Mitigating Time and Cost Overruns in Construction Projects: A Questionnaire Study on Integrating Earned Value Management and Risk Management

Abstract


Introduction
Numerous countries worldwide, particularly during economic development, prioritize infrastructure growth as a key component of their development strategies.This infrastructure is vital for societal progress and is the foundation for economic growth [1].However, construction projects frequently face delays, with many exceeding their original timelines by more than 50%, resulting in substantial costs and prolonged project durations [2].These delays lead to lost business opportunities in various sectors and make projects unsustainable because of changing technological, environmental, and social factors [3].
Development projects spearheaded by governments, notable for their vast scale, often face significant construction delays.These delays deeply affect the country's economic landscape and societal cohesion.
The underlying reasons for these delays range from inefficient management and insufficient legal safeguards to financing problems and logistical obstacles.Such interruptions highlight fundamental planning difficulties and threaten the nation's overall resurgence and economic progress.Promptly addressing even minor project setbacks is essential for improving productivity, efficiency, and cost-efficiency, ensuring projects stay within budget [4].Previous research indicates that the reasons for schedule delays and cost overruns are interconnected.Delays adversely affect production planning and control, particularly in construction projects, irrespective of a country's socioeconomic status [5].
Successful project completion relies on thorough planning and scheduling, incorporating construction methods, materials, and practices.Risk management about project tasks is crucial, underlining the importance of precise risk identification and mitigation [6].Considering the numerous uncertainties, risk management is vital for the smooth functioning of construction supply chains [7].Enhanced risk management can mitigate the significant impacts of identified risk factors on time and cost efficiency.The preparation and identification of risk factors directly and indirectly influence risk management [8].On-time project completion is a crucial requirement in the construction industry, where factors causing delays frequently overlap with time and cost challenges [9].Efficient coordination and wise resource distribution are essential in reducing stakeholder conflicts and guaranteeing smooth project execution.Delays can also arise from design hurdles, complexities in materials acquisition, and issues related to suppliers [10].
To improve project control and forecasting accuracy, it is crucial to leverage the capabilities of earned value management (EVM) and earned duration management (EDM) [11].Properly addressing major risks in infrastructure projects requires using a quantitative risk measurement approach, which provides essential insights for effective risk management tactics [12].Additionally, understanding the significance of organizational critical success factors (CSFs) and integrating them with ethical practices is paramount for construction business owners and senior management [13].However, using limited resources such as materials, funds, and labor effectively should be driven by identifying critical success factors (CSFs) [14].Ultimately, emphasizing risk management is crucial for enhancing different aspects of construction projects [15].The primary objective of this study was to improve the cost and timeline of construction projects.To achieve this, the research examined risk management and earned value management using an online questionnaire for factor analysis and assessment.

Study Site and Participants
A structured data collection method, using a questionnaire survey, was executed in specific regions of Iran: East Azerbaijan, West Azerbaijan, and Tehran province, between October 5 and November 10, 2022, as shown in Figure 1.The survey's primary objective was to investigate various facets of project management, emphasizing risk management, cost evaluation, and time management in construction projects.The comprehensive survey included twenty questions centered on the nuances of risk management, five questions addressing cost-related issues, and another five questions focused on time management topics.Through these inquiries, the study aimed to glean insights, data, and feedback from project managers and expert engineers in Iran, facilitating in-depth analysis and evaluation of these pivotal areas of project management in the mentioned regions .

Procedure and Instrument
The questionnaire was designed using Google Forms, a free electronic tool provided by Google for voluntarily gathering information via the structured survey.In compliance with the General Data Protection Regulation (GDPR), no personal information, such as first and last names, phone numbers, or emails, was requested or stored [16].The questionnaires were segmented into three main sections.The first section focused on demographics, encompassing gender, age, and education.The second section addressed risk management; the third delved into earned value management.After completing the questionnaire and removing the restricted questionnaires, 278 questionnaires were analyzed.

Earned value management
EVM has been applied consistently since its introduction by the American Department of Defense in 1967.EVM is grounded in three primary metrics: earned value (EV), which denotes the planned cost for completed tasks; planned value (PV), representing the budgeted cost for scheduled tasks; and actual cost (AC), which indicates the expense of work that has been done [18].The metrics outlined in Table 1 below can be used to evaluate a project's performance in terms of its schedule and costs at a specific project stage, usually during a designated tracking period.Specifically, a schedule variance (SV) or SV(t) (>0) and a schedule performance index (SPI) or SPI(t) (>1) indicate if the project is lagging or progressing ahead of the planned schedule.Similarly, cost metrics, such as cost variance (CV) > 0 (>0) and cost performance index (CPI) (>1), show whether the project is over or under its allocated budget.If a project's schedule and cost variances are zero, it signifies that the project is on track and within the budget [19].

