Integrating Risk Assessment and Scheduling in Highway Construction: A Systematic Review of Techniques, Challenges, and Hybrid Methodologies

Abstract

This study presents a comprehensive review of risk assessment and scheduling techniques in highway construction, addressing the complex interplay between uncertainty, project planning, and decision-making. The research critically reviews key risk assessment methods, including Probability–Impact (P-I), Monte Carlo Simulation (MCS), Fuzzy Set Theory (FST), and the Analytical Hierarchy Process (AHP), alongside traditional scheduling approaches such as the Critical Path Method (CPM) and the Program Evaluation and Review Technique (PERT). The findings reveal that, although traditional methods like CPM and PERT remain widely used, they exhibit limitations in addressing the dynamic and uncertain nature of construction projects. Advanced techniques such as MCS, FST, and AHP enhance decision-making capabilities but require careful adaptation. The review further highlights the growing relevance of hybrid and integrated approaches that combine risk assessment and scheduling. Bayesian Networks (BNs) are identified as highly promising due to their capacity to integrate both qualitative and quantitative data, offering potential for greater reliability in risk-informed scheduling while supporting improvements in cost efficiency, schedule reliability, and adaptability under uncertainty. The study outlines recommendations for the future development of intelligent, risk-based scheduling frameworks suitable for industry adoption.

1. Introduction

The construction industry is inherently complex and involves multiple uncertainties that can significantly impact project outcomes. Risk assessment and scheduling play a crucial role in ensuring project efficiency, reducing cost overruns, and preventing delays [1]. The growing complexity of modern construction projects, coupled with uncertainties in cost, time, and resource allocation, necessitates a structured and integrated risk assessment framework to enhance decision-making in project management. Aven [2] discusses the necessity of a comprehensive approach to risk management, integrating both qualitative and quantitative methods to address the complexities and uncertainties inherent in projects. He highlights the importance of understanding and managing these uncertainties to improve decision-making processes

Risk management is a crucial aspect of construction project planning, as effective strategies help mitigate unforeseen disruptions that could lead to severe financial and operational consequences. While Vose [3] extensively discusses risk analysis and management in various industries, his insights on quantitative methods remain valuable for assessing uncertainties in construction projects. Effective risk assessment not only identifies potential threats but also enables construction managers to develop proactive mitigation strategies that improve project performance [4].

Effective construction scheduling techniques are essential for optimizing resource allocation and maintaining workflow continuity, particularly in complex and uncertain project environments. Kaplan & Garrick [5] emphasize the importance of a structured risk analysis framework in decision-making, which can be applied to enhance scheduling reliability. Hubbard [6] highlights the need for integrating quantitative risk-based approaches with advanced analytical methods, reinforcing the importance of data-driven decision-making in risk management. The principles outlined can be applied to improving the resilience and efficiency of construction project schedules.

The research encompasses an extensive review of qualitative and quantitative risk assessment methods including traditional scheduling techniques, such as PERT and CPM. This study further highlights the impact of integrating risk-based decision-making with optimization algorithms, enhancing cost-efficiency and schedule reliability.

In support of a systematic approach to risk management, the Project Management Institute (PMI) outlines a widely adopted six-step risk management framework that serves as a foundation for many construction management strategies. These steps include (1) Risk Management Planning, (2) Risk Identification, (3) Qualitative Risk Analysis, (4) Quantitative Risk Analysis, (5) Risk Response Planning, and (6) Risk Monitoring and Control. This structured sequence ensures that risks are not only identified but also evaluated, prioritized, and continuously monitored throughout the project lifecycle [7]. To visually summarize this progression, a simplified version of the risk management process adapted from Ashley et al. [8] is presented in Figure 1. Figure highlights the decision-making flow between qualitative and quantitative analysis steps, illustrating the iterative and responsive nature of effective risk management.

2. Risk Assessment and Scheduling Techniques for Highway Construction Projects

While numerous risk assessment techniques exist, this section focuses on 13 qualitative and 13 quantitative methods that are widely used and applicable across different stages of risk management [1,9,10,11].

2.1. Qualitative Risk Assessment

Qualitative risk assessment techniques are subjective methods that rely on expert judgment and descriptive evaluations to identify and prioritize risks. These techniques are particularly valuable in the early stages of construction projects, where data are scarce, and the focus is on understanding risks that could affect project timelines, costs, or quality [1]. Heldman [12] provides insights into various risk management techniques, including qualitative methods that are particularly useful in the initial phases of projects. These techniques are widely applied in road and highway construction projects, as demonstrated by various studies summarized in

Table 1. Qualitative Risk Assessment Techniques.

Technique	Description	Studies Applying in Road/Highway Project Risk Assessment
Risk Matrix	Categorizes risks by likelihood and impact.	FHWA [8], Karaman et al. [13], Lv et al. [14]
SWOT Analysis	Analyzes Strengths, Weaknesses, Opportunities, and Threats.	Larsen et al. [15]
Checklists	Predefined list of risks based on experience.	FHWA [8], NCDOT [16], Le et al. [17]
Delphi Method	Iterative expert-based risk identification.	Lv et al. [14], Wu et al. [18], Mohamed [19], Fathi & Shrestha [20], Pratama et al. [21]
Expert Judgment	Leverages expert opinions for risk prioritization.	Karaman et al. [13], Lv et al. [14], Pratama et al. [21], Ariyanto et al. [22], Sharma & Trivedi [23], Gain & Mishra [24], Koulinas et al. [25], Senić et al. [26]
Brainstorming	Group-based generation of risks.	FHWA [8], NYSDOT [27]
Cause-and-Effect Diagram	Identifies root causes of risks.	* Oktaviani et al. [28]
Root Cause Analysis	Identifies fundamental causes of risks.	Al-Zwainy & Amer [29], Ahmed & Rezouki [30]
Scenario Analysis	Explores hypothetical risk scenarios.	Salling [31]
Failure Mode and Effects Analysis	Evaluates potential failure modes and impacts.	Lv et al. [14], Sharma & Trivedi [23], Gain & Mishra [24]
Risk Categorization	Group risks into categories for analysis.	Ariyanto et al. [22], Koulinas et al. [25], Senić et al. [26]
Risk Breakdown Structure	Organizes risks hierarchically.	El-Sayegh & Mansour [32], Jeon et al. [33]
HAZOP (hazard and operability study)	Structured evaluation of hazards.	Hanum et al. [34]

* The application of the Cause-and-Effect Diagram in highway construction risk assessment remains limited in existing literature. However, given its widespread use in general construction risk management, relevant studies from broader areas of construction are referenced to illustrate its applicability in highway projects.

. By facilitating systematic risk identification, qualitative approaches provide a foundation for informed decision-making in uncertain environments.

2.2. Quantitative Risk Assessment

Quantitative techniques are extensively used during the detailed planning and execution phases of construction projects to model uncertainties and optimize resource allocation. These data-driven methods apply probabilistic models and statistical analysis to assess risks numerically, enabling precise estimations of potential impacts on project cost, schedule, and performance. They are particularly effective in complex projects where risk factors can be systematically quantified and mitigated [1,3].

Various quantitative risk assessment techniques have been widely applied in road and highway construction projects, as summarized in

Table 2. Quantitative Risk Assessment Techniques.

