Analyzing Cost Overrun Risks in Construction Projects: A Multi-Stakeholder Perspective Using Fuzzy Group Decision-Making and K-Means Clustering

Abstract

The current research investigates cost overrun factors in structural projects, focusing on the Middle East and North Africa (MENA) region using Egypt as a model. A systematic literature review was conducted, analyzing 405 research papers published between 2000 and 2024, from which 69 relevant papers were selected to identify 48 key factors contributing to cost overrun. Using K-means clustering, these factors were grouped into three clusters based on their probability and impact, which were classified for their risk levels. To ensure robust analysis, a survey was conducted to gather expert opinions, resulting in 369 valid responses from owners, contractors/subcontractors, and management firms/consultants. The fuzzy group decision-making approach (FGDMA) was conducted to rank all 48 factors, offering a detailed assessment of their relative importance. Based on these rankings, the top 20 factors were identified for analysis to examine variations in stakeholder priorities, capturing differences in perspectives among multi-stakeholders. Sensitivity analysis and Tornado charts explored the critical variations among stakeholders, with management firms/consultants and owners prioritizing design-related risks, such as inconsistencies and delays in approvals, while contractors/subcontractors focused more on material waste. This novel integration presents a structured approach for analyzing, prioritizing, and mitigating cost overrun risks, offering a comprehensive framework that provides practical insights for stakeholders to improve cost and risk management strategies.

1. Introduction

In fact, the construction sector is a key driver of economic expansion worldwide, particularly in the MENA region. Through its obvious share of the GDP and facilitation of urban development, the construction industry is essential to advancing economic growth and infrastructural improvement [1,2]. In the MENA region, the construction sector accounts for an estimated USD 187 billion, or 8.4% of the GDP [3], and is projected to expand by 2.2% in 2024 [4]. Generally, a project is considered satisfying if it is delivered on time, under or within budget, and with considerable involvement of the participants in the project outcomes. However, one of the main challenges preventing the successful completion of most construction sector projects is cost overrun, which has a direct influence on task progress that leads to the abandonment of projects [5].

Cost or budget overrun is defined as the variance between estimated and actual costs, priced consistently, evaluated in local currency, and contrasted with a constant baseline [6]. In construction projects, this variance upon completion and the amount agreed upon by the parties at contract signing is known as the overrun amount [7]. Internal variables, such as design changes, the cost estimating process, location, duration, and project size, complexity or type, or external ones, including weather conditions, inflation, and regulations, can cause budget increases [8].

The complexity of construction projects, characterized by a lack of data, uncertainties, and limited awareness of the scope, often results in inaccuracies in estimating and subsequent over-budget [9]. Additionally, the intricate nature of construction processes, coupled with external factors such as economic fluctuations and internal inefficiencies, makes cost control a critical concern for stakeholders. Moreover, project complexity is directly proportional to the level of cost overrun, with higher complexity introducing greater uncertainties and leading to more frequent budget escalations [10,11]. This increasing technical and legal complexity in construction projects is further exacerbated by a growing number of risks, which negatively impact project execution [12]. To address these challenges, risk management is used to maximize the probability and consequences of positive opportunities while minimizing those of adverse occurrences or threats, ensuring that project objectives are achieved and positivism is leveraged [13].

Egypt is a key player in the MENA region and offers a unique lens to study budget overrun in construction. In Egypt, the construction sector is vital to its economic growth, contributing significantly to GDP and employment. However, it faces persistent challenges, including inflation, currency instability, and resource allocation inefficiencies, which frequently lead to delays and budget escalations [14,15]. Despite the increasing complexity and scale of construction projects, existing research often fails to address the localized risks specific to Egypt. This lack of focus leaves stakeholders without adequate tools to anticipate and manage cost-related challenges effectively.

Therefore, this study identified three critical research gaps in cost management practices within construction in Egypt as a model. These gaps are as follows: (1) systematic identification and clustering of cost overrun factors in the construction industry, considering its unique economic, managerial, operational, and external challenges; (2) integration of advanced analytical methodologies to address the challenges inherent in data-driven and stakeholder-based evaluations; and (3) employing advanced tools to compare and prioritize risks based on stakeholder perspectives. By addressing these gaps, this study identified critical causes of cost overrun and developed a structured framework for prioritizing each stakeholder. The following are this study’s specific contributions:

• Provides new insights into managing over-budget risks in Egypt by systematically identifying 48 critical factors.
• Develops an innovative methodological framework that integrates K-means clustering and FGDMA, combining quantitative analysis for clustering risks with qualitative prioritization based on stakeholder perspectives.
• Employs sensitivity analysis and Tornado charts to highlight differences and alignments among the priorities of stakeholders, enabling the design of more effective and targeted risk mitigation strategies.

2. Research Background

One of the most significant risks in construction projects is cost over-budget, which is challenging to totally prevent due to its complexity and dynamic nature [16,17]. The main cause is the resource-intensive nature of the construction sector, which results in fluctuations in the cost of materials and equipment, shortages of resources, unforeseen expenses, and accidents during construction [18]. Furthermore, the primary reasons for budget increase shift over time (every ten years) [19]. Therefore, it is necessary to continuously update the understanding of cost overrun risks in order to avoid or mitigate them and manage complexity efficiently [20].

