Urban rail projects are typically large-scale transport infrastructure projects (megaprojects) which have many potential risks that can influence the strategic goals of owners. However, there is a paucity of studies which explore the impact of risks on both “urban rail” project time and cost together considering quantitative assessments. Therefore, this paper focuses on investigating critical risks and quantifying such risk impacts on urban railway project schedule and cost in practice.
A combination of qualitative and quantitative research methods comprising semi-interviews with five experts and a questionnaire survey of 132 professional respondents is used. The data were modeled using Monte Carlo Simulation to predict the probability of project schedule and cost.
The results show that 30 risk variables are categorized into seven main groups which have significant impacts on both project time and cost. Outstanding five risk variables were highlighted as follows: (1) project site clearance and land compensation; (2) design changes; (3) physical project resources; (4) contractors’ competencies and (5) project finance. Such findings were supported by Monte Carlo simulation which predicted in the worst case that the project may suffer 11.03 months’ delays and have cost overrun with a contingency of US$287.68 million.
This study expands our knowledge about time and cost contingency of urban metro railway implementation across developing economies and particularly within the context of Vietnam. Policymakers will not only gain an understanding about risk structure but will also recognize the significant impacts of critical risk through risk impact modeling and simulation. Such an approach provides insights into risk treatment priorities for planners so that they can proactively establish suitable strategies for risk mitigation in practice.
1. Introduction
Transport infrastructure development plays a crucial role in creating a transport network. During infrastructure project development stages, cost overruns and project delays are two common factors mentioned in many studies (Park, 2021; Gashaw and Jilcha, 2022); and such issues often occur during the construction stage of transport infrastructure projects, significantly influencing the given project objectives. Even though different management approaches to infrastructure project implementation may be used, construction projects in many countries are still facing the serious problem of project delays and cost overruns (Memon et al., 2011). Recently, Flyberg also highlights this issue with his conclusion that infrastructure projects around the world often face worrying cost overruns and significant project delays (Flyvbjerg et al., 2018).
In practice, the level of control of cost overruns and delay issues depends greatly on many factors in accordance with the specific context of each country (Flyvbjerg, 2010). Recently, the Vietnamese government has implemented an infrastructure development strategy through huge investment in urban railway construction in two economic centers (Hanoi City capital in the north and Ho Chi Minh City in the south) to solve major problems regarding urban pressures existing in big cities, including traffic jams and environmental pollution. On Vietnamese social media during 2006–2010, both the local government and the public had overly optimistic expectations for the performance of the mega project, but the project implementation outcomes in practice were tragic and negatively impacted investors’ reputations.
The government of Ho Chi Minh City started to construct the HCMC Metro 1 in 2006 with an expected schedule of four years, but the given schedule has been extended, and the total budget has increased 2.5 times compared to the original approved budget (Nguyen et al., 2022). Similarly, the government of Hanoi started to construct the HNC Metro 2: Cat Linh-Ha Dong in 2010, and the local government expected to complete this important project within two years. Unfortunately, until now the trains are only at the testing stage, and the total budget spent on this project has nearly doubled compared to the initial approved budget ($552.86 million) (Vietnamnetplus, 2022).
These two iconic infrastructure projects in Vietnam have received much attention from both the public and local communities. In the context of Vietnam, both the local government and the community have high expectations regarding urban rail projects that can solve critical issues in big cities (e.g. reducing traffic congestion and air pollution and encouraging the use of public transport) as well as creating the backbone of the complex transport system. Even though building a simple metro line may cost from 2 to 4 billion US dollars, depending on the project scope (length, tunnel, metro width, or size), urban railway is a promising solution that has been proven to transport a large number of passengers many times during the day with a high speed of more than 100 kilometers per hour (HanoiMetro, 2023). In addition, urban railway development is an indispensable trend of all megacities in the world, which does not exclude 2 cities in Hanoi and Ho Chi Minh City; however, the delays and cost overruns of such mega projects lead to controversial debates on the efficiency of resource allocation and places a question mark on project feasibility, as well as highlighting the important role of and requirement for systematic risk assessment.
Thus, this study significantly focuses on metro railway project time and cost dimensions together, and it aims to investigate the structure of risk and associated critical risks that greatly influence the time and cost of metro railway projects and to provide a practical viewpoint on the major causes leading to such risks. The research particularly concentrates on the urban railway project in Vietnam, which is considered a typical context of a developing country with an urgent need for transport infrastructure development.
2. Time and cost causation risk factors
2.1 Urban rail projects
Urban rail is defined as rail in an urban area, including both heavy and light rail, which may be underground, at level, or elevated (Flyvbjerg, 2007). According to Pulido et al. (2018), urban rail projects are megaprojects with significant technical, institutional, financial, environmental, and social challenges and interdependencies. They are capital-intensive and frequently high-risk. As the number, variety, and complexity of urban rail projects increase, it is important to comprehend how to approach their development. Urban rail projects have the potential to reduce poverty and contribute to shared prosperity, but if they are inadequately planned or executed, they could be retrogressive and have negative effects (Pulido et al., 2018). Pulido et al. (2018) also showed that numerous developing cities have little to no expertise with the planning and implementation of such complex projects and often underestimate the actual cost required, leading to budget overruns and schedule slippage. Therefore, research on urban rail projects is essential and necessary for mitigating any potential negative effects, particularly those related to cost and schedule.