Technique	Description	Studies Applying in Road/Highway Project Risk Assessment
Monte Carlo Simulation	Uses random sampling to model uncertainties and generate probability distributions.	Salling [31], Arifani & Prakoso [35], Kameshwar et al. [36]
Expected Monetary Value	Calculates the weighted average of possible financial outcomes based on probabilities.	Arifani and Prakoso [35]
Probability–Impact	Quantifies risks using combined probability and impact scores.	Hillson [4], Karaman et al. [13], Ward [17], El-Sayegh & Mansour [32], Williams [37], Charette [38], Han et al. [39], Baccarini & Archer [40], Chapman and Ward [41], Cagno et al. [42], Franke [43]
Fault Tree Analysis (FTA)	Analyzes system failures through a top-down logical diagram.	Davis-McDaniel et al. [44], Gacevski et al. [45], Chun [46], Chen et al. [47]
Event Tree Analysis (ETA)	Analyzes scenarios branching from an initiating event.	Hong [48]
Decision Tree Analysis	Branching diagram of decisions and risks.	Arifani & Prakoso [35], Kameshwar et al. [36]
Analytic Hierarchy Process	Prioritize risks using pairwise comparisons in a structured process.	Lv et al. [14], Wu et al. [18], Ariyanto et al. [22], Koulinas et al. [25]
Sensitivity Analysis	Examines how input changes affect project outcomes.	(Salling, 2013) [31]
Fuzzy Set Theory	Handles vague data using Fuzzy Logic to quantify risks.	Mohamed [19], Pratama et al. [21], Andrić et al. [49]
Cost Risk Analysis	Quantifies variability in project budgets to estimate cost overruns.	Baek et al. [50], Molenaar [51]
Bayesian Networks	Model dependencies and calculates conditional probabilities using Bayes’ theorem.	Mohamed [19], * Trucco et al. [52]
Network Theory	Model dependencies and risks in interconnected systems.	Ahmed et al. [53], Deng et al. [54], Sasidharan et al. [55]
Markov Chains	Uses probabilistic models to predict system transitions based on current state conditions.	Obare & Muraya [56], Besenczi et al. [57], Salman, & Gursoy [58]

* The application of Bayesian Belief Networks (BBNs) for modeling organizational factors in highway construction risk assessment remains limited in the current literature. However, given their proven effectiveness in other safety-critical fields, the study by Trucco et al. (2008) [52], which applied BBNs in maritime transportation risk analysis, is referenced here to illustrate the method’s potential relevance and adaptability to highway projects.

, which outlines key methods along with their descriptions and corresponding studies that have implemented them.

2.3. Traditional Scheduling Techniques

A comprehensive understanding of the methods employed by planners (as presented in

Table 3. Traditional Scheduling Techniques.

Technique	Description	References
CPM	Identifies the longest path to determine minimum duration.	Wingate [59], Bordley & Tennent [60], Mohammed et al. [61], Aras & Surkis [62]
PERT	Uses three-point estimates for probabilistic durations.	Aras & Surkis [62], Onifade [63], Grunow [64]
Gantt Charts	Visual bar charts showing tasks, durations, and progress.	Kumar & Krishnamoorthi [65], Herbsman [66], Harris & Evans [67]
Line of Balance (LoB)	Graphical method for repetitive tasks ensuring continuity.	Khachi [68], Hegazy [69], Vargas & Moreira [70], Lutz & Halpin [71], Arditi et al. [72], Tongthong [73]
Precedence Diagramming Method (PDM)	Graphically represents task dependencies.	Romadhona et al. [74], Hambatan [75], Novitasari et al. [76]

) is of paramount importance in construction project management. By examining the specific approaches utilized, it becomes possible to identify the underlying strategies and techniques that contribute to the effective development and management of construction project schedules. This understanding enables a thorough evaluation of the feasibility and efficacy of the different methodologies, highlighting their inherent strengths and weaknesses. Such insights are essential for refining and optimizing project planning processes, supporting informed decision-making, and ultimately enhancing the performance and success of construction endeavors.

In addition to the techniques highlighted in Table 3, other scheduling methods are employed in highway and road construction.

Heuristic Scheduling provides quick, rule-based solutions for resource-constrained projects but is more suited for simple or time-sensitive situations. The study by Rubén Ruiz [77] provides an overview of heuristic methods used in production scheduling. It emphasizes how heuristic approaches can generate high-quality, near-optimal solutions efficiently, bridging the gap between scheduling theory and practice. In their study, Davis & Patterson [78] compared eight heuristic scheduling rules to optimal solutions. They focused on minimizing project duration under multiple limited resource constraints, highlighting the effectiveness of heuristic methods in practical scheduling scenarios.

The Linear Scheduling Method (LSM) is mainly applied to linear projects like highways and pipelines, particularly those involving repetitive and sequential activities. Harmelink [79] developed an LSM specifically for highway construction, highlighting its advantages over the CPM in handling linear and repetitive activities. His research introduced a graphical approach to identifying controlling activities, rate float analysis, and automated scheduling using Auto LISP, demonstrating LSM’s effectiveness in improving project planning and resource continuity. Recent research by Yogesh & Rao [80]. Illustrates potential advantages of LSM in optimizing project scheduling for highway construction, demonstrating how simulation modeling can improve production rate estimation and project duration forecasting. Their findings emphasize that LSM helps reduce resource idle time and enhance overall efficiency compared to traditional scheduling approaches. Earlier, Johnston [81] provided a critical assessment of LSM, emphasizing its applicability to linear infrastructure projects such as highways, railways, and pipelines. His study laid the foundation for evaluating LSM’s potential in highway construction and maintenance, reinforcing its value as a scheduling tool in the industry.

Additionally, Activity-On-Arrow (AOA) uses arrows for tasks and nodes for milestones. While AOA is a type of CPM representation, it has largely been replaced by PDM-based CPM in modern project scheduling due to its greater flexibility and efficiency [1].

3. Critical Review of Risk Assessment and Scheduling Techniques in Construction Project Management

The use and selection of risk assessment techniques in construction and industrial projects can vary depending on a variety of factors, including the needs of the organization, the type and complexity of the project, and the industry or sector involved.

Thaheem & De Marco [82] conducted a global survey to examine the adoption of project risk management techniques within the construction industry. Their findings reveal that practitioners predominantly rely on methods such as risk probability and impact assessment (66%), risk categorization (49%), and P-I matrix (35%). These approaches appear to align with the intuitive process of associating risks with their likelihood of occurrence and potential consequences, particularly during early planning stages. Additionally, the study highlights widespread use of expert judgment (64%) and interviewing (44%), while more analytically intensive techniques, such as modeling and simulation, expected monetary value, sensitivity analysis, and probability distributions, are employed less frequently, averaging around 30% usage. Notably, brainstorming was also cited by respondents as a commonly used tool, whereas 2% reported not applying any formal technique at all.

Renuka et al. [83] discussed various risk assessment models used in construction projects (Figure 2), highlighting that the AHP, Logistic Regression (LR), Fuzzy Logic and MCS models are among the most widely utilized. They conclude that due to its systematic approach to structuring risk assessment problems through a hierarchical framework, AHP proves to be an effective multi-criteria decision-making tool and is frequently employed among different assessment models.

Škrtić & Horvatinčić [84] conducted a comprehensive comparative analysis of commonly used risk assessment methods, including PERT, MCS, Decision Tree, Sensitivity Analysis, Probability Analysis, Brainstorming, Delphi, and AHP, highlighting their respective strengths and limitations. The study emphasizes that the suitability of a particular method largely depends on the project’s type, complexity, available data, and the specific setting in which it is used. Rather than recommending any single approach as universally optimal, the authors argue for the importance of aligning method selection with project-specific conditions and decision-maker needs. Their findings underscore that a clear understanding of each technique’s application conditions, as well as its limitations, enables more effective and tailored risk management strategies across varying project environments.