Numerous research has been conducted to identify the causes of cost overrun in construction projects. In the MENA region, the construction industry faces confrontations influenced by political, economic, and environmental factors [21]. Alsuliman [22] highlighted that awarding contracts based on the bidding price and prolonged project delays were the main reasons for cost increases in Saudi Arabia. In Iraq, Bekr [23] reported that security issues, regulatory changes, and delayed payments significantly impacted project costs. Additionally, Al-Hazim et al. [24] highlighted that Jordan’s high expenses were mostly caused by challenging terrain and unfavorable weather. Koushki et al. [25] conducted a study in Kuwait and concluded that change orders, owners’ financial constraints, and lack of experience were significant contributors to cost. Especially in Egypt, the construction sector faces unique challenges that reflect broader regional issues while highlighting local dynamics. According to Daoud et al. [26], scope modifications, payment delays, inadequate project planning, and a lack of skilled labor were the main causes of overrun in big construction projects. Yousri et al. [27] found 35 risk variables that make cost management even more difficult, such as material shortages, unrealistic estimates of costs, and funding problems. Similarly, Abdelalim et al. [28,29,30] highlighted additional risks that exacerbate budget overrun, emphasizing the complexity of cost management in the Egyptian construction industry.

Fuzzy logic was initially applied to construction risk assessment by Carr and Tah [31]. Numerous researchers have either utilized or modified fuzzy logic in various ways to enhance its applicability [32]. Modified fuzzy logic, along with combinations involving other methods such as the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), Analytical Network Process (ANP), and Analytical Hierarchy Process (AHP), has gained significant attention for risk assessment in infrastructure and power plant projects [32,33]. The Fuzzy AHP technique, a widely used Multi-Criteria Decision-Making (MCDM) method, can be burdensome due to the extensive pairwise comparisons required in complex projects [34]. ANP, a modified version of AHP, is particularly suited for capturing interdependencies between risks and their effects [32]. However, like AHP, ANP also demands numerous tedious pairwise comparisons. Fuzzy TOPSIS, another frequently used MCDM method for project selection and risk assessment in complex and uncertain environments, has limitations. It does not account for attribute interrelations, and maintaining consistent weighting for attributes can be challenging [35].

Novak [36] highlighted the general limitations of fuzzy methods, emphasizing the need to consider the ambiguity and uncertainty inherent in qualitative judgments from experts who may lack sufficient knowledge or relevant experience. Furthermore, developing proper aggregation techniques to quantify these expressions in order to rank risks and suitable linguistic expressions for risk evaluation is essential. Rather than evaluating risks based solely on probability and impact, applying fuzzy methods to risk assessment in healthcare projects can yield more reliable results by incorporating additional criteria such as the expertise and qualifications of experts and project characteristics [9,37,38]. This broader approach contributes to a more comprehensive and in-depth understanding of the risks associated with cost increases.

3. Research Methodology

This study first explored the relevant literature to identify possible risks of cost increase. A structured questionnaire was subsequently distributed throughout Egypt as a representative sample from the MENA region. K-means clustering was initially employed to group the identified factors into distinct clusters based on their probability and impact scores. These clusters then served as the basis for applying the fuzzy group decision-making approach (FGDMA), which refined the prioritization of risks through expert evaluations. Based on the results of the risk assessment, a comparative scenario was created to rank the most influential factors for owners, management firms/consultants, and contractors/subcontractors. The following subsections shown in Figure 1 provide a detailed step-by-step methodology.

4. Factor Identification

To explore the causes of cost increase in the construction industry, this paper used “cost overrun*”, “cost escalation”, “budget overrun*”, “budget exceedance”, “cost increase”, “budget inflation”, “cost acceleration”, “cost increment”, and “cost variation” as keywords in the ‘title’ section. Additionally, the ’topic’ section included keywords, such as “construction project*”, “construction management”, “construction industry”, and “construction sector”. In order to concentrate on identifying relevant factors, this paper uses “factors”, “causes”, “drivers”, “variables”, and “reasons” in the ’topic’ section to search papers from 2000 to 2024 using the Web of Science and Scopus databases.

After gathering the pertinent literature, the researchers acquired and reviewed 405 English-language papers to make sure there were no invalid records. After a thorough peer review, 128 duplicate entries were removed. The selection procedure followed two important inclusion criteria that matched the objectives of this study. First, only peer-reviewed articles were included to ensure reliability and validity, as these undergo rigorous evaluation compared to conference papers or books. Second, studies that rated cost variables were the focus, making sure the chosen articles not only identified the factors but also evaluated their relative importance [39].

The final step selected 69 papers from the databases. From the final set of 69 papers, efforts were made to harmonize and standardize the terminology used to describe the factors, ensuring consistency and accuracy across the analysis. Finally,

Table 1. List of the main cost overrun factors in construction.

