The structural equation model of risk factors influencing government irrigation project

Project management is the application of knowledge, skills, tools and various techniques to process project activities. Furthermore, to succeed in the proposal of a project, there are 10 pieces of project management knowledge. The 10-project management knowledge has consistent data integration, and project cost management is one important process. In addition, project cost management consists of four processes: cost management planning, cost estimation, budgeting and cost control. In planning such a working process, it requires well-prepared and well-managed cost management of a project, together with project scope and execution strategy development, schedule planning and development, quality planning, resource planning, risk management, procurement planning and stakeholder planning (Hollmann 2006; Al Shami 2018).

Project managers will recognise the classic systems methodology of input, process, output and feedback loop outlined above, which is vital to the effective control of a project; however, the risk is somehow different. It has to do with uncertainty, probability or unpredictability, and contingency planning according to the project management body of knowledge (PMBOK). Additionally, a project manager must be managing various risks, from the less complicated to the most complicated. Conflicts are common in construction projects. When each party has different motivations, attitudes, and opinions, it is possible to have conflicts. Therefore, project management must fix the problem as soon as possible; otherwise, other issues, such as an employee’s absence or resignation, may occur (Burke 2013). Risk management can reduce adverse effects in four ways: avoid, transfer, mitigate and accept (Lester 2007).

Irrigation is one of the crucial inputs to agricultural production, raising crop yields and increasing crop varieties in the region. The construction of irrigation schemes has also contributed to an increase in crop variety in the region. In consequence, the economic wellbeing of rural people is increased, and migration to big cities decreases. The overall level of spending on water-related infrastructure in developing countries amounts to about US$65 billion per year, with irrigation and drainage accounting for about US$25 billion (Briscoe 1999). The first irrigation scheme in Turkey was built in the Konya Plain by the Ottoman Empire between 1903 and 1913. Large-scale construction of irrigation and drainage schemes started with the foundation law numbered 6200 of the General Directorate of State Hydraulic Works (DSI). Classical open-channel irrigation schemes were built until 1950–1965. Since then, Western countries, especially Italy, have applied the flume (canalette), which is a type of prefabricated material. Sloped terrain requires a considerable number of drop-type constructions, needs to overcome water distribution problems during the operation phase, and needs close examination of low-pressure pipe schemes. In the 1980s, low-pressure concrete pipes were used. However, population growth in Turkey, together with the increase in rural to urban migration, brought an increase in water demand. As a result, in Turkey, since 1990, developing pipe production and water-saving tube technology, which allow usage of the optimum level of water resources, has started in high-pressure pipe schemes and sprinkler systems (Koç 2011).

Irrigation construction projects play a critical role in agricultural development by building infrastructure such as dams, canals, reservoirs and pipelines to manage and distribute water effectively. These projects are especially important in regions with inadequate rainfall, helping to enhance agricultural productivity, ensure food security, and support local economies. Additionally, irrigation projects are among the top three budget priorities in Thailand’s construction sector, as defined by the government. This prioritisation is due to the country’s predominantly agricultural economy, where a significant portion of the population consists of farmers. Additionally, most irrigation projects aim to mitigate flood risks and improve drainage infrastructure, which are critical concerns in Thailand. However, they often face significant challenges due to various risk factors that can disrupt progress and inflate costs (Abdaljader and Günal 2024).

Management-related risks are among the most impactful, including poor project planning, ineffective communication among stakeholders and delays in decision-making. These issues can lead to misaligned objectives and prolonged timelines. Financial risks are another critical concern, often stemming from budget constraints, delayed payments or inaccurate cost estimations, which can hinder progress and strain project resources. Material-related risks, such as shortages, delivery delays, and the use of substandard supplies, further exacerbate challenges during construction (Bahamid et al. 2020; Abdaljader and Günal 2024).

Labour and equipment issues also pose considerable risks. These include shortages of skilled workers, low productivity and equipment breakdowns or inefficiencies, all of which can disrupt workflows. Design risks, such as incomplete or unclear project plans, frequent changes or insufficient surveys conducted before construction, can lead to rework and delays. External risks, including adverse weather conditions, regulatory hurdles and shifts in government policies, add further complexity to these large-scale projects (McDermot et al. 2022; Abdaljader and Günal 2024).