Ouache et al. (2014) [85] provided a broad review of technical risk assessment techniques across industrial systems. Their work presents a diverse set of methods, including Risk Matrices, AHP, Decision Trees, Neural Networks, Regression Analysis, Event Tree and Fault Tree Analyses, and MCS, each evaluated for its strengths and limitations. The review echoes a central theme across literature, while numerous tools are available, their effectiveness is contingent upon careful selection tailored to the specific project environment and data conditions.

Doubravský & Doskočil [86] compared project scheduling techniques, specifically focusing on the PERT and MCS. They discussed the advantages and disadvantages of each method and concluded that the choice of technique should be based on the specific characteristics of the project and the preferences of the decision-makers, as no single method is universally superior. This suggests that the outcome varies depending on the method, which implies that method selection matters. While they do not assert a universally superior method, their findings show that PERT and MCS can yield significantly different results, indicating that method selection plays a critical role in project planning outcomes.

Nevertheless, for the purposes of this study, a synthesized review of the most widely cited and practically relevant methods is conducted, focusing on their strengths, limitations, and applicability within construction project management. The following subsection presents a critical appraisal of commonly used approaches, both risk assessment and scheduling, as outlined in Table 1, Table 2 and Table 3. The methods reviewed include the following:

• P-I;
• MCS;
• FST;
• AHP;
• CPM;
• PERT.

3.1. Probability–Impact (P-I)

P-I is the technique most utilized by researchers and analytics for a comprehensive assessment of risk impact. It is an approach that considers the impacts of a risk on project objectives, for example, averaging the likelihoods and the impacts of a risk on project cost, time, and quality, and calculating the overall risk Baccarini & Archer [40].

Various authors have discussed the limitations of the P-I risk model and suggested improvements by extending it into a three-dimensional risk model or proposing alternatives:

• Probability–Impact–Predictability as recommended by Charette [38], Williams [37];
• A six-step ‘minimalist’ approach for estimating uncertainty, see Chapman & Ward [41];
• The degree of exposure of risk, see Jannadi & Almishari [87];
• Risk controllability, see Cagno et al. [42];
• Risk manageability, see Dikmen et al. [88];
• Significance–Probability–Impact, see Han et al. [39].

According to Ouache et al. [85], Risk Matrix offers a simple and intuitive way to visualize risk by categorizing events according to their probability and impact. It aids in decision-making by increasing risk visibility and is especially useful as a screening tool for identifying events that may require further quantitative assessment. However, the method is inherently qualitative, which limits its analytical depth. Notably, it cannot be used independently to determine risk premiums or support multi-criteria decision-making. Additionally, its categorical approach makes it less effective for allocating resources based on precise risk levels and can misrepresent severity when consequences are uncertain.

3.2. MCS

MCS is a probabilistic technique used for simulating events. A notable foundational overview of MCS has been provided by Vose [89], offering valuable insights into its theoretical underpinnings.

Ross [90], Rubinstein et al. [91], Fishman [92], and Owen [93] provide a comprehensive introduction to the MCS, discussing its applications in various fields, including finance, engineering, physics, and statistics, as well as its advantages and limitations, and its implementation in practice. A number of authors have presented a methodology based on MCS; it is used to estimate project cost, to assess project cost and duration, to aggregate probabilities of cost-related, duration-related risk factors, etc., i.e., in one word, to estimate the outcome of an uncertain process, see Guyonnet et al. [94], Oztas & Okmen [95], Molenaar [51], Baudrit et al. [96], Dikmen et al. [88], and Sadeghi et al. [97].

However, despite its widespread adoption, MCS has notable limitations. Ferson [98] identified four key shortcomings of classical MCS methods: their high demand for empirical data, inability to represent partial ignorance under frequentist probability, limitations in estimating exceedance risks, and ineffectiveness in solving backcalculation problems like those encountered in remediation planning. These limitations can lead to incorrect conclusions in complex risk assessments.

While MCS has proven effective in modeling uncertainty through probabilistic analysis, it has limitations in handling the multi-dimensional nature of project risk. As noted by Oztas & Okmen [95], MCS does not natively support the integration of multiple risk dimensions such as cost, time, and quality. To address this, multi-criteria decision analysis techniques like the Analytic Hierarchy Process (AHP) have been proposed to manage and prioritize complex, interrelated risk factors. At the same time, traditional deterministic scheduling methods such as PERT are known to produce overly optimistic projections by relying on expected values derived from three-point estimates, without propagating uncertainty throughout the project network. The limitations of such approaches are evident in a case study by Oztas and Okmen, where a fixed-price design–build project with a contractual duration of 131 days was estimated deterministically at 162 days using traditional methods. However, the actual completion took 190 days, aligning more closely with the 80–90th percentile predictions from MCS. Similarly, the project’s final cost of TRY 130.6 billion exceeded both the contractor’s bid (TRY 84.5 billion) and the MCS 50th percentile estimate (TRY 117.4 billion), demonstrating MCS’s superior capacity to capture cost-related uncertainty. These findings underscore the optimistic bias inherent in methods like PERT and the value of combining probabilistic tools like MCS with multi-criteria approaches such as AHP to achieve a more comprehensive and realistic assessment of project risk.

According to Ouache et al. [85], the MCS method is recognized for its robustness in modeling uncertainty using probability distributions and repeated random sampling. It effectively supports the evaluation of cost, schedule, and technical risks, producing probabilistic outputs that inform decision-making. However, the method is highly dependent on the accuracy of input data and correlation assumptions. Misrepresenting variable relationships can compromise the validity of results. Furthermore, because MCS relies on randomly sampled scenarios, it may overlook low-probability but high-impact risks unless explicitly modeled.

Dorp & Duffey [99] critique the traditional application of MCS in project risk analysis, particularly the common assumption of statistical independence among activity durations. They argue that this assumption is often unrealistic, as many activities share common risk factors, leading to positive dependence between activity durations. Neglecting these dependencies can result in underestimating the uncertainty in project schedules and costs. To address this issue, they propose a methodology that models and quantifies positive dependence between uncertain activity durations, enhancing the accuracy of risk assessments in project networks. This limitation is further illustrated in a case study by Dorp and Duffey, where failure to account for such dependencies resulted in significant underestimation of schedule risk. In a small shipbuilding project, ignoring activity correlations led to a more than 5% underestimation of the 95th percentile completion time; in a larger 250-activity network, the discrepancy approached 10%. These results highlight how rigid MCS inputs, similar to the deterministic optimism observed in PERT, can severely misrepresent project risk unless dependency structures are explicitly modeled.

While MCS has been widely applied in project scheduling and risk assessment, certain underlying assumptions may limit its effectiveness in capturing complex project dynamics. As highlighted in previous research, many deterministic and probabilistic scheduling methods, including traditional MCS applications, often assume that activity duration is independent and that the overall project duration follows a normal distribution. These assumptions may not always align with real-world project conditions, where activity durations can be correlated due to shared risk factors. To address this limitation, Wang & Demsetz [100] introduced the NETCOR model, which integrates correlation effects into network schedules to provide more refined sensitivity analyses. This suggests that while MCS remains a powerful tool, its standard implementation may not fully capture interdependencies among activities unless explicitly modified to account for correlated uncertainties. Such advancements underscore the importance of refining simulation techniques to enhance the accuracy of project risk assessments.