In the context of urban rail projects, however, only a small number of research papers have been published in prominent scholarly journals, despite the subject matter’s economic and social significance (Li and Love, 2020). There have been some studies relating to identifying risk factors in rail projects in countries around the world. In Thailand, Ghosh and Jintanapakanont (2004) identified and assessed the critical risk factors in underground rail projects. In Singapore, Zhao et al. (2013) identified the critical risks in underground rail international construction joint ventures. Recently, identification of risk in metro rail projects in India was also undertaken by Patil et al. (2019) and Karmarkar et al. (2022). In another study in China, Feng et al. (2021) evaluated the risks of urban rail transit projects. However, such studies identified the risk factors in general, but did not focus on specific aspects of the project performance regarding cost and time overruns.
Thus, this study aims to identify the potential risk factors leading to cost and time overruns in urban rail projects through a comprehensive review of three resources: (1) factors affecting schedule delays in rail projects; (2) factors leading to cost overruns in rail projects; and (3) risk factors in rail projects.
2.2 Factors affecting schedule delays
Mittal and Paul (2018) identified the five most critical delay factors in India’s metro rail projects: delay in land acquisition and site handover to the contractor, utility relocation and contingency work, scope change, payment delay, and subsurface and changing ground conditions. Meanwhile, Tavassolirizi et al. (2020) conducted a study on factors affecting delays in rail transportation projects in Iran and came to completely different conclusions. Specifically, the existence of numerous decision-making stations, the lack of a central role, the lack of sufficient authority given to project managers, the lack of timely issuance of work permits by operators, and the lack of a correct prediction of the contract schedule were considered the most important delay factors. Moreover, the lack of a correct prediction of the contract schedule was also found to be one of the top five ranked delay factors in Kassa’s (2020) study on federal road and railway construction projects in Ethiopia. The remaining factors in the top five delay factors were incomplete studies prior to project approval and right-of-way acquisition issues.
In urban railway projects in Vietnam, Tran et al. (2020) identified 25 factors affecting implementation progress: including changing materials and construction procedures, problems in project management and administration, a weak and inexperienced project management team, and deferring decisions about changes in the contract. The results of the literature review indicate that there is a paucity of research on the factors that cause schedule delays in rail projects and that almost all of it was conducted in developing nations, such as India, Iran, Ethiopia, and Vietnam. Moreover, major factors influencing schedule delays vary across geographic regions. In general, key problems with schedule delays are due to deficiencies in implementing projects; for example, lack of materials and labor, lack of experience, lack of a correct prediction about the schedule, or delays in works such as decision-making, payment, or land acquisition.
2.3 Factors leading to cost overrun
Hua et al. (2004) conducted a study on cost risk management for the West Rail project in Hong Kong and found that resource factors, management factors, and parent factors were the three main categories of risk factors pertaining to project cost management. In a study on trends in US rail transit project cost overruns, Dantata et al. (2006) asserted that there are numerous causes of cost overruns, including, but not limited to, optimistic underestimation of costs at the conceptual phase, the lengthy project approval and construction process, the omission of project components during early phases, additions to project scope during project development, and difficult-to-predict latent conditions. Lee (2008) identified causes of cost overruns in Korean social overhead capital projects, such as roads, rails, airports, and ports. The findings indicated that 100% of rail projects had a maximum cost overrun of 50%, and cost overruns were due to inefficient budget usage by the national government for social overhead capital investments. Cost overruns could be categorized into several primary categories: scope changes, construction delays, unreasonable assessment and modification of project costs, and no practical application of the earned value management system. Recently, within Sudanese railway projects, Abdorahman (2020) indicated that the top five factors contributing to cost overruns were: economic inflation, change orders, project delays, inaccurate estimates of quantities, and poor budget preparation. Similarly, the findings from the study by Kassa (2020) agreed that material cost inflation, change orders, poor bill of quantity, and poor design were significant factors causing cost overruns. Other significant factors mentioned by Kassa (2020) in the top five were incomplete study project approval and poor project performance monitoring.
In summary, research on the factors that contribute to cost overruns is limited. Regarding the primary causes of cost overruns, the findings of the previous studies appear to be similar. Change orders, construction delays, inflation, inadequate bills of quantities, and inadequate project performance monitoring were found to be significant contributors to cost overruns in rail projects.
2.4 Risk factors in rail projects
Ghosh and Jintanapakanont (2004) identified 35 risk factors in underground rail projects in Thailand: for example, unavailability of funds, delay in solving contractual issues, ecological constraints, design change, unforeseen site conditions, and so on. In a study on risk evaluation of China urban rail transit PPP projects, Feng et al. (2021) placed 27 risk factors into 9 groups: political risk, legal risk, economic risk, environmental risk, market risk, financing risk, project promotion risk, relationship risk, and force majeure risk. They indicated that political risk, legal risk, and project promotion risk were the most significant risk indicators, whereas relationship risk and force majeure risk were the least significant.