Information gathering tools
This study comprehensively explored the challenges of implementing the earned value management methodology and suggested practical solutions to address these challenges for effective methodology use.
The reliability of the questionnaire was validated with a Cronbach's alpha coefficient of 0.82, and its face and content validity were established.Chapter four presents the results of a confirmatory factor analysis undertaken to verify the construct validity of the questionnaire-the research questionnaire comprised thirty items rated on a 7-point Likert scale.Expert feedback was incorporated in the development of the questionnaire to ensure its relevance, with the scoring system detailed in Table 2. From the table presented above, it is clear that the questionnaire exhibits high reliability since all the values surpass the preferred threshold of 0.7.

Data Analysis
After collecting data to address the research questions and hypotheses, the researcher needs to use suitable statistical methods for analysis, aiming primarily to harness the data effectively to tackle research issues.This study's results were analyzed using both descriptive and inferential statistics.In the descriptive phase, multiple elements were evaluated, including frequency distribution, score distribution graphs, data skewness, and kurtosis.Central tendency measures like mean and standard deviation were also considered.Overall, SEM is instrumental in enabling researchers to evaluate intricate models and relationships across diverse data sets.

Stages of the structural equation modeling
Covariance structure analysis follows a systematic procedure that progresses sequentially.This methodological approach consists of several essential stages, each integral to the complete analysis.The process begins with the formulation of a sample statement.Then, the model estimation phase takes place.After that, the model's fit or accuracy is evaluated.Depending on the assessment outcomes, the model might be tailored or refined to match the gathered data more closely.Once an optimal model is discerned, the next stage is interpreting its results, extracting significant insights from the determined parameters and relationships.The culmination of the process is consolidating the outcomes and interpretations into a comprehensive research report.

Descriptive statistics
For the data analysis, we begin by presenting the descriptive statistics for the demographic variables of the study, which include gender, age, and education level.The presentation of the analytical statistics follows this.

Gender
The descriptive statistics in Table 3, detailing the frequency distribution for gender, indicate that of the study's respondents, 4% were women, while 96% were men.Based on Table 4, only 4% of respondents have less than five years of experience.Those with experience ranging from 5 to 10 years account for 17.2%.A significant 43.9% have experience spanning 10 to 15 years, and 34.9% have experience between 15 to 20 years.

Education
Descriptive statistics related to the education level variable are given in Table 5.

Level of Education Frequency Frequency Percentage
Bachelor's degree 116 41.7 Master's degree 162 58.3

Total 278 100
Regarding the level of education, the results show that 58.3% of the respondents have a master's degree, which is the most frequent among those with a bachelor's degree, which is 41.7%.• Values below 0.49 suggest avoiding factor analysis.

Descriptive statistics of research variables
• Values between 0.50 and 0.69 mean factor analysis is feasible, but data adjustments are advised .
• Values of 0.70 or above advocate for factor analysis without hesitations .
Bartlett's test evaluates if the correlation matrix is an identity matrix.A significance level for Bartlett's test below 5% denotes the aptness of factor analysis to determine the factor model, rejecting the unity assumption of the correlation matrix.This means significant interrelations exist between variables, enabling the extraction of inherent structures from the data.The risk management questionnaire has a KMO index of 0.96, while the EVM questionnaire stands at 0.92.
Both surpass the 0.7 threshold, signifying the sample size is apt for factor and path analysis using the structural equation model.Additionally, Bartlett's test has a significance value under 5%, confirming a meaningful interrelation between variables, making factor analysis fitting for deducing the structural model from the data.

Confirmatory factor analysis
Confirmatory factor analysis of risk management questionnaire: In Figures 2 and 3, confirmatory factor analysis is presented in two standard and non-standard formats.Furthermore, the root means square error of estimation (RMSEA) index, which gauges the model's fit, has a value of 0.37, reaffirming the model's fitting aptness.The data aligns within acceptable bounds, and the constructed model has been validated as apt for the study.
Confirmation factor analysis of earned value management questionnaire: In Figures 4 and 5, confirmatory factor analysis is presented in two standard and non-standard formats.Examining the final research models: In Figures 6 and 7, the final models of the project are presented in standard and non-standard formats.053 for the final model, further underscoring its fitting aptness.The gathered data sits within the acceptable parameters, and the constructed model has been affirmed apt for the study.These findings solidify the adequacy and validity of the suggested model, cementing its value for probing the interrelation between risk management, earned value management, and assorted performance metrics in civil engineering ventures.

Examining research assumptions
In order to evaluate the hypothesis that the integration of earned value management with risk management techniques significantly impacts the reduction of project time, Table 11 presents the path coefficients (or regression weights) of the variables in the research model along with their respective probability values.From the information in the preceding table, both earned value management and risk management positively influence construction project duration, showing a significant effect of 0.53 and an error level below 0.05.This confirms the research hypothesis that integrating earned value and risk management significantly reduces project time.
To evaluate the hypothesis that the combination of earned value management with risk management techniques notably impacts project cost reduction, Table 12 offers the path coefficients (or regression weights) of the variables in the research model, along with their respective probability values.From the information provided in the preceding table, it is apparent that earned value management and risk management exert a positive and significant influence on construction project costs, represented by a value of 0.60, with an error level below 0.05.This data corroborates and confirms the research hypothesis: integrating earned value and risk management significantly curtails project expenses.