3.3. FST

FST is commonly applied in situations where data are imprecise or uncertain, making it particularly valuable in risk analysis and decision-making in complex and uncertain situations, see Zimmermann [101].

FST is a widely used decision-making technique applied across various domains, particularly in risk assessment and multi-criteria decision-making within construction projects. Numerous studies have demonstrated its effectiveness in addressing uncertainties and optimizing decision-making in complex environments, such as highway and road construction projects.

While numerous books provide an in-depth introduction to FST, the focus here is to highlight key studies that have applied FST in risk assessment and multi-criteria decision-making within the construction industry. Researchers have continuously refined and developed fuzzy-based methodologies to enhance risk evaluation and decision-making processes in construction projects. These advancements have contributed to more effective and reliable frameworks, particularly within highway and road construction projects, see Dikmen et al. [88]; Kangari & Riggs [102]; Paek et al. [103]; Wirba et al. [104]; Tah & Carr [105]; Thomas et al. [106]; and Jin & Doloi [107].

Kangari & Riggs [102] utilized linguistic variables within FST to assess construction risks, providing a framework for qualitative risk evaluation. Paek et al. [103] applied FST to price construction risks, demonstrating its effectiveness in quantifying uncertainties in cost estimation. Tah & Carr [105] proposed a risk assessment model using fuzzy logic, highlighting its applicability in evaluating complex project risks. Thomas et al. [106] applied fuzzy fault tree analysis within their proposed framework, demonstrating the utility of FST in assessing and managing risks in BOT road projects.

While FST has been widely applied in construction and project risk assessment due to its ability to handle ambiguity and uncertainty in decision-making, recent scholarly discourse has raised important concerns about its theoretical and practical limitations.

In a critical review, Sotoudeh-Anvari [108] challenges the methodological foundations upon which many fuzzy approaches are built. He argues that despite its popularity, fuzzy methods often rest on mathematically incorrect assumptions, particularly when fuzzy arithmetic is applied without proper validation. A further concern raised in the study is the heavy dependence on subjective judgments during the formulation of membership functions, an aspect that can undermine the consistency and repeatability of model results. The paper also points out that the increasing sophistication of newer fuzzy methodologies, such as type-2 fuzzy sets and hesitant fuzzy approaches, while theoretically robust, often results in computational inefficiencies that hinder their practical implementation. These concerns suggest that while FST remains a powerful conceptual tool, its application should be approached with caution and a critical understanding of its limitations, particularly in domains where decision accuracy and methodological robustness are essential.

Toth [109] offers an early and critical perspective on the foundations of FST, arguing that key misconceptions have influenced its development from the beginning. His work challenges conventional interpretations of fuzziness, particularly those based on Hisdal’s TEE model, and proposes an alternative grounded in operational and set-theoretical reasoning. Through a series of examples, he demonstrates how traditional fuzzy models may fall short in representing the true nature of ambiguity, calling for a reassessment of core assumptions within the theory. Guth [110] explored both the practical benefits and inherent limitations of applying fuzzy logic within artificial intelligence systems, based on the development of a prototype expert system for reactor monitoring. Drawing on real-world insights from engineers at Oak Ridge National Laboratory, the study found that while fuzzy logic allowed for nuanced reasoning under uncertainty, its implementation revealed challenges in integrating expert knowledge, structuring rule-based inference, and maintaining reliability in complex technical environments.

While FST is recognized for handling uncertainty better than classical methods, Dey [111] notes that such conceptual models often rely on subjective expert inputs and may lack precision when multiple risk factors interact. Thus, they may not offer a sufficiently robust framework for complex risk management. Similarly, he emphasizes that classical probabilistic techniques such as MCS require detailed quantitative data, which are rarely available during the early stages of project planning, limiting their practical application.

3.4. AHP

AHP is a reliable, rigorous, and robust method for eliciting and quantifying subjective judgments in multi-criteria decision-making (MCDM), see Tavana [112]. AHP is a decision-making method that is widely used in many investigational studies. Risk assessment methodology based on AHP has been presented by many researchers, including Lv et al. [14], Wu et al. [18], Ariyanto et al. [22], Koulinas et al. [25], Mustafa & Al-Bahar [113], Ogunlana [114], Hastak & Shaked [115], Dikmen & Birgonul [116], Zayed et al. [117] and Taherdoost [118].

Although the application of AHP in road construction continues to gain traction and offers structured support in evaluating and ranking project risks, it still faces notable limitations. Dey [119] recognized the strengths of AHP in integrating both qualitative and quantitative risk information and facilitating structured group decision-making. However, he also acknowledged that AHP may fall short in capturing the vagueness and ambiguity inherent in expert judgment. In his proposed framework combining AHP with a risk map, he noted the method’s limitations in addressing uncertainty in early-stage project evaluations and suggested that the incorporation of fuzzy logic could provide a more suitable and flexible alternative for modeling imprecise risk inputs.

According to Ouache et al. [85], AHP is praised for its ability to structure complex decision problems hierarchically, enabling both subjective and objective assessments across multiple criteria. It is especially beneficial in scenarios requiring a systematic and logical framework for prioritizing risks. Nonetheless, AHP has limitations, including the rigidity of its scale, difficulty in assigning precise numerical values in pairwise comparisons, and sensitivity to changes requiring recalculation if new criteria are added. Its outputs assist in comparative decision-making but do not yield direct estimates of risk premiums or monetary values.

A related concern has been raised by Zhang & Zou [120], who identified that traditional AHP-based models may struggle with ambiguity and subjectivity in expert judgments when assessing risks in joint venture construction projects in China. To address these limitations, they integrated fuzzy set theory into the AHP framework, creating a Fuzzy Analytical Hierarchy Process (FAHP) model. To address these shortcomings, they proposed integrating FST to enhance the model’s capacity to manage uncertainty and support more informed decision-making.

3.5. CPM (Deterministic Scheduling Approach)

CPM has long been recognized as a dominant approach for project scheduling and monitoring, as highlighted by prominent scholars including Wingate [59], Bordley & Tennent [60], Mehany & Grigg [61], Aras & Surkis [62], and Galloway [121].

Bordley & Tennent [60] explored the feasibility of applying the CPM to highway right-of-way acquisition processes. Their study, initiated by the U.S. Bureau of Public Roads, aimed to improve efficiency in managing the complex and time-sensitive task of acquiring thousands of individual land parcels. Through the development and pilot testing of a prototype CPM network, the authors demonstrated that CPM could enhance planning, coordination, and control across multiple divisions involved in right-of-way activities. The model facilitated better communication, streamlined operations, and exposed procedural inefficiencies, ultimately offering a systematic approach to managing one of the most resource-intensive phases of highway development.

Aras & Surkis [62] provided one of the early integrated treatments of PERT and CPM techniques, emphasizing their fundamental network-based structure and practical application in project management. By illustrating how nodes and arrows represent events and activities respectively, they highlighted the utility of these models in mapping the logical sequence and interdependencies of project tasks. Their work also addressed emerging developments in cost–time optimization and demonstrated the applicability of both techniques through a simplified highway construction case study. Notably, the integration of work breakdown structures and account code systems further reinforced the relevance of CPM and PERT for structured planning and control in infrastructure projects.