Zhao et al. (2013) identified the critical risks in underground rail international construction joint ventures, and they found that the most critical risk factors were identified as: disagreement on some conditions in the contract, disagreement on accounting of profit and loss, distrust between partner employees, labor, material, and equipment import restrictions, and the partner’s lack of management competence and resourcefulness. In the metro rail project of Pune in India, Patil et al. (2019) identified 25 risk factors and four main groups: technical risk, construction risk, socio-political-environmental-surrounding risk, and financial risk. Historical building preservation problems, relocation risk of people surrounding the project, traffic diversion problems, insurance for all projects and surrounding areas, and political pressure were asserted to be the top five risk factors. In another study in the same context in India on the risk assessment of underground and elevated metro projects, Karmarkar et al. (2022) also identified 25 risk factors. The study showed that delay in project approvals and the economic crisis were the two most critical risks from the perspectives of both clients and contractors among the two types of metro systems, elevated and underground.
Based on the relevant review of the above-mentioned three resources, it can be seen that in research areas related to risk factors in rail projects, the factors identified were general and not mainly concentrated on cost and time overruns. Regarding the research on factors affecting schedule delays and cost overruns in rail projects, previous studies have separated cost overruns and time overruns instead of focusing on both cost and time overruns. Consequently, it may be inadequate to provide solutions that simultaneously address cost and time overruns, as it is evident that cost and time are inextricably linked. Also, most earlier studies concentrated on the identification of risk factors but failed to uncover the underlying relationship between risk factors, which can aid parties in swiftly perceiving the risk factors’ primary characteristics. This study will address the above-mentioned gaps. As a result of the literature review, 25 potential risk factors leading to cost and time overruns in urban rail projects were identified, as shown in Table 1. Table 1 also includes five risk factors added by experts through a pilot test, which are shown in the following section.
List of risk factors leading to cost and time overruns in urban rail projects
| ID | Risk factors | Related research |
|---|---|---|
| R01 | Sketchy and incomplete design | Ghosh and Jintanapakanont (2004), Zou and Li (2010), Mittal and Paul (2018), Patil et al. (2019), Tran et al. (2020), Kassa (2020), Feng et al. (2021), Karmarkar et al. (2022) |
| R02 | Slow design approval and appraisal process | Han et al. (2009), Tavassolirizi et al. (2020), Mittal and Paul (2018) |
| R03 | Unexpected changes in the design approved | Ghosh and Jintanapakanont (2004), Hua et al. (2004), Han et al. (2009), Patil et al. (2019), Tavassolirizi et al. (2020), Abdorahman (2020) |
| R04 | ODA (Official Development Assistance) projects suffering from tight constraints of main sponsors | Experts’ opinion |
| R05 | Slowly disbursed project capital | Tavassolirizi et al. (2020) |
| R06 | Payment delays | Ghosh and Jintanapakanont (2004), Zou and Li (2010), Mittal and Paul (2018), Patil et al. (2019), Tavassolirizi et al. (2020), Abdorahman (2020), Karmarkar et al. (2022) |
| R07 | Exchange rate change | Ghosh and Jintanapakanont (2004), Zou and Li (2010), Zhao et al. (2013), Feng et al. (2021), Karmarkar et al. (2022) |
| R08 | Changes in material, labor and machinery costs | Hua et al. (2004), Lee (2008), Patil et al. (2019), Tran et al. (2020), Feng et al. (2021) |
| R09 | Lack of coordination and support from government agencies at all levels | Zou and Li (2010), Tavassolirizi et al. (2020), Mittal and Paul (2018), Patil et al. (2019), Tran et al. (2020) |
| R10 | Interface problems between contractors due to lack of information and consensus | Tran et al. (2020) |
| R11 | Changes in local policies and regulations | Zou and Li (2010), Zhao et al. (2013), Patil et al. (2019), Tavassolirizi et al. (2020), Tran et al. (2020), Feng et al. (2021), Karmarkar et al. (2022) |
| R12 | Lack of standards for design, construction and verification for new projects first implemented | Experts’ opinion |
| R13 | Lack of mechanisms and regulations for managing EPC (engineering, procurement and construction) contract type | Experts’ opinion |
| R14 | Delays during the bidding and contractor selection from the project management unit | Tran et al. (2020) |
| R15 | The project management unit’s lack capacity and experience in managing and operating complex projects | Tran et al. (2020), Zhao et al. (2013), Kassa (2020) |
| R16 | Delays in decision regarding contract changes and project adjustments from the project management unit | Ghosh and Jintanapakanont (2004), Tran et al. (2020) |
| R17 | Consultants’ shortcomings during project quality supervision | Tran et al. (2020) |
| R18 | Errors, accidents and work problems at construction site due to contractors’ poor management | Ghosh and Jintanapakanont (2004), Mittal and Paul (2018), Tran et al. (2020), Abdorahman (2020), Karmarkar et al. (2022) |