Conclusion
The research study presented in this paper has explored and confirmed various findings that offer significant benefits in improving time and cost efficiencies in construction projects.The study examined a research model comprising Earned Value Management and Risk Management, finding it relatively acceptable and satisfactory for construction projects.This structural model is recommended for its ability to predict project performance, encompassing time, cost, and scope.
Further analysis of the data reveals that Earned Value Management, when used in conjunction with Risk Management, positively influences the timing of construction projects.This effect was quantified at a rate of 0.53, and the error level was less than 0.05.This significant finding supports the hypothesis tested in the research, highlighting the importance of integrating Earned Value Management and Risk Management to reduce the duration of construction projects.
Additionally, the study indicates that the cost of construction projects can be significantly reduced by applying this model.This conclusion was supported by employing regression weights and probability values.The impact of the model on project cost reduction was notable, with a rate of 0.60 and an error level of less than 0.05.The positive effect of Earned Value Management and Risk Management on reducing costs and improving project performance in the construction industry is thus emphasized, suggesting their effective application in future projects.

Figure 2 .Figure 3 .
Figure 2. Confirmatory factor analysis of risk management questionnaire in standard mode

Figure 4 .Figure 5 .
Figure 4. Confirmatory factor analysis of the earned value management questionnaire in standard mode

Figure 6 .Figure 7 .
Figure 6.The final model in standard mode

Table 2 .
Reliability of the research questionnaire using Cronbach's alpha

Variables Number of questions Cronbach's alpha Scoring method
the earned value approach was assessed through interviews with senior managers, supervisors, and selected experts-those demonstrating exceptional expertise advanced to the development and validation phase of the study questionnaire.The questionnaire was then administered to those qualified interviewees, with a group of up to eight individuals selected as the study's statistical sample.Finally, SPSS and AMOS software were used to analyze the gathered data .Structural equation modeling (SEM) is a robust statistical method designed to test hypotheses about relationships between observed and dependent variables.Also termed causal modeling, structural analysis of covariance, or simply SEM, it commonly employs the AMOS software for model evaluation and exploring intricate relationships.SEM encompasses two types of variables: latent (unseen) and observed (manifest).The observed variables serve as measures for the latent ones.Like other research types, the model's variables are classified into exogenous (independent) and endogenous (dependent) categories.SEM integrates two models: the structural model, delineating causal links between latent variables, and the measurement model, a facet of confirmatory factor analysis that leverages observable variables to gauge latent ones.
The structural equation modeling (SEM) technique was used to examine the research hypotheses.The analysis employed both AMOS and SPSS 27 software.The process unfolded: initially, expertise in using MSP software andThe goodness-of-fit indices offered by AMOS assess the model's compatibility with the observed data.The root mean square error is one of various goodness-of-fit tests, with values under 0.05 regarded as excellent, values between 0.05 and 0.08 as moderately acceptable, those between 0.08 and 0.1 as relatively poor, and above 0.1 as weak.Models with a root mean squared error of approximation over 0.10 are deemed notable.

Table 3 .
Frequency distribution of gender variableDescriptive statistics related to the work experience variables are given in Table4.

Table 4 .
Variable frequency table of work experience

Table 6
presents the descriptive statistics for each research variable, including the mean and standard deviation.

Table 6 .
Descriptive statistics of research variables

. Checking the adequacy of the sample
In inferential statistics, assessing the research data's adequacy is vital before undertaking factor analysis.This assessment determines if data can be distilled into a few underlying factors.For this evaluation, two tests are utilized: the Kaiser-Meyer-Olkin (KMO) index and Bartlett's test.The KMO index gauges the degree of partial correlation between variables.It highlights how much the shared variance of certain latent factors influences the variance of research variables.This index ranges between 0 to 1. Values nearing 1 indicate that the sample size is suitable for factor analysis.Based on the KMO value.

Table 7
lists the KMO value, Bartlett's statistic, degrees of freedom, and Bartlett's test significance.

Table 7 .
KMO index and Bartlett's test to check sample adequacy

Table 8 .
Indicators of the risk management model of construction projectsBased on the results shown in Table8, all fit indices align with the adequacy benchmarks, reflecting a satisfactory and reasonably acceptable alignment of the structural model.The RMSEA index, under 0.08, confirms an acceptable fit for the research model.Comparative fit indices, such as CFI, NFI, GFI, TLI, and IFI, underscore the model's acceptability.Values above 0.90 in these indices suggest a favorable model fit.

Table 9 .
Fitness indicators of the performance model of construction projects

Table 10 .
Fit indices of the final research models and IFI validate the model's suitability, with scores surpassing 0.90, marking a desirable fit.Additionally, the root means square error of estimation (RMSEA) index, a gauge for the model's fit, registers a value of 0.

Table 11 .
examination of the first hypothesis of the research

Table 12 .
examination of the second hypothesis of the research