Wingate [59] presented an early and detailed evaluation of the CPM in road construction projects, highlighting both its strengths and limitations. The report includes multiple case studies and reveals that, while CPM facilitates logical planning, resource allocation, and progress tracking, its practical application often requires adaptation to project conditions. Notably, the study found that rigid, large-scale CPM networks were often less practical than simpler, manually computed diagrams, particularly in environments subject to high uncertainty such as road construction. The report underscores that the effectiveness of CPM depends heavily on the quality of input data and flexibility in implementation.

In his comprehensive survey, Galloway [121] examines the construction industry’s utilization of CPM scheduling, highlighting its widespread adoption and significance in project management, while also identifying areas for improvement. Despite CPM’s long-standing presence since the 1950s, the study revealed that its adoption remains inconsistent, with no universal standards guiding its development or use. The findings highlight a lack of consensus among stakeholders regarding the qualifications required for scheduling personnel and the need for best practices or standardization. Galloway [121] emphasizes that enhancing consistency and clarity in the application of CPM could significantly improve scheduling practices and overall project management within the industry.

Mehany & Grigg [61] underscore the foundational role of the CPM in delay claim analysis within highway construction projects. While CPM remains the most widely adopted scheduling framework, the authors highlight significant inconsistencies in how it is applied across various forensic delay analysis methods. Through a detailed case study of a highway bridge project, they demonstrate that the use of different CPM-based techniques, such as As-Planned vs. As-Built, Impacted As-Planned, Collapsed As-Built, Schedule Window Analysis, and Time Impact Analysis, can lead to divergent conclusions for the same delay scenario. This variability reveals the potential for manipulation and dispute, emphasizing the need for standardized CPM-based practices. Their findings advocate for the use of Time Impact Analysis (TIA) as a best-practice model due to its real-time integration with CPM networks and its enhanced reliability in capturing schedule dynamics, delay causality, and concurrency.

Senior [122] questioned the widespread use of CPM, highlighting the paradox between its institutional prominence and practical underuse. Drawing on Action Theory, the study suggests that this disconnect may stem from fundamental shortcomings in the method itself, including its limited adaptability to the dynamic realities of construction sites. Senior argues that unless these drawbacks can be addressed meaningfully, continued reliance on CPM may be misguided, and alternative approaches such as Lean Construction should be considered. The study underscores the urgency of re-evaluating the method’s practical value, particularly given the inefficiency of developing detailed plans that ultimately go unused in the field.

Bhatt et al. [123] investigated the practical challenges of CPM implementation in real-world construction projects. While CPM remains a widely used project scheduling tool, the authors argue that it often fails to align with on-site realities due to its rigid, deterministic structure. The study emphasizes a disconnect between upper management’s pre-planned schedules and the dynamic, variable conditions encountered by field teams. Drawing on Action Theory, specifically the contrast between Deliberative and Situated Action perspectives, the paper highlights the inefficiency of strictly pre-defined plans and the need for adaptability in project execution. A case study further illustrates how CPM’s inability to accommodate unexpected changes, such as unforeseen ground conditions, can lead to cost overruns, delays, and reputational damage. The authors suggest enhancements such as involving field personnel in planning, limiting initial scheduling detail, and expanding short-term task granularity to improve CPM’s effectiveness.

Ragel et al. [124] critically examined the limitations of the PERT/CPM approach within construction management, particularly in large-scale projects. Their findings highlight that while PERT/CPM remains a valuable tool for planning and scheduling, its effectiveness is constrained by several practical challenges. These include the method’s inability to accommodate sudden changes in project execution, the high demand for frequent updates and revisions, and the complexity of managing extensive networks in large projects. Moreover, the authors note that the deterministic assumptions of PERT/CPM often fail to reflect the uncertainty and interdependencies present in real-world construction environments, thereby calling for more adaptive and resource-sensitive planning tools.

To address the limitations of traditional CPM in resource-constrained environments, Lu & Li [125] developed the Resource–Activity Critical-Path Method (RACPM). This method explicitly integrates the resource dimension into project scheduling, alongside time and activity, allowing for more realistic planning in construction projects. By redefining start/finish times and floats as resource–activity attributes and incorporating both resource availability and technological constraints, RACPM improves the accuracy of project duration estimates and supports more practical, resource-sensitive scheduling. Through illustrative examples, including a footbridge construction case, the authors demonstrate the method’s capacity to align resource allocation with construction sequencing, offering a more effective alternative to conventional CPM scheduling frameworks.

3.6. PERT (Probabilistic Scheduling Approach)

To address the limitations of deterministic scheduling methods like CPM, PERT was introduced as a probabilistic tool capable of incorporating risk and uncertainty into project scheduling. Zhong & Zhang [126] proposed an improved method for calculating path float in PERT networks to better manage uncertainty in construction scheduling. They argue that traditional float calculation methods may mislead site managers, potentially increasing the risk of project delays. By introducing a revised approach for evaluating non-critical path float based on required completion probability and duration, the authors provide a more accurate tool for project planners. Their findings suggest that the new method offers more consistent and informative guidance for managing schedule risks in uncertain construction environments.

Malcolm et al. [127] introduced the PERT model as a probabilistic scheduling tool developed for the U.S. Navy’s Polaris missile program. Designed to manage large-scale research and development efforts, PERT introduced a probabilistic approach to project scheduling, incorporating uncertainty through three-point time estimates, optimistic, most likely, and pessimistic for each activity. The paper established the statistical foundation of PERT and demonstrated its effectiveness in improving planning, control, and decision-making under uncertainty, marking a major advancement in project management methodology. Later works, such as those by Miller [128], Van Slyke [129], and Kuklan et al. [130], contributed to its refinement and broader application, while more recent studies have extended its capabilities in modeling uncertainty in project scheduling.

PERT is used in road construction projects, particularly for identifying task dependencies, calculating critical paths, estimating project durations, and supporting resource planning, see Aras & Surkis [62], Onifade et al. [63] and Grunow [64].

Grunow [64] explored the practical application of the PERT within highway management, particularly in the Washington State Highway Department project programming. The study demonstrated how PERT could model complex sequences of events and activities across multiple projects, enhancing schedule visibility, resource allocation, and decision-making under uncertainty. Despite initial challenges in implementation, the research emphasized PERT’s potential as a valuable planning and control tool in managing multi-project highway programs.

Additionally, the application of PERT in highway construction planning has been exemplified in early integrated studies, such as that by Aras & Surkis [62], who demonstrated how probabilistic scheduling could be effectively incorporated into complex infrastructure projects alongside CPM.

Onifade et al. [63] conducted a survey-based study among professionals in the Ogun State Ministry of Works and Transport to assess the application of the PERT in road construction management. Their findings affirm that PERT serves as a valuable planning and control tool, enabling project teams to achieve defined objectives within specified time and budget constraints by optimizing the use of available resources.

Despite its advantages, PERT is not without limitations. Scholars have criticized it for producing overly optimistic or risky time estimates, which may create a false sense of confidence in project outcomes. Additionally, PERT can be sensitive to changes in activity precedence or sequence, often lacking the flexibility to adapt to evolving project conditions. To overcome these drawbacks, several researchers have proposed hybrid approaches that combine PERT with MCS, aiming to improve the robustness and precision of schedule forecasting. Notable examples include the works of Van Slyke [129], Kirytopoulos et al. [131], Ganame & Chaudhari [132], and Hendradewa [133], who demonstrated that such integrations enhance the reliability of risk-informed project scheduling.