| R19 | Delays in making and approving dossiers regarding plan, design, quality, acceptance documents | Tavassolirizi et al. (2020), Tran et al. (2020), Abdorahman (2020), Karmarkar et al. (2022) |
| R20 | Changes in project scope | Ghosh and Jintanapakanont (2004), Hua et al. (2004), Lee (2008), Mittal and Paul (2018), Patil et al. (2019), Abdorahman (2020), Kassa (2020), Karmarkar et al. (2022) |
| R21 | Lack of raw materials and new equipment necessitated importation | Tavassolirizi et al. (2020), Mittal and Paul (2018), Patil et al. (2019), Kassa (2020), Karmarkar et al. (2022) |
| R22 | Obstacles regarding heavy traffic obstruct and the limitation of construction site | Mittal and Paul (2018), Patil et al. (2019), Karmarkar et al. (2022) |
| R23 | Lack of skilled labor and top tier professionals in field required | Han et al. (2009), Tavassolirizi et al. (2020), Mittal and Paul (2018), Abdorahman (2020), Kassa (2020), Karmarkar et al. (2022) |
| R24 | Delays in import and shipping goods and machineries for project implementation | Zhao et al. (2013), Mittal and Paul (2018) |
| R25 | Complicated construction technology | Experts’ opinion |
| R26 | Slow land compensation and construction site clearance | Zou and Li (2010), Tavassolirizi et al. (2020), Mittal and Paul (2018), Patil et al. (2019), Tran et al. (2020), Han et al. (2009), Kassa (2020), Karmarkar et al. (2022) |
| R27 | Difficulties in relocating and protecting existing works required from local governments | Mittal and Paul (2018), Patil et al. (2019), Tran et al. (2020), Karmarkar et al. (2022) |
| R28 | Litigation and disputes between parties | Ghosh and Jintanapakanont (2004), Hua et al. (2004), Han et al. (2009), Zou and Li (2010), Zhao et al. (2013), Mittal and Paul (2018), Patil et al. (2019), Abdorahman (2020), Kassa (2020), Karmarkar et al. (2022) |
| R29 | Unforeseen geological conditions | Ghosh and Jintanapakanont (2004), Lee (2008), Zou and Li (2010), Tavassolirizi et al. (2020), Mittal and Paul (2018), Tran et al. (2020), Han et al. (2009), Abdorahman (2020), Kassa (2020), Feng et al. (2021), Karmarkar et al. (2022) |
| R30 | Interruption of construction and associated work volume verification due to epidemic | Experts’ opinion |
Source(s): Table by authors
3. Research methodology
The framework of this study is described in Figure 1. The study begins with a literature analysis to identify risk factors leading to cost and time overruns. The outcome of the literature review is a list of probable risk factors that were used to construct the questionnaire to gather the respondents’ views. According to Thomas (2003), the questionnaire survey is widely acknowledged and utilized in social science, general management, and project management studies. Besides, with appropriate participants, it can produce high-quality useful data, obtain high response rates, and enable anonymity, which encourages more honest and frank responses than, for example, interviews. This can aid in the reduction of prejudice (Marshall, 2005).
To validate the questionnaire, a pilot test was conducted with five experienced experts in urban rail projects, including two project managers from a project management unit, two consultant managers, and one contract manager. The experts were contacted to examine the clarity of the questions in the questionnaire and the adequacy and suitability of risk factors in the context of the urban rail projects. This means that the experts can remove or add risk factors as needed. As a result of this process, five new risk factors were added (ODA projects suffering from tight constraints of main sponsors; lack of standards for design, construction, and verification for new projects first implemented; lack of mechanisms and regulations for managing EPC contract types; complicated construction technology; and interruption of construction and associated work volume verification due to epidemic). The final questionnaire, comprising two main parts, contains 30 risk factors leading to cost and time overruns. Based on five-point Likert scales, respondents were asked to rate the likelihood of occurrence (ranging from 1 – rare to 5 – almost certain), impact on the cost of the project, and impact on the schedule of the project (ranging from 1 – insignificant to 5 – very significant) of each risk factor.
The finalized questionnaire was distributed to parties directly involved in urban rail projects in Vietnam through face to face, email, and Zalo Messenger, including project management units, consultants, contractors, and governmental agencies. The convenience sampling method is adopted in this study since the number of metro railway projects in Vietnam during the last five years has been limited to only two mega projects in Hanoi and Ho Chi Minh City, and the number of direct participants in metro railway projects in Vietnam is limited. Such urban rail projects are firstly implemented and face significant challenges for implementation in practice. According to Sekaran (2000), this sampling method is preferable when it is difficult to obtain responses using the random sampling method. All 150 questionnaires were sent to the target organizations in the Vietnamese construction industry. A total of 135 completed questionnaires were received, yielding a response rate of 90%. Out of 135 responses, three were rejected due to all questions in the questionnaire having the same answers. Therefore, this study is based on 132 valid responses.