Code	Factor	Code	Factor
F01	Define the Scope	F25	Delayed Decision-Making by Owners
F02	Frequent Changes in Design	F26	Political Interference
F03	Increase in Additional Work Orders	F27	Contractual Claims and Disputes
F04	Project Specifications and Feasibility	F28	Bureaucratic and Unethical Behavior
F05	Labor Availability and Skills	F29	Lack of Communication between Parties
F06	Labor Productivity	F30	Lack of Coordination and Cooperation
F07	Contractor Experience	F31	Delay in Materials
F08	Subcontractor Experience and Performance	F32	Planning and Scheduling Issues
F09	Poor Site Management Supervision	F33	Cost of Labor
F10	Availability of Materials	F34	Cost of Equipment
F11	Equipment Suitability and Availability	F35	Market and Supply Conditions
F12	Resource Planning and Allocation	F36	Delays in the Design Stage
F13	Contractual Ambiguity and Contradiction	F37	Economic Challenges
F14	Bidding and Tendering Difficulties	F38	Currency Issues
F15	Cost of Materials	F39	Market Fluctuations
F16	Inflation	F40	Owner Interference
F17	Financial Impediments by Owner	F41	Force Majeure
F18	Delayed Payments by Owner for Contractors	F42	Unclear or Changes in Specification
F19	Financial Issues by Contractor	F43	Mistakes/Deficiency in BOQ
F20	Bad Execution Problems	F44	Cash Flow During Construction
F21	Unforeseeable Site and Soil Conditions	F45	Poor Monitoring/Controlling of Costs
F22	Project Estimation Problems	F46	Poor Project Management
F23	Design Inefficiencies	F47	Adverse Weather Conditions
F24	Wastage of Materials on Site	F48	HSE Management

summarizes 48 factors of construction project cost increase, and Figure A1 contains their details.

5. Questionnaire Design

The creation of the research questionnaire began with semi-structured interviews, where expert engineers discussed the 48 selected factors. A crucial stage in conducting the survey correctly was the pilot survey with expert engineers [40]. In this study, 15 expert engineers evaluated budget overrun factors identified from previous research to assess their applicability in Egyptian construction projects.

The surveys were separated into the following sections:

5.1. Section One: Demographic Information

This section included details about the participants and their companies, such as job titles, company size by using employees in their company, company type (owners, contractors/subcontractors, and management firms/consultants), level of education, and years of experience in construction projects. This is required to ensure that all the respondents involved in this study meet the participant criteria.

5.2. Section Two: Survey Questions

This section was a structured type of survey to collect ratings for the final refined cost factors. This section contains forty-eight (48) questions that illustrate the affecting factors. Measurement of the probability of occurrence frequency and impact of each factor was based using linguistic terms such as very low, low, medium, high, and very high. The corresponding numerical values of these linguistic terms are 1, 2, 3, 4, and 5, respectively, evaluating each factor based on both likelihood and consequence, specifically in terms of cost impact [41]. Each factor is analyzed twice in a row; the first is for the probability of the factor and the second is for the impact of the same factor.

6. Distribution and Collection of Questionnaire

6.1. Questionnaire Distribution

Online questionnaires have the benefit of saving time and money in carrying out research. The poll was distributed to Egyptian construction practitioners via LinkedIn and emails. The practitioners addressed in this survey were employees of the main stakeholders in the construction sector (i.e., owners, contractors/subcontractors, and management firms/consultants). The survey asked the participants to rank the factors from different perspectives. The questionnaires were distributed with a variety of experiences, levels of education, titles, and company sizes to obtain a clear understanding of the construction field. As a result, 369 valid questionnaires were successfully completed and collected.

6.2. Data Collected

This section presents the demographic information of the respondents. The survey questionnaires were collected from 37, 110, and 222 owners, management firms/consultants, and contractors/subcontractors, respectively, as presented in Figure 2. In terms of the highest level of education achieved by the participants, the results indicate that 79.7% held a bachelor’s degree, 17.3% held a master’s degree, and 3% held a doctorate, as shown in Figure 3.

The size of the participants’ companies based on the number of employees in their companies was divided into five categories, namely Micro (1–10 employees), Small (11–50 employees), Medium (51–250 employees), Large (251–1000 employees), and Enterprise (1000+ employees), and the number of participants related to these categories was 5, 15, 60, 85, and 204, respectively, as illustrated in Figure 4.

Furthermore, based on the questionnaire results, 86 of the respondents have 0–4 years of experience, 106 have 5–8 years of experience, 85 have 9–13 years of experience, 52 have 14–20 years of experience, and 40 have more than 20 years of experience in the construction field, as shown in Figure 5. Additionally, the respondents were chosen from a variety of sections/positions to obtain comprehensive responses to the cost factors associated with Egyptian construction projects, as illustrated in Figure 6.

7. K-Means Clustering

This section examines the K-means clustering algorithm, a vital data analytics tool known for its effectiveness and simplicity [21]. K-means clustering is a widely used and proven technique in clustering [42]. This method is particularly valuable for this research as it offers a unique perspective on the dynamics of the factors affecting budget overrun in projects.

To calculate the proper number of clusters (k), numerous methodologies including the Dunn index, Davies Bouldin index, Hubert statistic, elbow plot, score function, and silhouette plot have been devised [43]. The elbow plot approach, which is renowned for its dependability [44,45], was used in this investigation to determine the cluster count.

The k-means algorithm’s primary aim is to minimize the within-cluster sum-of-squares criteria, also known as cluster inertia. By showing distortion scores for a chosen number of clusters according to Equation (1), the elbow plot validates the choice of an appropriate number of clusters. The “elbow” point is the number of clusters at the point at which the WCSS value is not appreciably decreased. Notably, this research found that three clusters were the optimal number, as seen in Figure 7:

$$WCSS=\sum _{k=1}^K{\sum }_{i\in C_k}{‖x_i-{\mu }_k‖}^2$$

(1)

where K = the total number of clusters, $C_k$ = the set of data points in the $k^{th}$ cluster, $x_i$ = a data point in the cluster, ${\mu }_k$ = the centroid of the $k^{th}$ cluster, and ${‖x_i-{\mu }_k‖}^2$ = the squared Euclidean distance between the data point $x_i$ and the centroid ${\mu }_k$.