In his seminal work, Van Slyke [129] introduced the application of MCS methods to address the inherent uncertainties in PERT networks. He highlighted that while PERT offers a structured approach to project scheduling, its deterministic nature can lead to overly optimistic time estimates. By incorporating MCS, Van Slyke demonstrated a more robust analysis of project completion times, accounting for variability and uncertainty in activity durations.

The study by Kirytopoulos et al. [131] compared traditional PERT with MCS across 20 small- and medium-sized construction projects. The authors found that MCS provided more accurate and reliable estimates of project durations, emphasizing the importance of selecting appropriate distributions and utilizing historical data. Their findings suggest that integrating MCS with PERT can mitigate some of PERT’s limitations, particularly its sensitivity to activity sequence changes and its potential for generating optimistic forecasts.

Ganame & Chaudhari [132] highlight the limitations of relying solely on PERT, particularly its use of most likely durations that often fail to capture the impact of uncertainty in construction projects. To address this, they propose integrating MCS, demonstrating through a case study that MCS provides more realistic estimates of project completion time by accounting for variability in activity durations and enhancing the reliability of schedule risk analysis.

Hendradewa [133] critiques the deterministic nature of PERT, highlighting its limitations in accounting for uncertainties in construction schedules. To address this, the study integrates MCS with CPM-PERT, demonstrating that the combined approach provides more realistic and probabilistic estimates of project completion times, thereby enhancing the accuracy of risk-informed scheduling.

3.7. Summary

The diverse landscape of risk assessment and scheduling methodologies in construction project management reflects an evolving understanding of uncertainty, data complexity, and decision support needs. Each technique contributes uniquely to the field, yet no singular method can comprehensively address all dimensions of construction risk analysis.

Table 4. Comparative Overview of Key Risk Assessment and Scheduling Techniques.

Technique	Type	Strengths	Limitations
P-I Matrix	Qualitative	Simple, intuitive, widely used in early project phases	Over-simplified, lacks dynamic risk behavior modeling
MCS	Quantitative	Handles uncertainty probabilistically; effective for cost/duration estimation	Assumes independence among variables; computational load; correlation issues
FST	Hybrid (Quali–Quant)	Captures imprecision and ambiguity in expert judgment	Subjective membership functions; complex validation and implementation
AHP	Qualitative	Structured decision-making; supports pairwise comparison	Sensitive to bias; fails to capture ambiguity in judgment
CPM	Deterministic Scheduling	Well-established; clear sequence planning	Rigid, ignores uncertainty; lacks feedback mechanisms
PERT	Probabilistic Scheduling	Incorporates uncertainty with three-point estimation	Optimism bias; sequence sensitivity; assumes beta distribution

provides a synthesized overview of widely adopted risk assessment and scheduling techniques based on the critical review conducted in this section, outlining their types, strengths, and limitations. This comparison underscores the importance of selecting methods tailored to the specific characteristics and requirements of each project and illustrates how each approach supports decision-making under uncertainty.

The comparative analysis in Table 4 reveals that combining techniques such as MCS with PERT can enhance the realism of duration estimates by incorporating probabilistic variability. However, despite the strengths of individual methods, the integration of risk assessment techniques with scheduling frameworks remains limited. Traditional methods like CPM and PERT are often applied in isolation, reducing their effectiveness in the presence of uncertainty. These findings underscore the need for adaptive, data-driven approaches that dynamically align risk evaluation with scheduling decisions.

4. Integrated Approaches

Combining Risk Assessment and Scheduling: Examples of Hybrid Methodologies

This section presents a brief overview of selected examples illustrating the integration of risk assessment methodologies into scheduling problems. A more comprehensive analysis and in-depth methodological discussion covering both the integration aspects and the scheduling problems will be provided in a separate publication. This forthcoming study will specifically focus on the implementation and empirical validation of the proposed hybrid risk-scheduling model. It will demonstrate, through real-world or simulated case studies, how the model addresses the limitations of traditional methods (e.g., CPM, PERT, and conventional MCS) by effectively integrating qualitative expert judgment with quantitative analysis, particularly through Bayesian Networks and multi-criteria decision-making techniques. The study will also highlight how the developed model enhances predictive accuracy, adaptability, and decision support in highway construction projects, thereby substantiating its advantages over existing approaches. To effectively bridge the gap between qualitative insights and quantitative decision-making, Bayesian Networks (BNs) utilize Conditional Probability Tables (CPTs), which convert expert judgment into structured probabilistic models. This allows for the integration of subjective assessments with empirical data in a mathematically rigorous framework. Similarly, in the Analytic Hierarchy Process (AHP), expert inputs are structured through pairwise comparisons, with consistency ratios used to evaluate the logical soundness of judgments. Aggregation methods are then applied to ensure reliable and reproducible prioritization of risk factors. These methodological features enhance the transparency and validity of qualitative–quantitative integration within hybrid models.

The integration of risk assessment and scheduling methodologies has emerged as a transformative approach to addressing the complexities of construction project management. Hybrid methodologies, which combine qualitative, quantitative, and scheduling techniques, have shown significant promise in enhancing the robustness of project planning and execution. By synchronizing risk identification and analysis with dynamic scheduling models, these integrated approaches enable construction managers to proactively mitigate uncertainties and optimize resource utilization [1].

Razaque et al. [134] proposed an integrated methodology combining risk assessment and scheduling techniques to enhance project reliability. Their approach incorporates fuzzy Failure Mode and Effects Analysis (FMEA), MCS, and critical chain scheduling to identify and evaluate project risks, simulate schedule variability, and mitigate potential delays. The model also employs Adaptive Procedure with Density (APD) to refine buffer sizing and improve schedule resilience. By merging analytical risk modeling with dynamic scheduling, the framework enables more accurate forecasting and proactive risk control throughout the project lifecycle.

In his paper, Hulett [135] discusses how incorporating MCS into CPM allows for the quantification of uncertainties in activity durations and highlights potential delays with confidence intervals, thereby offering a more resilient scheduling framework.

Hubbard [6] critiques traditional risk management practices and advocates for the use of rigorous quantitative metrics over conventional qualitative methods to develop more predictive and empirically grounded risk frameworks. He emphasizes the importance of data-driven analysis to enhance decision-making under uncertainty.

Acebes et al. [136] introduced a method that integrates aleatory uncertainty into project scheduling to support more informed selection of baseline schedules. Rather than assuming all feasible schedules with equal duration are equally viable, their approach quantifies and compares the total risk associated with each schedule using MCS. By developing the Schedule Risk Baseline (SRB) and deriving the Schedule Risk Value (SRV), they demonstrate that schedules with identical durations can differ significantly in terms of risk exposure. This risk-informed selection framework enables project managers to choose schedules that maximize the probability of meeting project deadlines while accounting for the inherent variability in activity durations.

Isah & Kim [137] developed a stochastic multiskilled resource scheduling (SMSRS) model that integrates schedule risk analysis (SRA) with resource-constrained scheduling algorithms. By incorporating MCS and triangular probability distributions into a heuristic-based scheduling framework, their model enables the generation of more realistic schedules that account for both uncertainty in activity durations and limited resource availability. This integrated approach enhances decision-making under uncertainty and improves the robustness of construction project planning.

Some studies have explored the application of AHP in project scheduling and risk prioritization. For instance, Moselhi & Roofigari-Esfahan [138] present a method for compressing project schedules using AHP, considering multiple factors beyond cost to optimize project timelines. Dey [111] argues that most traditional tools evaluate risks in isolation and lack a mechanism to integrate risk analysis with project objectives, such as cost, schedule, and quality. He emphasizes the need for a decision-support system that aligns risk assessment with overall project goals. To overcome these shortcomings, he proposes a hybrid model that combines the AHP for prioritizing risks and Decision Trees (DT) for evaluating alternatives. This approach incorporates both qualitative and quantitative data, supports multi-risk analysis, and aligns decisions with project objectives.