Firstly, risk factors are ranked based on their risk level. The risk level (RL) has been recognized as a function of the likelihood of occurrence (P) and impact of each risk factor (I) (Cooper et al., 2005). The RL of a certain risk factor can be calculated as follows:
In addition, Monte Carlo Simulation (MCS) was used to provide forecasts on contingencies regarding risk impacts on cost and time. The inputs used for MCS are cost and time variables identified at the project feasibility analysis stage, and such variables need to reflect the structure of the project cost and project schedule. In this study, Crystal Ball software was selected to provide the entire range of possible outcomes as well as the likelihood of achieving such outcomes. The cost and time baseline of HCM Metro 2 was selected from the official reports of the local government. The results from MCS were used to confirm whether the risk variables have significant impacts on project time and cost.
4. Data analysis
4.1 Demographics of respondents
Most respondents, 80 (60.6%), come from construction companies, while 16 (12.1%) are from the investor or sponsor sectors. 19 respondents (14.4%) come from consulting units such as design consultants and supervision consultants, and the remaining group, including experts and staff from local government, accounted for 12.9%.
The majority of respondents (50) have more than 15 years of experience, which accounts for 37.9% in total, and the percentage of respondents with work experience between 11 and 15 years is 33.4%. Respondents with 6–10 years of experience accounted for 18.9%, in comparison with 9.8% of participants who had less than five years of work experience in the construction industry.
4.2 Reliability analysis
The value of Cronbach’s Alpha coefficient of 30 risk variables was examined with risk probability and risk impact on infrastructure project time and cost. In this study, the Cronbach’s Alpha coefficients for 30 risk variables are 0.814 and 0.806, which are greater than the acceptable threshold of 0.7, indicating that the items have relatively high internal consistency (Ghosh and Jintanapakanont, 2004).
4.3 Analyzing the risk level of factors leading to delays and cost overruns
The combination between the risk likelihood and risk impact on project objectives, considering time and cost, is calculated following the risk level (RL) formula (1) presented above. The minimum benchmark used for ranking risk variables is 85% (0.85). As a result of the calculation, ten major risks are presented in Table 2.
Ten critical risks affecting to both project time and cost
| List | Risk variables | Risk impact on time | Risk impact on cost | ||
|---|---|---|---|---|---|
| Mean | Ranking | Mean | Ranking | ||
| 1 | R27. There are difficulties in relocating and protecting existing works required from local governments | 0.94 | Very high | 0.93 | Very high |
| 2 | R26. Land compensation and construction site clearance is complicated and slow in practice | 0.94 | Very high | 0.94 | Very high |
| 3 | R08. Material, labor and machinery costs have significant change | 0.92 | Very high | 0.94 | Very high |
| 4 | R04. ODA (Official Development Assistance) projects have tight constraints imposed by main sponsors | 0.92 | Very high | 0.93 | Very high |
| 5 | R22. There are obstacles regarding heavy traffic obstruct and the limitation of construction site | 0.92 | Very high | 0.91 | Very high |
| 6 | R30. The epidemic may disrupt the project execution stage and associated work volume verification | 0.92 | Very high | 0.91 | Very high |
| 7 | R15. Investors (sponsors) lack capacity and experience in managing and operating complex projects using new technology | 0.90 | Very high | 0.90 | Very high |
| 8 | R05. Project capital is slowly disbursed and complicated | 0.90 | Very high | 0.90 | Very high |
| 9 | R03. Unexpected changes in the design approved | 0.89 | High | 0.90 | Very high |
| 10 | R16. Delays in decision regarding contract changes and project adjustments from the main investor | 0.90 | Very high | 0.89 | High |
Source(s): Table by authors
4.4 Factor analysis
In the current study, the Bartlett’s test of sphericity was 2401.075 (p < 0.001), which indicates that the data collected are appropriate for such analysis. In addition, the results of the KMO test to measure the sampling adequacy were 0.907 and 0.871, which are greater than the required value (KMO >0.6) (Ghosh and Jintanapakanont, 2004).
The cumulative variances explained value for seven factors identified (time and cost) are 72.595 and 68.980%, respectively, which are both greater than 50%, and such values are considered sufficient for research in the humanities (Williams et al., 2010). Next, the study used varimax rotation to maximize the distribution of loadings within factors (Weide and Beauducel, 2019). In this study, seven main factors, after rotation, were extracted from 30 risk variables. From the results analyzed, seven main factors affecting project time can explain up to 72.595%; and seven risk factors can also explain up to 68.980% of the variation. The factor loading ranges from 0.501 to 0.798, which is higher than the minimum value of 0.3, which is a good rule of thumb (Brown, 2015). This indicates that all correlations between variables and factors are important. Based on the extracted groups, the authors re-coded the names for the groups as shown in Table 3.
