findings and discussions, the following conclusions are suggested:

• In this study, the authors should consider validating their factor analytical results using factor analysis approaches to enhance confidence in their findings, and it concluded that the correlation matrix shows the adequacy the questionnaire.

• The KMO measure verified the sampling adequacy for the analysis; KMO of .676 is above Kaiser’s recommended threshold of 0.6, and Bartlett’s test of 0.00 shows that factor analysis is appropriate. The extracted component represents the variables well. The six factors explained 79.430% of the information contained by the 23 items (variables).

• Then, after using factor analytical techniques, authors should use the validated questionnaire to develop the fuzzy AHP model to prioritize the newly developed questionnaires for the variables by decision experts in terms of their relative impotence, subjectivity, and uncertainty of human assessment are taken into account fuzzy set theory in a fuzzy environment.

• From the proposed method, fuzzy AHP helps to find out that project-related and construction parties can be identified as the most significant group variables that affect the performance of building construction projects.

• The study also concluded that the other top five factors that affect the performance of building construction projects were project location (0.083), changes in design (0.068), market fluctuation (0.067), financial difficulties (0.066), and size of the project (0.064).

• It was also revealed, among others, that the factors affecting project performance are complexity of project, rework, lack of experience, delays in design work, and inaccurate quantity take off.

• The result of this study motivates the authors to formulate recommendations to improve the performance of a construction project. One recommendation is the results of the current study confirm that the fuzzy AHP technique is a useful and simple method for evaluating MCDM regarding project performance factors.

• Moreover, this paper recommends that the project professionals and stakeholders must develop and implement ways to reduce the cost and the time factors of the construction projects to complete within budgeted cost and defined period.

• Finally, the findings of this study provide academicians and practitioners with insightful information on the factors affecting the performance of building construction projects by integrating factor analysis and fuzzy AHP to demonstrate its practicality would be the best area of exploration.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The source of data used to support the findings of this study is available from the corresponding author upon request.

Additional information

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Notes on contributors

Girmay Getawa Ayalew

Girmay Getawa Ayalew is a lecturer at the Woldia Institute of Technology, Woldia University, P.O. Box 400, Woldia, Ethiopia. He is a former lecturer at the Gondar Institute of Technology, University of Gondar, P.O. Box 196, Gondar, Ethiopia. He received his MSc degree in Construction Engineering and Management from the University of Gondar, Ethiopia. His research interest includes Fuzzy AHP, Performance Management, and Emerging Technologies in Construction.

Genet Melkamu Ayalew

Genet Melkamu Ayalew is a lecturer at the Gondar Institute of Technology, University of Gondar, P.O. Box 196, Gondar, Ethiopia. She received her MSc degree in Construction Engineering and Management from the University of Gondar, Ethiopia. Her research interest includes Performance Management, Fuzzy AHP, and Regression Modeling.

Meseret Getnet Meharie

Meseret Getnet Meharie is a senior lecturer at Adama Science and Technology University, Adama, Ethiopia. He is an Assistant professor at Adama Science and Technology University, Adama, Ethiopia. His research interest includes Cost Estimation of Construction Projects, Fuzzy AHP, and Machine Learning algorithms in predicting the cost of highway projects.