Mitigating these risks requires comprehensive planning and effective collaboration among all stakeholders, including clients, contractors and consultants. This involves clear communication, accurate cost and resource estimations, timely procurement of materials, and robust risk assessment processes. By addressing these challenges proactively, irrigation construction projects can achieve their objectives, delivering sustainable water management solutions that benefit agriculture and local communities. Effective risk management is essential to ensuring that these projects are completed on time, within budget, and to the required standards (Abdaljader and Günal 2024; Adedokun and Egbelakin 2024).

An overview of risk factors in construction projects reveals several key issues identified in previous studies that stakeholders should be aware of. Many studies have categorised the group of risk factors that affect construction projects such as grouping of environmental, physical, logistic, legal, political, construction, design, financial and management (Bahamid et al. 2020). In the other hand, several researchers define eight components of risk factors: cost, construction knowledge, design, economic, estimator, resource, geographic and project-related (Ekung et al. 2021). Specifically in irrigation projects, five groups of risk factors identified through a relative importance index analysis include management, design, finance, materials, labor and equipment, and external factors.

Construction risk is the umbrella term for all possible losses that could occur during a construction project. Construction risk can be caused by various factors, including the environment, project delays, safety concerns and other challenges. Throughout the entire building process, risk is an unavoidable component, according to the PMBOK. Table 1 lists the 56 criteria that were compiled from the literature evaluation process and used in this investigation. Many risk factors are used in this study, and they focus only on risk factors that relate to construction projects, infrastructure projects and irrigation projects.

The conceptual framework in this study is a structural framework built from the relationship between risk factors affecting cost management, time management and quality management. The factors will be analysed before developing the structural model that could be identified by exploratory factor analysis, as shown in Figure 1. With assumption 1 (H1), the risk factors affecting construction irrigation project management are related to three main aspects of project management.

Fig. 1:

Conceptual framework in this study.

Since the entire population of the irrigation work group in Thailand is unknown, Cochran’s formula was used, calculated at a confidence level of 95%, a tolerance of 5%, and a proportion of the characteristic of interest in the population of 0.5. The required population size was 384 units (Burrows 1953), and the researcher allowed 50% (Chaitongrat et al. 2021), totalling 576 units in determining the sample size, which is consistent with the concept of Pallant (2020) that specified the acceptable size to be more than 150 samples and 460 responses and complete sets, which is a small proportion because most construction companies are not convenient to provide information. The response rate of the questionnaire (Respond Rate) of more than 20% is considered acceptable (Ehrlinger and Wöß 2022).

The sample for this research comprised 460 irrigation construction projects funded by government departments, classified as a type of public construction project under the Government Procurement and Supplies Management Act B.E. 2560 (2017), within the 2024 fiscal year. In addition, the representatives of each construction project were engineers who were experienced in construction management and supervision.

A questionnaire was used as a tool in the research. In addition, the questionnaire was divided into two parts: (1) participants’ general information and (2) issues about problems and conditions of the contract for architects and construction supervisors. Both parts were a closed-end questionnaire with five rating scales to evaluate the chance of having risk factors. Moreover, the research instrument employed in this study underwent rigorous validity and reliability testing to ensure its effectiveness. The validity test was conducted by three experts in construction management and government project supervision, following the methodology described by Kangpheng et al. (2016). As a result, this process identified and refined the final set of 56 factors used in the study. Additionally, the instrument’s reliability was evaluated through a try-out method involving 30 individuals who possessed similar characteristics to the sample population but were not included in the actual sample group. The tool achieved a Cronbach’s α value of 0.95, which significantly exceeds the standard reliability threshold of 0.75, as outlined by Vorakitkasemsakul (2011). Consequently, the questionnaire was deemed both valid and reliable, making it suitable for data collection from the study’s target sample.