Risk factors coding
| ID | Risk factors | Variables | |
|---|---|---|---|
| IT1 | Project time | Project implementation | R02, R09, R10, R14, R15, R16, R17, R18, R19 |
| IT2 | Project context uniqueness | R20, R21, R22, R23, R24, R25, R29 | |
| IT3 | Project finance | R04, R05, R06, R07 | |
| IT4 | Project site preparation and design | R26, R27, R01, R03 | |
| IT5 | Law and local regulations | R13, R12, R11 | |
| IT6 | Social environment | R28, R30 | |
| IT7 | Physical project resources (e.g. labor, machinery and material) | R08 | |
| IC.1 | Project cost | Project planning | R01, R02, R03, R14, R15, R16 |
| IC.2 | Project planning competency | R10, R17, R18, R19 | |
| IC.3 | Project context uniqueness | R20, R25, R24, R21, R29, R23, R22 | |
| IC.4 | Project finance capacity | R04, R05, R06, R07, R08 | |
| IC.5 | Law and local regulations | R12, R11, R13 | |
| IC.6 | Social environment | R28, R30 | |
| IC.7 | Site clearance and land compensation | R26, R27 |
Source(s): Table by authors
In terms of validation, the KMO values of these factors range from 0.87 to 0.91, and the average variance explained values of extracted factors range from 62.89% to 87.62%, which are higher than the minimum required value (0.5) referred to in a similar assessment in the study of Ghosh and Jintanapakanont (2004). This indicates that all factors are acceptable in terms of construct validity.
4.5 Monte Carlo Simulation
The simulation results are 10,000 times iterated to provide statistical data and the probability scenarios (P) with the degree of certainty that are shown through the distribution histograms and the cumulative frequency histograms. At the time of the model simulation, the HCMC project was in the stage of site clearance and prequalification of contractors. The contingency estimation values for time and cost were calculated based on the time and cost 'baseline', which was in accordance with the investment plan approved by the local government in 2013 after several revisions. Clark (2001) proposed that 80% or 90% probability levels should be chosen for redundancy considering the large scale of construction projects. Thus, the probability levels P5 (5%), P95 (95%), and a confidence level of 90% are selected for HCMC Metro 2, Ho Chi Minh City, due to the complexity level of a mega infrastructure project.
Firstly, MCS applied for time contingency estimation is illustrated in Figure 2. In the case of the most optimistic probability (P-Minimum), the project is likely to be ahead of schedule (2.40%), which is approximately 1.15 months, and the total complete time is 46.85 months. On the other hand, considering P-Maximum with a provision time of 22.97%, the project will be completed with a total construction time of 59.03 months, which means the project is behind schedule at 11.03 months later than the original plan (22.97%*48 months). The average probability (P-mean) of the hedging period is 9.77%, and the total project completion time is 52.69 months, which means 4.69 months later than planned (9.77%*48 months).
Secondly, MCS applied for cost contingency estimation is illustrated in Figure 3. For the most optimistic probability (P-Minimum), the cost savings for the project are 0.91%, which is around $12.51 million (1371.5 million * 0.91%), and the total cost for the project is $1361.9 million. On the other hand, with P-Maximum with a provision cost of 20.93%, the project cost has a significant increase with additional costs ($287.68 million), which leads to a high budget of $1662.18 million compared to the cost baseline of $1371.5 million. In the case of P-Mean, the percentage of the contingency cost is 9.72%, which is around $133.6 million (9.72%*1371.50), and the total budget estimated is 1508.14 million, which is still higher than the cost baseline approved.
Next, the author investigated the sensitivity of risk variables using a tornado diagram to measure the contribution of risk variables to project time and cost estimation in the MCS. The top five risk variables affecting time estimation are R05-Slowly disbursed project capital (12.9%); R26- Slow land compensation and construction site clearance (12.2%); R15- The project management unit’s lack capacity and experience in managing and operating complex projects (11.4%); R14- Delays during the bidding and contractor selection from the project management unit (10.6%); and R16- Delays in decision regarding contract changes and project adjustments from the project management unit (10.5%).
As Figure 4 shows, the top five risk variables affecting cost estimation are R26- Slow land compensation and construction site clearance (12.2%); R05-Slowly disbursed project capital (11.1%); R08- Changes in material, labor and machinery costs (10.2%); R30 – Interruption of construction and associated work volume verification due to epidemic (10.2%); and R27- Difficulties in relocating and protecting existing works required from local governments (10.2%).
5. Findings and discussion
There are some similarities when considering critical risk impacts on both project time and cost. In this study, the main causes leading to critical risks are investigated, including: (1) Project Site Clearance and Land Compensation; (2) Design Changes; (3) Physical Project Resources; (4) Contractors’ Competencies; and (5) Project Finance.
(1) Project Site Clearance and Land Compensation: IT4 is a critical risk factor that significantly affects project time. IT4 includes four main risk components: R26, R27, R01 and R03. R26 – slow land compensation and construction site clearance was confirmed in the Monte Carlo Simulation (MCS) as the most significant risk variable contribution to the change in the time and cost estimation (12.2%). Such a figure was confirmed by the study of Idrees and Shafiq (2021) carried out in Pakistan, which indicated that land acquisition and payment settlements were ranked 2nd in the top ten delay factors. R26, in practice, takes a long period of time due to three main causes: (1) resistance of local people to be relocated; (2) the valuation of land pricing based on a pricing frame provided by the states is very low compared to the land pricing of the current market; and (3) threats to traditional business of citizens within the construction zone. These causes lead to conflicts in negotiation between landowners and local governments in making agreements on the land prices used for compensation. In addition, R27-difficulties in relocating and protecting existing works, is also a critical risk affecting project time. This variable also was confirmed by the MCS which indicated that R27 contributes to 9.1% of the project time estimation and 10.2% of total project cost.