While this section primarily focuses on the integration of risk assessment methodologies with scheduling techniques, it also includes references to studies that address the integration of multiple risk assessment methods independently of scheduling. These examples are included due to their methodological relevance and potential to inform future developments in risk-informed scheduling frameworks.

Some researchers attempt to combine the strengths of FST and AHP to assess the level of criticality of risk events in risky conditions of the road construction projects, assessing the relationship between impact and probability of occurrence, ranking risk factors, etc., see Zhang & Zou [120], Zeng et al. [139], Subramanyan & Sawani [140], Enny & Purba [141], and Tüysüz & Kahraman [142].

Zeng et al. [139] applied a fuzzy-based decision-making methodology to construction project risk assessment, combining FST and AHP to prioritize risk factors. Subramanyan & Sawant [140] assessed risk conditions in the Indian construction industry using a fuzzy AHP approach, identifying and prioritizing 93 risk factors based on expert input and statistical analysis to support risk mitigation strategies. Enny & Purba [141] conducted a literature review examining the application of FAHP in construction project risk assessment. By analyzing 30 academic sources, their study highlights FAHP’s effectiveness in evaluating complex and uncertain risk factors, particularly in capturing the subjectivity of expert judgment. The review concludes that technical risks are the most dominant risk category across construction projects, reinforcing the importance of systematic and structured risk prioritization using fuzzy-based methods. Zhang & Zou [120] proposed a fuzzy AHP model to evaluate risks in joint venture construction projects in China, addressing the need for a more structured, quantitative approach to support decision-making. Their model integrates expert judgment into a hierarchical framework, offering a practical method for assessing JV-related risks and demonstrating its usefulness through a real-world case study. Tüysüz & Kahraman [142] demonstrated the practical value of the fuzzy AHP method for evaluating project risks under uncertainty. By applying Chang’s extent analysis approach, they highlighted the method’s ability to incorporate expert judgment and manage ambiguity in complex decision-making environments. Their findings suggest that fuzzy AHP offers a structured and effective framework for prioritizing risks in project management.

Regarding the reliability of combining FST and AHP, while the combination of both techniques can be considered to be the best available option for handling risk analysis, the following studies have raised concerns (Zhu et al. [143], Wang et al. [144], Shapiro & Koissi [145]).

Zhu et al. [143] raised concerns about the extent analysis method in fuzzy AHP, pointing out issues such as inconsistency in pairwise comparisons and the method’s sensitivity to the choice of fuzzy numbers, particularly due to violations of fundamental axioms such as reciprocity in pairwise comparisons when using triangular fuzzy numbers. They highlighted the absence of a robust mechanism for consistency checking, which undermines the credibility of the decision-making process. Additionally, Zhu and Jing pointed out that the method’s outcomes are highly sensitive to the selection of fuzzy numbers and that its application often adds complexity without offering clear advantages over traditional AHP. These limitations led them to question the overall reliability of fuzzy AHP, especially in scenarios requiring high decision accuracy.

Wang et al. [144] critically revisited Chang’s [146] extent analysis method for fuzzy AHP and demonstrated that it may not accurately reflect the true priority weights derived from fuzzy comparison matrices. Their analysis revealed that this method can misrepresent the relative importance of decision criteria or alternatives, potentially leading to flawed decisions. Through illustrative examples, they emphasized the risk of misapplication and urged caution in using extent analysis for fuzzy AHP problems, especially in decision-making processes that require reliable prioritization.

Similarly, Shapiro & Koissi [145] raised several concerns about the use of fuzzy AHP, particularly the extent analysis method introduced by Chang [146]. They noted that the method’s normalization approach can violate fundamental assumptions of AHP and may lead to zero weighting of important criteria, potentially excluding relevant factors from the decision-making process. Additionally, they questioned the theoretical soundness of the method, pointing out that it can overlook essential decision data and lead to unreliable outcomes in real-world risk assessment scenarios.

In recent years, BNs have attracted significant attention from researchers due to their versatility and applicability in project risk management. Key studies by Fan & Yu [147], Trucco et al. [52], Khodakarami et al. [148], and Mohamed [19] have explored the potential of BNs as a robust methodological framework.

Khodakarami et al. [148] present a novel approach that integrates BNs into the CPM for project scheduling. This integration allows for the modeling of uncertainty and causality within project schedules, enabling dynamic updates as project conditions evolve. By incorporating probabilistic dependencies, their method enhances traditional scheduling techniques, providing a more robust framework for managing uncertainties in project timelines. Hybrid approaches like these demonstrate the potential to create predictive and adaptive scheduling frameworks that address both qualitative insights and quantitative metrics.

In his 2022 doctoral dissertation, Mohamed [19] developed a comprehensive risk-based inspection (RBI) framework for highway construction projects, integrating FST and BBNs into a unified fuzzy Bayesian belief network (FBBN) model. This approach enables the prioritization of quality assurance inspection activities based on risk criticality, considering factors such as cost, safety, and service interruption while explicitly addressing the uncertainty and subjectivity inherent in qualitative data. Through a case study conducted with the Kansas Department of Transportation, the study demonstrated the model’s capacity to infer causal relationships among inspection variables and optimize inspection strategies under resource constraints, marking a significant advancement in data-driven construction quality management.

Fan & Yu [147] introduced a BBN-based framework for software project risk management that incorporates predictive feedback loops to identify potential risks and their sources. Their model supports dynamic resource adjustment and enhances transparency and consistency in decision-making processes. Through both analytical and simulated case studies, the authors demonstrate how the BBN approach improves the visibility, adaptability, and repeatability of risk assessment across project stages.

In their 2008 study, Trucco et al. [52] developed a BBN model to assess the impact of organizational factors on safety risks within maritime transportation systems. The paper presented a structured and probabilistic approach that integrates expert knowledge and empirical data to evaluate causal relationships among risk-contributing factors. By applying the model to a case study of the Italian port state control system, the authors demonstrated the model’s capacity to improve risk understanding, support informed decision-making, and identify leverage points for safety improvement interventions. This application highlights the broader utility of BBNs in complex, safety-critical domains.

As evidenced by the studies, BNs offer significant methodological advantages, particularly in their ability to simultaneously process both quantitative and qualitative data. This dual capacity enhances reasoning and supports more robust decision-making in the presence of uncertainty. Furthermore, BNs facilitate the integration of diverse information sources, thereby improving the accuracy and reliability of predictions when compared to more traditional, deterministic models.

Although BNs offer powerful capabilities for modeling uncertainty and causality, Trucco & Leva [149] highlight several methodological limitations that can challenge their application in operational risk analysis. These include the rapid complexity growth of models with multiple parent nodes, difficulties in eliciting reliable expert judgments, and the need for discretization of continuous variables, which can inflate computational demands. Moreover, the lack of a consistent, theory-driven framework for incorporating organizational factors often leads to ambiguous models that are difficult to validate empirically. These challenges underscore the importance of adopting rigorous modeling practices when applying BNs in complex socio-technical systems.