The analysis of the general data from the questionnaire and the evaluation of the research instruments revealed that the sample group comprised 70.87% male and 29.13% female participants. Most respondents were aged between 31 years and 50 years (48.48%), followed by those <30 years (38.91%) and >50 years (12.61%). In terms of educational background, the largest proportion held a bachelor’s degree (42.83%), while 36.74% had completed a higher vocational certificate, 17.39% had a vocational certificate, and 3.04% held a master’s degree or higher.

Regarding work experience, 32.39% of respondents had <5 years of experience, 30.65% had 5–10 years, 31.52% had 11–30 years, and 5.43% had >30 years of experience. Most participants worked in the private sector (74.13%), while 25.87% were employed in government organisations. In terms of job position, 26.30% of respondents were engineers, 11.52% were officers, and 6.30% were managers. A significant portion, 51.96%, held other roles such as supervisors or subcontractors, while 3.91% were directors, as summarised in Table 2.

One limitation of this study is the private-sector sample bias, with 74.13% of the respondents representing private construction firms. This skewed distribution is primarily because private-sector organisations typically employ a larger workforce in construction projects compared with government agencies. In many construction projects, private-sector firms handle on-site management, subcontracting and execution, which require a higher number of personnel than governmental bodies that focus more on regulatory oversight and funding allocation (Ekung et al. 2021). Consequently, the findings of this study may not be fully generalised to government-led projects, where administrative and bureaucratic challenges may differ. Future research should incorporate a more balanced sample, ensuring greater representation from government stakeholders to provide a more comprehensive perspective on risk factors in construction projects. The mean and standard deviation values for each factor are presented in Table 3.

This study employs SEM to analyse the effects of 56 identified risk factors on three critical dependent variables: budget, duration and quality. SEM is a sophisticated statistical tool particularly suited for this research because it allows for the simultaneous examination of multiple dependent variables, enabling a holistic understanding of the complex interrelationships within construction projects. Additionally, SEM supports the integration of latent variables that are unobservable constructs such as ‘management effectiveness’ or ‘resource availability’, which are derived from measurable indicators. This feature enhances the depth of analysis by capturing nuanced aspects of risk dynamics. Moreover, SEM is capable of modelling both direct and indirect effects, providing insights into how risk factors influence project outcomes through intermediary pathways. Another significant advantage of SEM is its ability to account for measurement errors in the observed variables, thereby increasing the precision and reliability of the results. Through goodness-of-fit indices, SEM ensures that the theoretical model aligns with empirical data, reinforcing the validity of the findings. By leveraging these capabilities, this study provides a robust framework for identifying critical risk factors and their impacts, offering valuable insights for project managers to mitigate risks and enhance the success of construction projects.

The initial SEM shown in the Figure 2 highlights the hypothesised relationships among multiple risk factors, risk management and project outcomes (duration, budget and quality). Six primary risk factors were considered: inefficiency in legal administration, inefficiency in construction management, inefficiency of contractors, inefficiency of design in the construction phase, discovery of antiquities and special work conditions. These factors were expected to contribute to overall project risk, which influences risk management strategies and, subsequently, project outcomes.

Fig. 2:

Management structure model before the adjustment. GFI, goodness of fit index; NFI, normal fit index; RMR, root mean square residual; RMSEA, root mean square error of approximation.

The path coefficients reveal varying degrees of influence. Notably, inefficiency in legal administration exhibits the highest factor loading (1.00), indicating its significant contribution to overall project risk. By contrast, factors such as the discovery of antiquities and special work conditions show minimal impact, with factor loadings of 0.03 each. The influence of inefficiency in construction management is negative (–0.20), suggesting a counterintuitive relationship that may require further investigation. The pathways connecting risk management to project outcomes suggest a strong influence, with coefficients of 0.98, 1.02 and 1.00 for duration, budget and quality, respectively. However, the model’s fit indices indicate poor overall fit (GFI = 0.813, CFI = 0.864, RMSEA = 0.175), suggesting the need for refinement. High values for normed chi-square (15.024) and RMR (6.619) further indicate discrepancies between the hypothesised model and the observed data.

This initial model underscores the need for iterative adjustments to better align theoretical assumptions with empirical data and improve overall model fit.