The main cause leading to this risk is because there is a lack of a collaboration mechanism between official departments of local government in evaluating the status of current works and controlling the quality of such works while moving such works to another place in practice. To be similar to IT4, IC7-land clearance and compensation consists of two main risk components (R26 and R27), which are like those affecting project time. Given the very high RLT of both IT4 and IC7 at 0.91 and 0.93, these are clearly two main risk factors that significantly influence both urban railway project time and cost. These results imply that the government must enhance its compensation regulations and policies: for example (1) the compensation price must be assessed fairly and based on the actual value of the affected property; (2) local governments should provide programs and opportunities to help resettled individuals obtain new employment. More importantly, it is essential that resettled people have access to the necessary services such as hospitals, schools, medical facilities, and other social services.
The findings regarding land compensation programs for citizens in the construction zone have been researched by several scholars, particularly in the context of developing countries. Typically, Liu et al. (2016) conducted many interviews regarding the land compensation program for villagers in West China to investigate landowners’ perspectives. Interestingly, most landowners want to receive more land compensation than money compensation, because they want to continue with their sustainable jobs in the agricultural industry compared with becoming workers in industrial zones. Land acquisition was indicated as one of the major causes of delay in mega projects (Sarkar and Singh, 2022) that influences construction time performance and often leads to cost overruns. In terms of delays in site clearance and providing the site for construction, Williams (2003) highlighted that a three-month delay in a project can lead to a one-year delay in execution, resulting in substantial costs. Delays in one core activity can lead to a change in a series of subsequent tasks and directly require a significant change in resource planning for project implementation (Danial and Misnan, 2022). Thus, it is essential to carry out a detailed survey regarding the status of the project area to provide statistics and to specifically evaluate the barriers or difficulties in relocating existing works (e.g. houses, old trees, traditional landmarks, and other underground systems) to other places. All issues should be documented in the preliminary planning (Kumaraswamy et al., 2017) to make sure that all key parties involved (investors, contractors, and local government) have sufficient awareness about difficulties in practice.
(2) Design Changes: R01 (The design is sketchy and lacks important details) and R03 (Unexpected changes in the design approved) belong to IT4. From the factor analysis, these are two main risks that affect project time and cost. R03 also received strong confirmation from the MCS when it indicated that R03 contributes to 9.1% of total cost estimation. This finding is also confirmed in the study of Idrees and Shafiq (2021) conducted in Pakistan, which indicated that non-availability of drawings and rework due to errors in design were ranked fifth out of ten delay factors. The main causes leading to IT4 in practice are insufficient surveys, sketchy appraisals, and lack of experienced staff. These causes often lead to delays in the project implementation schedule and to an increase in the total budget approved for changes and contingencies in practice (Mathar et al., 2020). Even though both sponsors (project investors) and contractors in infrastructure projects have arranged a fixed amount of budget for contingencies (5%), such practices as R03 and R01 often lead to significant cost overruns and time delays. Therefore, the appropriate risk mitigation strategies should be applied from the outset, including setting up a competent team of experts, having a clear and comprehensive design brief, investing time and effort in the design development phase, and enabling the early involvement of key stakeholders thorough the project stages. Importantly, defining the project’s objectives, scope, requirements, and constraints at the outset is a crucial step for establishing a firm design foundation, as well as for reducing the need for significant changes later.
(3) Physical Project Resources (E.g. Labor, Machinery and Material): IT7 is a critical risk factor that significantly influences both project time and cost. IT7 directly refers to the change in labor cost, machinery cost, and material cost in practice. IT7 includes R08 which received strong confirmation from the MCS. From the simulation of MCS, R08 contributes to 5.6% project time estimation; and 10.2% total cost estimation. The cost components mentioned in R08 take significant account of the total construction project cost; and any changes in costs estimated for labor, machinery and material may greatly affect the financial allocation planning of contractors. Recently, Andrić et al. (2019) used the statistical data of 102 infrastructure projects in Asia, including 58 highway projects, and found that the key causes of cost overruns are the increase in the cost of resources (construction materials, equipment, and labor). For some ODA projects, the request for changes in the input costs is complicated and takes time for approval, so the factor IT7 should be considered as a main risk leading to project delays in practice. Similarly, IC3 also consists of two main risk variables, namely R21-Lack of raw materials and new equipment necessitated importation, and R23- Lack of skilled labor and top tier professionals in the required field, which are the main risk variables contributing to the cost overrun of urban railway projects. Practically, the main cause leading to IT7 is the rising inflation in labor and materials’ costs which is a major concern of both investors and contractors considering the profitability and project effectiveness. Thus, the inflation rate in construction materials and labor costs should be considered a mandatory factor that must be mentioned in the risk assessment and associated mitigation strategies for critical infrastructure projects.