Figure 8 and Figure 9 provide a graphic representation of this study by listing the cost over-budget factors within each cluster and displaying the network model colors by cluster.

Table 2. K-means clustering analysis summary.

Cluster	Factor IDs	Characteristics	Probability Range	Impact Range	Number of Factors
1	F15, F16, F37, F38, F39	Highest probabilities and impacts (High Risk)	0.61–0.67	0.66–0.73	5
2	F01, F02, F03, F04, F07, F08, F09, F10, F12, F13, F14, F17, F18, F19, F20, F22, F23, F24, F25, F26, F27, F30, F31, F32, F34, F35, F36, F40, F42, F43, F44, F45, F46	Moderate probabilities and impacts (Medium Risk)	0.47–0.58	0.56–0.65	33
3	F05, F06, F11, F21, F28, F29, F33, F41, F47, F48	Lowest probabilities and impacts (Low Risk)	0.33–0.49	0.43–0.58	10

shows these results and provides a summary of each cluster’s characteristics.

Every cluster has a unique impact on risk, influencing how we should react to possible difficulties. The most significant threats are seen in Cluster 1, such as Inflation (F16) and Currency Issues (F38), which are factors that are very likely to occur and have a significant impact. These are the kinds of risks that demand immediate attention and careful planning because ignoring them could seriously disrupt operations. Cluster 2 feels like the middle ground; risks such as Project Estimation Problems (F22) and Poor Monitoring/Controlling Costs (F45) are not as urgent but still need consistent attention to avoid becoming bigger problems down the line. Lastly, factors in Cluster 3, while they are less likely to occur, having a backup plan is prudent since they might have a cascading impact if they do, such as Adverse Weather Conditions (F47) and Labor Availability and Skills (F05). These are not pressing issues right now, but they might be worth keeping an eye on in the long run. Each cluster shows us where to concentrate our attention, ensuring that we are taking care of the critical factors first while continuing to monitor the other factors.

8. Model Set and Analysis (Fuzzy Group Decision-Making Approach)

Evaluating the factors affecting complex construction projects is crucial because of the complexity involved in understanding the properties of the risks, handling response evaluations of risks and subjective biases, and the relevance of the selected factor evaluation technique. Among many other fuzzy and hybrid methods, a modified fuzzy group decision-making approach was applied in this study [46] because of its capacity to handle the subjective biases for evaluated factors and its suitability for inference systems using limited data sets [34,47,48]. This entails participants assessing risks using the most popular probability–impact (P–I) approach [38] and their ability to judge individual factors based on their “professional competence” [49] to evaluate individual factors to increase the reliability of decision-making [37].

The sequential procedure of the modified FGDMA [9,46] is as follows:

1.

The fuzzy triangular number (FTN) for the associated linguistic term is extracted following the scenarios and descriptions shown in

Table 3. Linguistic variables and the associated fuzzy numbers.

Level of Risk Probability or Impact	Fuzzy Triangular Number (FTN)	Defuzzified Number Range	Description
Very High	0.7, 0.9, 1	0.7 to <0.9	Extremely high probability of event occurrence, with catastrophic impact on cost.
High	0.5, 0.7, 0.9	0.5 to <0.7	High probability of event occurrence, with a significant impact on cost.
Medium	0.3, 0.5, 0.7	0.3 to <0.5	Moderate probability of event occurrence, with notable impact on cost.
Low	0.1, 0.3, 0.5	0.1 to <0.3	Low probability of event occurrence, with a minor impact on cost.
Very Low	0, 0.1, 0.3	0.025 to <0.1	Very low probability of event occurrence, with negligible impact on cost.

. Triangular fuzzy numbers, which provide a three-point estimate (e.g., 0.5, 0.7, and 0.9, which represents high severity) rather than an exact value. This three-point estimate for each factor allows the flexibility to carry out appropriate decision-making to manage cost-increase risks in the project.

2.

Using the FTN, the following Equation (2) provides a fuzzy decision matrix (FDM) for risk probability (p) or impact (i) of cost overrun for each risk factor (f):

$${FDM}_{p/i}^f=\left[\begin{array}{ccc}{L}_1& M_1& U_1\\ ..& ..& ..\\ L_n& M_n& U_n\end{array}\right]$$

(2)

where L = lower, M = medium, and U = upper values of a risk’s probability or impact cost, and n = means the number of participants evaluating the risks.

3.

Participants’ skills and the validity of their assessments might differ for a variety of reasons in a given situation; thus, they must be weighted accordingly. This is a function of their participant position (PP), experience years (EY), and education level (EL) [49]. The weighted judgment of each participant is included in the evaluation of crucial criteria to improve decision reliability [37]. Aboshady, Elbarkouky et al. [50] calculate the weight of each participant ($W_i^{lnd})$ as follows:

$${W_i^{lnd}=\left(W_{PP}+W_{EY}+W_{EL}\right)}_i$$

(3)

where $W_{PP}$, $W_{EY}$, and $W_{EL}$ = each participant’s weights for PP, EY, and EL, respectively. Then, to evaluate the participant’s weights, each criterion (i.e., $W_{PP}$, $W_{EY}$, and $W_{EL}$) is assumed to be equal. The global weight of a participant ($W_i^g$) is calculated as follows [51]:

$$W_i^g={\displaystyle \frac{W_i^{lnd}}{\sum _{i=1}^n{W}_i^{lnd}}} ; \sum _{i=1}^n{W}_i^g=1$$

(4)

According to Jung et al. [49], to meet the principle aggregated fuzzy score, the global weights of all participants must equal unity.