The results of analysing the SEM of risk factors affecting irrigation construction projects have found that the initial model was inconsistent with the empirical data, making it impossible to draw conclusions and interpretations, as shown in Figure 2. In addition, there were the under-criteria statistical values, which were chi-square significant (p-value <0.05), relative chi-square value at 20.744 (higher than criteria), NFI value at 0.893 (lower than criteria), GFI value at 0.825 (lower than criteria), CFI value at 0.897 (lower than criteria), standardised RMR value at 0.024 (pass criteria) and RMSEA value at 0.243 (higher than criteria). This indicated that the initial model was inconsistent with the empirical data and needed to be adjusted. Later, the model was adjusted following the model modification indices (MI) value suggestion, and the results are shown in Figure 3. In addition, the statistical values were as follows: chi-square significant (p-value = 0.050), relative chi-square value at 0.212, NFI value at 0.996, GFI value at 0.990, CFI value at 0.999, standardised RMR value at 0.080 and RMSEA value at 0.030. These statistical values passed the criteria, indicating that the SEM of risk factors affecting irrigation construction projects was consistent with the empirical data, as shown in Figure 3. In addition, analysing the consistency and harmony of a measurement model for each latent variable to find the relation was demonstrated in Table 8. Considering the estimation of the parameters from regression analysis (R²), the result of analysing the structural equation found that the R² value was 0.688, which means those factors can interpret the SEM.

Fig. 3:

Risk management structure model after the adjustment. AGFI, adjusted goodness of fit index; GFI, goodness of fit index; NFI, normal fit index; RMR, root mean square residual; RMSEA, root mean square error of approximation.

To adjust the SEM following the model MI suggested value, the correlation was adjusted based on a review of the relevant literature and theories. Additionally, there were five relations among the variables that were adjusted as shown in Figure 3, which was the relation between the duration of construction projects and the condition of special work, such as when the construction contractor experiences a condition on the site that differs materially from the conditions indicated in the contract with the owner of the project or from what is normally expected in the site area. Almost all construction projects have risks. One of the major risks in construction projects arises when the contractor encounters a different site conditions (Srinavin et al. 2021). Moreover, this risk has been related to the inefficiency of legal administration and the discovery of antiquities at construction sites, which is crucial to construction project management according to the PMBOK. However, most countries try to apply several technologies to avoid risks that affect their projects. Several tools have been developed, such as machine learning and big data (Taylor and Littleton 2008).

The SEM highlights the relationships among identified risk factors, risk management and project outcomes (duration, budget and quality). The model reveals that risk factors such as inefficiency in legal administration, discovery of antiquities at construction sites and special work conditions impact overall project risk, which in turn influences the effectiveness of risk management strategies. Among these, inefficiency in legal administration is the most significant contributor, with a factor loading of 2.24, emphasising the critical need for streamlined legal processes to reduce delays and uncertainties. This is followed by the discovery of antiquities (factor loading 1.41), which underscores the importance of pre-construction assessments in historically significant areas, and special work conditions, which have the least impact (factor loading 0.05), although careful planning is still essential to mitigate minor risks. The model further demonstrates the direct impact of risk management on project outcomes. A path coefficient of 1.00 to project duration indicates that effective risk management plays a crucial role in avoiding delays, while a coefficient of 1.03 to budget highlights its importance in controlling costs. The slightly lower path coefficient of 0.98 to quality reflects that, while critical, risk management has a marginally lesser effect on ensuring project quality compared to cost or time. Overall, the relationships in the model signify that proactive and efficient risk management strategies are indispensable for mitigating the effects of project risks, particularly on financial and temporal outcomes. Additionally, the model stresses the relative importance of addressing legal inefficiencies, as they are the dominant risk factors, and tailoring risk management strategies to address specific high-priority risks effectively. The practical implications include prioritising legal and administrative process improvements, conducting thorough site assessments to identify archaeological risks early, and ensuring that risk management systems are integrated into project planning and execution phases. This approach not only minimises disruptions but also enhances the likelihood of achieving project goals in terms of duration, budget and quality. By understanding the varying influence of risk factors and leveraging efficient risk management, project teams can develop more targeted and effective strategies to ensure success. The SEM thus serves as a valuable tool for identifying critical areas of intervention and optimising project performance through structured risk management practices.