(4) Contractors’ Competencies: IT1 mainly refers to competencies in project planning and implementation. Even though this risk factor was ranked 7th in terms of influencing project time, this factor significantly influences project cost, with a ranking of 2nd (IC.2). Several main causes contributing to the likelihood of IT1 are: (1) the investor lacks the capacity and experience to manage complex projects (R15); (2) delays in making decisions for design approval (R02); and (3) contract changes (R16). R15 and R16 were also strongly confirmed by the MCS, which indicated that these risk variables contribute to 11.4 and 10.5% of the project time estimation, and R15 also contributes to 9.8% of the total project cost estimation. R15 and R16 may directly contribute to an increase in project costs due to unexpected compensation for contractors because delay errors are not caused by the contractors. In addition, a sketchy and flawed design survey (R01) at the schematic design step reflects investors’ incompetence, which also contributes to the high level of risk impact. Thus, it is crucial to consult or work with experienced contractors with a good reputation to have a holistic picture of the project as well as to identify challenges and barriers (e.g. technology and associated construction methods, financial allocation, and contract type selection) in implementing complicated infrastructure projects in practice.
(5) Project Finance: Risk factors including IT3 and IC4 were ranked at 4th and 3rd among the critical risks that influence project time and cost. There is a similarity between IT3 and IC4: R05, the status of slowly disbursed project capital. The Tornado Diagram shows that R05 significantly contributes to 12.2% of total cost estimation, while it also has a considerable contribution to total project time of 12.9%. Similarly, R04 (Official Development Assistance projects have tight constraints on their main sponsors) is also an important risk variable since this risk contributes highly to both the likelihood and impact of risk. This risk variable was also confirmed by the MCS when the Tornado Diagram showed that R04 contributes to 8.5% of the total time estimation and 9.2% of the total cost estimation. The main reason for this is the use of ODA loans financed by different capital sources and therefore bound by the constraints presented in the Framework Agreement between Vietnamese and foreign investors.
There are many constraints that refer to the use of technology, machinery, and materials, as well as the selection of contractors involved in bidding activities. In practice, the negotiating process to get contract approval and get the necessary loan takes a long time, which explains the high level of likelihood and impact for R04 in practice. This crucial point is also mentioned in the study of Mahmud et al. (2021) conducted in Nigeria, in which they asserted that limiting the sustained availability of funds to key projects often leads to stoppage of work pending the availability of funds. Thus, it is essential to raise awareness about sponsors’ funding requirements as well as to have a sufficient understanding of complicated collaboration mechanisms among parties in practice. This is key to reaching consensus between sponsors and contractors as well as to mitigating financial risks.
In summary, the findings presented above highlight critical issues of local government in developing countries in delivering mega railway projects, including: (1) lack of practical experience in assessing potential risks of mega projects, which entail considerable uncertainty; (2) limitation of sustained availability of funds to key projects; (3) underestimating the complexity of technical aspects and the technology involved in project implementation; and (4) lack of coherent policy and transparent framework for stakeholders’ collaboration, especially for mega projects with the involvement of overseas parties. These critical issues should be examined within a holistic investment picture considering the context of developing countries, which often have many limitations (e.g. finance, technology, and experience in project delivery) to set up priorities to ensure the feasibility of mega projects and also to maximize project impacts on the whole society. In addition, these findings provide a different perspective for international joint ventures in identifying and assessing critical risks that significantly impact urban railway project success in the context of developing countries. In line with this, international joint ventures can improve their awareness of local project contexts as well as their deep understanding of key aspects of megaprojects in practice.
6. Conclusion
This study confirms the critical risks and outstanding problems of urban railway project implementation in the context of developing countries from a risk perspective. The outstanding risks that have great impacts on both time and cost mainly come from issues of land compensation programs and handing over the clear construction site to contractors. Typically, the collaboration mechanism and implementation in practice are considered to contain potential risks that have the same impacts on both project time and cost. The major difference between risk impact on time and risk impact on cost is the availability of project resources available to be spent on project implementation. Lack of physical project resources can also lead to significantly more delays compared to cost overruns in practice. Another different point for risk impact measurement on time and cost comes from the project context, which is often unique and highly dependent on local traditions and regulatory constraints.
Since few studies have been conducted to investigate risk impacts on both project cost and time dimensions together in the context of a developing country, our study has added new knowledge through: (1) not only ranking potential risks but also quantitatively measuring critical impacts on urban railway time and costs; (2) clarifying the risk structure considering the nature of risk factors; and (3) confirming the significant similarities between the traditional risk ranking approach and the modeling approach. The findings shed light and provide insights for both decision-makers and practitioners involved in transport infrastructure projects in the context of developing countries. Understanding potential risks in the early stages of megaprojects brings considerable benefits to active planning as well as to developing a sufficient risk management plan.
Because the data were collected in Vietnam, there may be limitations in providing applications for mega construction projects in other countries. Thus, future studies should be extended and adapted by taking into consideration the local and cultural conditions of the host environment, which may lead to changes in the risk structure and associated critical risks affecting infrastructure project cost and time.