4.

A weighted FDM (WFDM) for each factor (f) is transformed from the FDM by the following equation:

$$\begin{gathered} W{FDM}_{{\displaystyle \frac{p}{i}}}^f=\left({FDM}_{{\displaystyle \frac{p}{i}}}^f\right)\times W_i^g \\ =\left[\begin{array}{ccc}{L}_1& M_1& U_1\\ ..& ..& ..\\ L_n& M_n& U_n\end{array}\right]\times \left[\begin{array}{c}{W}_1^g\\ ..\\ W_n^g\end{array}\right] \\ =\left[\begin{array}{ccc}{L}_1{W}_1^g& M_1{W}_1^g& U_1{W}_1^g\\ ..& ..& ..\\ L_n{W}_n^g& M_n{W}_n^g& U_n{W}_n^g\end{array}\right] \end{gathered}$$

(5)

5.

The fuzzy score ($FS$) for each cost factor is the sum of each column of the Equation (5) matrix.

$${FS}_{{\displaystyle \frac{p}{i}}}^f=\sum _{i=1}^n{L_{i}W}_i^g, \sum _{i=1}^n{M_{i}W}_i^g,\sum _{i=1}^n{U_{i}W}_i^g$$

(6)

6.

The fuzzy risk score, which is derived by likelihood and consequences on cost, can be calculated using the fuzzy synthetic evaluation approach [52] as follows:

$${\left({FRS}_f\right)}_{L,M,U}={\left(\sqrt{\left({FS}_p^f\right)\times \left({FS}_i^f\right)}\right)}_{L,M,U}$$

(7)

where ${FS}_p^f$ and ${FS}_i^f$ are fuzzy scores for probability and impact on cost for each factor, respectively. Despite traditional fuzzy if–then rules being typically utilized to draw conclusions about the significance of critical factors, they have been criticized for their incapacity to address subjective biases [36]; as a result, an alternate method proposed by Xu et al. [52] is employed.

7.

The level of each factor (i.e., low to very high) is defined by the defuzzification, which is calculated as follows [53]:

$${\left({FS}_x\right)}_{Deff.}={\displaystyle \frac{{\left({FS}_x\right)}_L+4\times {\left({FS}_x\right)}_M+{\left({FS}_x\right)}_U}{6}}$$

(8)

9. Fuzzy Results

After applying the FGDMA to the 48 selected cost overrun factors, the analysis provided a refined prioritization of risks by addressing biases and integrating stakeholder evaluations, building on K-means clustering, which grouped factors into risk categories by probability and impact. This integration ensured a comprehensive evaluation that combined objective clustering with practical stakeholder considerations, offering a more balanced and actionable prioritization of risks. This study highlights the overall cost-related risk scenario for Egypt’s construction projects, derived by analyzing 369 responses. Financial, management, and operational difficulties emerged as the most significant risks, while external influences played a comparatively minor role, emphasizing the significance of addressing systemic challenges within the industry.

The high-risk cost-related factors in projects are Inflation (F16), Currency Issues (F38), Cost of Materials (F15), Market Fluctuations (F39), and Economic Challenges (F37). Of these top five factors, there are highlights of the pervasive impact of financial challenges and economic instability on project budgets. Notably, those financial factors are also present in the top 20 high-risk factors, such as Cash Flow During Construction (F44), Financial Issues by Contractor (F19), and Financial Impediments by Owner (F17), as shown in

Table 4. FGDMA outcomes of cost overrun factors.

Code	Factor	Probability Score	Impact Score	Risk Score	Rank
F16	Inflation	0.672	0.735	0.703	1
F38	Currency Issues	0.659	0.719	0.688	2
F15	Cost of Materials	0.662	0.711	0.686	3
F39	Market Fluctuations	0.629	0.691	0.660	4
F37	Economic Challenges	0.608	0.664	0.636	5
F02	Frequent Changes in Design	0.582	0.649	0.615	6
F44	Cash Flow During Construction	0.562	0.644	0.602	7
F18	Delayed Payments by Owner for Contractors	0.562	0.633	0.596	8
F19	Financial Issues by Contractor	0.538	0.650	0.591	9
F46	Poor Project Management	0.532	0.650	0.588	10
F22	Project Estimation Problems	0.530	0.647	0.585	11
F31	Delay in Materials	0.549	0.622	0.585	12
F45	Poor Monitoring/Controlling of Costs	0.527	0.644	0.582	13
F03	Increase in Additional Work Orders	0.563	0.591	0.577	14
F36	Delays in the Design Stage	0.550	0.605	0.577	15
F17	Financial Impediments by Owner	0.525	0.633	0.577	16
F12	Resource Planning and Allocation	0.522	0.626	0.572	17
F08	Subcontractor Experience and Performance	0.533	0.613	0.572	18
F32	Planning and Scheduling Issues	0.533	0.598	0.564	19
F25	Delayed Decision-Making by Owners	0.516	0.617	0.564	20
F23	Design Inefficiencies	0.5311	0.5987	0.5639	21
F10	Availability of Materials	0.5200	0.6113	0.5638	22
F13	Contractual Ambiguity and Contradiction	0.5037	0.6165	0.5572	23
F07	Contractor Experience	0.5015	0.6172	0.5564	24
F26	Political Interference	0.5030	0.6113	0.5545	25
F27	Contractual Claims and Disputes	0.5179	0.5934	0.5544	26
F43	Mistakes/Deficiency in BOQ	0.5061	0.6048	0.5532	27
F09	Poor Site Management Supervision	0.5090	0.5918	0.5489	28
F20	Bad Execution Problems	0.4954	0.6046	0.5472	29
F14	Bidding and Tendering Difficulties	0.5008	0.5958	0.5463	30
F01	Define the Scope	0.4860	0.6118	0.5453	31
F42	Unclear or Changes in Specification	0.4976	0.5919	0.5427	32
F40	Owner Interference	0.5072	0.5682	0.5368	33
F35	Market and Supply Conditions	0.4980	0.5779	0.5365	34
F34	Cost of Equipment	0.5104	0.5636	0.5363	35
F24	Wastage of Materials on Site	0.5026	0.5699	0.5352	36
F04	Project Specifications and Feasibility	0.4691	0.6050	0.5327	37
F30	Lack of Coordination and Cooperation	0.4915	0.5737	0.5310	38
F06	Labor Productivity	0.4905	0.5427	0.5160	39
F21	Unforeseeable Site and Soil Conditions	0.4404	0.5845	0.5074	40
F05	Labor Availability and Skills	0.4690	0.5399	0.5032	41
F11	Equipment Suitability and Availability	0.4504	0.5391	0.4927	42
F33	Cost of Labor	0.4801	0.5031	0.4915	43
F29	Lack of Communication Between Parties	0.4576	0.5231	0.4892	44
F48	HSE Management	0.4586	0.5067	0.4821	45
F28	Bureaucratic and Unethical Behavior	0.4308	0.5175	0.4721	46
F41	Force Majeure	0.3332	0.5711	0.4363	47
F47	Adverse Weather Conditions	0.3530	0.4308	0.3900	48