In construction project management, budget and quality are highly related to the status of projects, such as the high quality of work in construction, which should incur a lot of cost if the owner would like to perfect the building (Peterzens-Nysten 1998). This relationship has influenced the discovery of antiquities at construction sites, where remains of ancient civilizations have occasionally been uncovered (Ramaswamy, 2013). It has been studied that Egyptian antiquities are being protected using the dewatering technique. These tools are very expensive and require expertise in this technique. Moreover, some locations are very hard to find (Kusonkhum et al. 2023). When this situation happens in a construction project, it can result in significant costs. The prevention process will save their company, such as risk management (PMBOK), and some technologies will be used to predict the risk and feasibility of users or management, even though some of these tools cannot be applied for some reason in some countries.

Finally, risk management focuses on the time, cost and quality of projects. However, most irrigation projects are looking at soil moving work because they always build dams or canals to serve their people. Thus, the discovery of antiquities at construction sites and the condition of special work are very important in this case. Most of the population in this study proves that most firms in construction industries should be concerned about this result because it will decrease damage to the government and improve their construction management knowledge according to the PMBOK.

The SEM demonstrates the complex interrelationships among risk factors, risk management and project performance outcomes (duration, budget and quality) in construction projects. The model identifies three major risk factors: inefficiency in legal administration (1.00), discovery of antiquities at construction sites (0.03) and the condition of special work (0.03), all of which collectively influence overall risk exposure. Among these, legal inefficiency exerts the strongest impact on risk factors, indicating that bureaucratic obstacles, regulatory delays and contract disputes are dominant sources of uncertainty in construction projects. By contrast, the discovery of antiquities and specialised work conditions have a much smaller effect, suggesting that while these risks exist, their frequency and overall impact on construction projects are relatively low. These risk factors contribute to the overall risk environment, which directly affects risk management (0.06), implying that as risk factors increase, the need for structured risk management approaches becomes more critical. However, the small coefficient suggests that risk management effectiveness can mitigate the negative influence of these risk factors if implemented properly.

The model further demonstrates that risk management serves as a crucial intermediary variable, significantly influencing project outcomes. The strongest impact is observed on budget (1.03), followed by duration (1.00) and quality (0.98), which indicates that financial management is the most sensitive aspect of risk exposure in construction projects. This relationship suggests that poor risk management leads to budget overruns, while effective risk strategies can reduce cost escalations. The slightly lower coefficient for quality (0.98) implies that while risk management plays a critical role in maintaining project standards, other external factors, such as material quality and labour expertise, may also influence quality outcomes independently. Additionally, the interconnections between project performance variables reveal that budget and quality (0.01) are weakly correlated, meaning that cost fluctuations slightly impact quality but do not directly dictate project standards. The model also illustrates a feedback loop between risk factors and project outcomes, emphasising how legal inefficiency (2.24) and financial constraints (0.05) create reinforcing cycles of risk exposure, meaning that poor legal administration not only introduces risks but also exacerbates delays, budget issues and quality concerns over time.

The high model fit indices (chi-square = 1.652, p-value = 0.799, CFI = 1.000, GFI = 0.999, RMSEA = 0.000) confirm that the proposed relationships between variables are statistically significant and accurately reflect real-world project dynamics. These findings reinforce the need for proactive legal frameworks, robust risk assessment mechanisms and financial planning strategies to improve the efficiency of construction project. Future research should explore alternative risk variables, such as fluctuating material prices or labour shortages, and integrate qualitative insights through interviews or case studies to provide a more comprehensive understanding of how risk factors interact over time. Additionally, emerging technologies like machine learning and Internet of Things (IoT)-driven risk monitoring systems could be integrated into risk management frameworks to enhance predictive capabilities and improve real-time decision-making. Overall, this SEM highlights the intricate causal relationships among risk factors, management interventions and project outcomes, emphasising the necessity for a structured and data-driven approach to risk mitigation in construction projects.