Kyrimi et al. [150] conducted a comprehensive scoping review of 123 studies to assess the use of BNs in healthcare, highlighting a striking gap in systematic evaluations and real-world adoption. Despite their theoretical strengths, BNs are underutilized, partly due to the absence of standardized development frameworks and inconsistent presentation in literature. These issues limit understanding, methodological coherence, and practical implementation. While the study focuses on healthcare, its insights are broadly applicable across fields, emphasizing the critical need for structured methodologies and better translation of technical models into decision-support tools.

Integrated risk-scheduling methodologies have a profound impact on project outcomes, primarily through their ability to balance competing constraints of cost, time, quality, safety, etc. By merging risk assessment tools with advanced scheduling algorithms, such as Genetic Algorithms (GE) or reinforcement learning, construction managers can simulate multiple scenarios, evaluate trade-offs, and optimize decision-making under uncertainty. As illustrated in Figure 3, this integration of qualitative and quantitative risk assessment with mathematical and metaheuristic scheduling techniques enhances project resilience and minimizes disruptions caused by unforeseen risks.

Addressing resource-constrained scheduling problems under uncertainty has been a longstanding challenge in project management. Furthermore, hybrid methodologies improve communication and stakeholder engagement by providing a transparent and comprehensive view of project risks and schedules. For instance, Fuzzy Logic models combined with resource-constrained scheduling have been shown to effectively address uncertainties in resource availability while maintaining project timelines. With foundational contributions dating back to the work of Lorterapong & Moselhi [151], who introduced a fuzzy heuristic method for resource-constrained project scheduling, addressing uncertainties in activity durations and resource availability. Long & Ohsato [152] proposed a fuzzy critical chain method for project scheduling under resource constraints and uncertainty, enhancing project timeliness’ robustness. Song et al. [153] extended traditional Schedule Risk Analysis (SRA) by incorporating resource constraints into the risk assessment process, introducing novel sensitivity metrics tailored for resource-constrained environments. Their proposed Resource-Constrained SRA (RC-SRA) framework supports more accurate identification of critical activities and facilitates dynamic corrective actions during project execution. This integration exemplifies how risk assessment tools can be synchronized with advanced scheduling techniques to enhance project control and mitigate uncertainties in both time and resource dimensions.

In conclusion, the fusion of risk assessment and scheduling techniques represents a paradigm shift in construction project management. These integrated approaches not only provide a holistic understanding of risks but also empower managers to develop adaptive strategies, ensuring projects are delivered on time, within budget, and to the desired quality standards.

5. Conclusions

Qualitative and quantitative techniques have been reviewed to identify those that are most relevant to road and highway construction risk management. The identified techniques were chosen based on their widespread adoption within Highways or Road Construction, frequent citation in academic literature, and ability to address specific challenges in construction projects, such as schedule delays, cost overruns, and project risks. This review highlights that hybrid methods such as Bayesian Networks and Fuzzy-AHP offer practical advantages over traditional scheduling tools, including improved adaptability, better integration of expert judgment, and enhanced schedule risk modeling. The analysis of risk assessment and scheduling techniques led to the following observations:

• Fuzzy Logic and AHP are commonly utilized techniques in risk assessment and analysis. While Fussy logic and AHP provide valuable tools for risk assessment in road construction projects, they require careful adaptation and refinement to address their respective shortcomings and improve decision-making reliability.
• This review demonstrates that no single risk assessment or scheduling technique is universally sufficient to address the multifaceted challenges of highway construction project management. Each method contributes uniquely to capturing uncertainties, managing resources, and supporting informed decision-making. The study underscores the need for context-driven method selection, where the choice of technique is tailored to the specific characteristics of the project, such as its complexity, data availability, and risk profile.
• Integrating risk assessment techniques with scheduling frameworks provides a more realistic and proactive basis for construction planning. Emerging methods, including probabilistic simulations and machine learning-assisted risk models, show promise in enhancing the responsiveness and accuracy of project scheduling. Ultimately, advancing highway construction management requires the continued development of interoperable, hybrid approaches that combine expert judgment with data-intensive tools to manage risk holistically.
• Existing techniques such as CPM, MCS, and PERT exhibit inherent limitations in effectively simulating and managing schedule risks, particularly in handling uncertainties, activity correlations, and sensitivity to risk factors. Given these challenges, BNs offer significant potential to enhance traditional scheduling methods by integrating probabilistic reasoning and dependency modeling. Further investigation into BN-based hybrid models could help overcome current limitations and substantially improve the robustness and reliability of project scheduling.

6. Discussion and Future Research Directions

The comparative analysis in this study highlights the diverse range of techniques available for risk assessment and scheduling in construction projects, each with distinct strengths, limitations, and domains of applicability. While deterministic scheduling approaches like CPM remain widely used, they often fall short in capturing uncertainty, underscoring the need for probabilistic models such as PERT and MCS. Likewise, while qualitative techniques are crucial in early-stage planning, quantitative models such as AHP and FST support more structured and data-driven evaluations. Future research should prioritize the development of integrated frameworks that bridge the gap between risk assessment and construction scheduling. Key areas for advancement include the following:

• Exploration of Multi-Objective Optimization for Time–Cost–Risk Trade-Offs.

A promising direction for future research involves the development of advanced multi-objective optimization models that address trade-offs among time, cost, and risk in construction scheduling. While traditional models often prioritize time and cost, real-world projects require balancing multiple conflicting objectives. Future studies should explore metaheuristic algorithms such as genetic algorithms and evolutionary strategies to support robust decision-making under uncertainty.

• Application of Bayesian Networks in Construction Scheduling.

Given CPM’s limited flexibility and PERT’s optimism bias, hybrid models that incorporate dynamic risk reasoning, such as BNs, are essential for real-time responsiveness in highway construction environments. Hybrid models such as Fuzzy-AHP and PERT-MCS emerge as promising strategies, bridging gaps between intuition-based assessments and rigorous analytics. However, current industry practice often reveals a disconnect between risk assessment and scheduling tools. This separation undermines the capacity of project managers to adapt plans in real time based on evolving risks. The evidence presented in this paper supports an integrated, situationally aware approach to tool selection, promoting alignment between precise scheduling and risk-informed decision-making. Of particular interest is the integration of BNs with established scheduling frameworks such as the CPM and PERT. This hybridization enables more accurate and dynamic evaluation of schedule risks, supporting the development of intelligent, risk-informed scheduling systems. Future research should further explore the application of BNs in highway construction to model complex dependencies and uncertainties inherent in project environments. BNs provide a robust methodological foundation for risk analysis by integrating both quantitative and qualitative data, enabling a more comprehensive and nuanced understanding of uncertainty. This dual capability supports informed decision-making under conditions of incomplete or variable information. Further research into the use of BNs for constructing risk knowledge bases could substantially enhance strategic planning and operational resilience in infrastructure projects.

• Advancing Decision Support through AI and Interoperable Risk-Schedule Systems.

Future research should focus on the integration of artificial intelligence (AI) and machine learning (ML) to enhance predictive risk modeling, anomaly detection, and adaptive scheduling in construction projects. These technologies offer significant potential to improve forecasting accuracy and enable proactive decision-making under uncertainty.

A key area of development involves the creation of interoperable systems that dynamically link risk registers with scheduling algorithms. Such integration would support real-time responsiveness and facilitate continuous risk-schedule alignment throughout the project lifecycle. Additionally, future efforts should prioritize the development of advanced decision support tools capable of automating expert input validation and synthesizing both qualitative judgments and quantitative data. By leveraging AI and ML capabilities within these systems, construction managers can achieve more reliable, data-informed decisions, particularly in complex and evolving project environments.