. This highlights the vital role that stable economies and prudent financial management play in guaranteeing the successful completion of building projects.

Alongside financial risks, managerial challenges also emerged as critical among the high-risk factors contributing to cost increase. Key issues, like Frequent Changes in Design (F02), Poor Project Management (F46), and Inadequate Cost Monitoring and Control (F45), highlight the essential need for strong project management and well-structured planning, as outlined in Table 4. Additionally, operational challenges, including Delays in Material Delivery (F31), Execution Problems (F20), and Issues with Material Availability (F10), underscore the complexity of managing construction processes.

While external factors, such as Adverse Weather Conditions (F47) and Bureaucratic and Unethical Behavior (F28), are recognized, their impact is less significant compared to the more pressing systemic challenges. This analysis underscores the need for comprehensive risk management strategies that address the interconnected nature of these challenges to improve project resilience and outcomes. To effectively tackle these challenges and ensure successful and efficient project delivery, it is essential to adopt a holistic risk management approach that accounts for the combined effects of operational challenges, managerial weaknesses, and financial constraints.

10. Sensitivity Analysis

Sensitivity analysis is an important part of risk analysis that enables prioritizing critical risk factors and developing effective mitigation strategies. This study analyzed the risk score (RS) of the top 20 risks, considering different clusters of stakeholders, to observe the sensitivities of the cost-related risks from owner, contractor/subcontractor, and management firm/consultant perspectives in Egypt. The results from this analysis are detailed in Figure 10 and

Table 5. Ranking of the top 20 factors based on stakeholder perspectives.

ID	Factor	GE Score	Rank	CS Score	Rank	MC Score	Rank	OW Score	Rank
F16	Inflation	0.703	1	0.694	1	0.714	1	0.717	1
F38	Currency Issues	0.688	2	0.674	3	0.710	2	0.709	2
F15	Cost of Materials	0.686	3	0.683	2	0.695	3	0.676	3
F39	Market Fluctuations	0.660	4	0.648	4	0.679	4	0.667	5
F37	Economic Challenges	0.636	5	0.628	5	0.648	5	0.645	7
F02	Frequent Changes in Design	0.615	6	0.591	9	0.641	6	0.668	4
F44	Cash Flow During Construction	0.602	7	0.604	7	0.592	9	0.618	13
F18	Delayed Payments by Owner for Contractors	0.596	8	0.607	6	0.584	11	0.576	28
F19	Financial Issues by Contractor	0.591	9	0.576	13	0.607	7	0.627	10
F46	Poor Project Management	0.588	10	0.594	8	0.565	20	0.623	11
F22	Project Estimation Problems	0.585	11	0.591	10	0.570	17	0.599	18
F31	Delay in Materials	0.585	12	0.570	16	0.601	8	0.621	12
F45	Poor Monitoring/Controlling of Costs	0.582	13	0.582	11	0.566	19	0.629	9
F03	Increase in Additional Work Orders	0.577	14	0.575	14	0.576	15	0.590	22
F36	Delays in the Design Stage	0.577	15	0.562	20	0.581	12	0.648	6
F17	Financial Impediments by Owner	0.577	16	0.580	12	0.560	23	0.606	17
F12	Resource Planning and Allocation	0.572	17	0.565	18	0.580	13	0.586	23
F08	Subcontractor Experience and Performance	0.572	18	0.563	19	0.573	16	0.615	14
F32	Planning and Scheduling Issues	0.564	19	0.566	17	0.561	22	0.568	33
F25	Delayed Decision-Making by Owners	0.564	20	0.549	26	0.577	14	0.608	16
F23	Design Inefficiencies	0.564	21	0.548	28	0.568	18	0.639	8
F10	Availability of Materials	0.564	22	0.544	32	0.586	10	0.610	15
F13	Contractual Ambiguity and Contradiction	0.557	23	0.551	24	0.557	24	0.596	19
F07	Contractor Experience	0.556	24	0.547	30	0.563	21	0.592	20
F24	Wastage of Materials on Site	0.535	36	0.572	15	0.470	44	0.525	40

Note: GE means general for all responses, CS means contractor/subcontractor responses, MC means management firm/consultants, and OW means owner responses.

.

A Tornado chart is a very useful tool for explaining the relative importance of various risks. It is a visual depiction of the model’s sensitivity analysis. A plot of horizontal bars arranged in descending order represents the risk scores. When there is a positive connection, the bars move to the right; when there is a negative correlation, they move to the left.

11. Stakeholder Perspective Discussion

11.1. Contractor/Subcontractor vs. Owner Perspectives

The Tornado graphic in Figure 11 illustrates the disparities between owner and contractor/subcontractor perspectives of the top 20 cost overrun factors. Positive differences imply that contractors view the factor as being more critical, whilst negative differences show that owners view the factor as more important than contractors.

For instance, Design Inefficiencies (F23) shows the largest negative difference (−0.091), Delays in the Design Stage (F36) has a notable negative difference (−0.086), and Frequent Changes in Design (F02) has a large difference (−0.076), meaning that owners view design factors as significantly more important than contractors. This could reflect owners concerns about how the design stage might affect the project budget. Moderate differences, such as Availability of Materials (F10) (−0.066), Delay in Materials (F31) (−0.051), and Subcontractor Experience and Performance (F08) (−0.052), show that owners prioritize these issues more than contractors/subcontractors, possibly due to their direct impact on cost and schedule planning for the project.

On the other hand, positive differences like Wastage of Materials on Site (F24) have a positive difference (+0.047), implying that contractors are greatly affected by waste on site, leading to an increase in the planned budget. The differences for factor (F18) Delayed Payments by Owner for Contractors (+0.03) show the extent to which contractors are focused on receiving payments and that owners’ payment delays have a significant impact on the project.

There are also areas where contractors and owners are largely aligned. Factors like Inflation (F16) and Cost of Materials (F15) show minimal differences, indicating that both parties recognize the impact of Economic Challenges on the project. Similarly, Market Fluctuations (F39) and Economic Challenges (F37) exhibit small differences, reflecting a shared understanding of the broader financial risks involved in the project.

Overall, the differences between owners and contractors/subcontractors emphasize the need for improved communication and clearer alignment of priorities, particularly regarding design and financial risks. Ensuring transparency during the design phase can minimize frequent changes while adopting proactive financial planning and monitoring mechanisms can help contractors manage cost-related challenges. Bridging these gaps can enhance collaboration and reduce delays, ultimately supporting the successful delivery of projects.

11.2. Contractor/Subcontractor vs. Management Firm/Consultant Perspectives

The Tornado chart shown in Figure 12 compares the perspectives of contractors and consultants regarding the importance of various factors affecting the project. Negative differences indicate that consultants consider a factor more critical than contractors, while positive differences show the opposite.

The most significant positive difference is for Wastage of Materials on Site (F24) (+0.101), highlighting that contractors place far greater importance on managing material waste than consultants. This factor also ranked as the most significant for contractors in the comparison between contractors and owners, which confirms contractors’ consistent prioritization of resource optimization and cost minimization across different perspectives, which is critical from an operational perspective. Additionally, factors like Poor Project Management (F46) and Project Estimation Problems (F22), although less obvious, further emphasize contractors’ concerns with managerial efficiency to mitigate potential risks to lead project cost increments.

In contrast, several factors show negative differences, indicating higher importance for the consultant. For example, Frequent Changes in Design (F02) (−0.049) and Availability of Materials (F10) (−0.042) are prioritized more by the consultant. This reflects consultants’ concerns about how the design stage and resources management may impact the project cost. Interestingly, these priorities align closely with those of owners, as both consultants and owners place significant emphasis on those factors. Similarly, Currency Issues (F38) (−0.036) and Delayed Decision-Making by Owners (F25) (−0.028) also have negative differences, reflecting consultants’ focus on financial stability and decision efficiency, which are essential for maintaining project timelines.

Some factors show minimal differences negative or positive, such as Contractor Experience (F07), Poor Monitoring/Controlling of Costs (F45), Cost of Materials (F15), and Cash Flow During Construction (F44). These alignments mean that both parties share a similar understanding of their impact on the project and are likely to collaborate effectively in addressing them.

Overall, the differences between contractors/subcontractors and management firms/consultants highlight critical areas where alignment is necessary to mitigate cost overrun. Enhancing cost control mechanisms and optimizing resource allocation are essential for addressing material waste and improving project management practices. Similarly, leveraging advanced analytical tools can support consultants in providing data-driven insights that align with contractors’ priorities. Addressing these differences fosters better coordination and reduces the likelihood of disputes arising from budget overrun.

11.3. Owner vs. Management Firm/Consultant Perspectives

Figure 13 compares the priorities of owners and management firms/consultants regarding the top 20 project cost-related factors. Positive differences indicate that owners view the factor as more critical than management firms/consultants, while negative differences show that management firms/consultants assign higher importance.

The most significant positive difference is for Design Inefficiencies (F23) (+0.070) and Delays in the Design Stage (F36) (+0.067), both highlighting owners’ strong focus on the design phase. This analysis confirms owners’ prioritization of ensuring efficiency and timeliness in design, recognizing its critical impact on project timelines and costs. In addition to the design phase, owners also place notable importance on factors such as Subcontractor Experience and Performance (F08), Contractor Experience (F07), and Contractual Ambiguity and Contradiction (F13). These factors reflect owners’ concerns with ensuring that all parties involved in the project are competent, well-coordinated, and operating under clear and consistent contractual terms to minimize risks and enhance project performance.

In contrast, the consultants place slightly more emphasis on factors like Cost of Materials (F15) (−0.019) and Financial Issues by Contractor (F19) (−0.020). This suggests that the consultants are more focused on addressing specific financial risks and ensuring efficient resource allocation. Similarly, Currency Issues (F38) (−0.001) and Market Fluctuations (F39) (−0.008) also have negative differences, reflecting the consultants’ focus on financial stability and external risks, which are essential for maintaining project timelines.

Some factors show minimal differences, such as an Increase in Additional Work Orders (F03) (0.014) and Planning and Scheduling Issues (F32) (−0.007). These alignments suggest that both parties recognize the importance of managing additional work orders and addressing scheduling issues, reflecting a shared understanding of their role in maintaining project timelines and minimizing disruptions.

Overall, the relationship between owners and management firms/consultants highlights the need for enhanced collaboration to align strategic goals with operational execution. Owners benefit from clear and timely information provided by management firms, enabling better decision-making, while consultants should focus on delivering actionable insights and proactive risk management strategies. Strengthening this partnership can ensure smoother project execution, mitigate risks, and support the achievement of project objectives within budget and schedule.

12. Limitations and Future Directions

This study provides valuable insights into cost-related factors in construction projects, offering a general framework applicable across the industry. However, certain limitations should be acknowledged. First, the data for this study were collected from stakeholders in Egypt, which may limit the generalizability of the findings to regions with differing economic, regulatory, and operational contexts. Since this study reflects conditions specific to the data collection period, the priority of cost-related factors may also vary across regions and over time due to evolving industry dynamics. Future research could address these limitations by applying the framework to diverse regions and time periods, enabling cross-regional comparisons, and capturing changes in cost-increase priorities. Second, the analysis focuses on the perspectives of key stakeholders, including owners, contractors, and consultants, but does not consider other important parties, such as designers, regulatory authorities, and government entities. Expanding the stakeholder scope in future research could provide a more comprehensive understanding of diverse priorities and potential conflicts, ultimately contributing to more inclusive cost management strategies. Finally, while this study adopts a general approach appropriate for all construction projects, applying the framework to specific project types such as residential, commercial, or infrastructure could yield tailored insights for stakeholders managing risks unique to each project category. By addressing these limitations, future studies could improve the generalizability, relevance, and practical impact of the findings across various contexts and project types.

13. Conclusions

This study provides a detailed analysis of factors that affect cost overrun in construction projects across the MENA region using Egypt as a model by employing advanced analytical methodologies, including K-means clustering, the fuzzy group decision-making approach (FGDMA), sensitivity analysis, and Tornado charts.

A systematic literature review of 405 research papers from the Web of Science and Scopus databases, spanning the years 2000 to 2024, was conducted. This yielded a final selection of 69 relevant studies. From these, 48 distinct cost-related factors were identified and analyzed. Key factors include Inflation (F16), Currency Issues (F38), Cost of Materials (F15), Market Fluctuations (F39), and Economic Challenges (F37). These were assessed using a structured questionnaire completed by 369 construction practitioners, distributed among 37 owners, 110 project management firms/consultants, and 222 contractors/subcontractors in the region.

The K-means clustering analysis grouped these factors into three separate clusters based on their probability and impact scores: Cluster 1 (High Risk) includes Inflation (F16), Currency Issues (F38), Cost of Materials (F15), Market Fluctuations (F39), and Economic Challenges (F37). These factors exhibit the highest probabilities (0.61–0.67) and impacts (0.66–0.73), emphasizing their critical influence on cost increases. Cluster 2 (Medium Risk) encompasses 33 critical factors. These factors have moderate probabilities (0.47–0.58) and impacts (0.56–0.65). Cluster 3 (Low Risk) includes 10 factors with the lowest probabilities (0.33–0.49) and impacts (0.43–0.58).

The fuzzy group decision-making approach effectively addressed subjective biases, providing weighted evaluations of each factor based on participant expertise. The results highlighted those financial factors, such as Inflation and Currency Issues, pose the most important risks, emphasizing the need for economic stability and strong financial planning. Managerial factors, including Poor Project Management (F46) and Frequent Design Changes (F02), also play a critical role, while operational challenges, such as Material Delays (F31) and Execution Problems (F20), add further complexity.

The Tornado charts provided insights into stakeholder-specific perspectives, revealing alignment and discrepancies among owners, contractors/subcontractors, and management firms/consultants. For instance, owners prioritized design-related factors like Design Inefficiencies (F23) and Delays in the Design Stage (F36), while contractors emphasized operational issues such as Material Wastage (F24) and Delayed Payments (F18).

This paper offers a thorough analysis of cost-related issues in Egyptian construction projects, combining literature review, clustering techniques, and decision-making approaches. By comparing and ranking the identified challenges, this study highlights the most critical financial, managerial, and operational risks. This structured evaluation provides stakeholders with actionable insights to prioritize efforts and address the majority of impactful factors, contributing to improved cost control and project outcomes in a complex construction environment.