Risk Assessment Tool (Associated with Cost)
Abstract
Cost estimation in transportation infrastructure projects continues to face challenges from uncertainty, variability, and systemic data limitations. Inaccurate early estimates often lead to significant cost overruns, straining public resources and complicating project delivery. Previous studies have focused on creating parametric cost models, allocating fixed contingencies, and using risk registers to improve estimate accuracy. However, these methods have limitations. For example, parametric models usually lack accuracy once specific design details are set. Fixed contingencies do not fully account for variability across different project types, work classifications, or regions, and few studies have explored how cost estimates change over time. Additionally, risk registers rely heavily on subjective input, which can require extra time and effort and may be less reliable than objective data from historical records. This study introduces a cost adjustment framework to address these issues. Based on a comprehensive analysis of over a decade of project data, the framework integrated statistical modeling to enhance the accuracy of preliminary cost forecasts. It utilized key statistical indicators, such as the mean and standard deviation, to measure variability across work types, program categories, and geographic regions, and to track how costs evolve through different project phases. An Excel-based cost adjustment tool was developed based on this framework, improving cost estimation by providing a repeatable, data-driven approach that accounts for uncertainty. Validation using an independent dataset confirmed the tool’s ability to improve the accuracy of separate estimates. In doing so, it bridges a methodological gap between deterministic estimates and the complex realities of infrastructure delivery. This research contributes to a growing body of knowledge advocating the use of probabilistic methods in public sector project planning and provides a scalable framework adaptable to other agencies and project types.
Start Time
15-4-2026 1:30 PM
End Time
15-4-2026 2:30 PM
Room Number
311
Presentation Type
Oral Presentation
Presentation Subtype
Grad/Comp Orals
Presentation Category
Science, Technology, and Engineering
Student Type
Graduate
Faculty Mentor
Joseph Shrestha
Risk Assessment Tool (Associated with Cost)
311
Cost estimation in transportation infrastructure projects continues to face challenges from uncertainty, variability, and systemic data limitations. Inaccurate early estimates often lead to significant cost overruns, straining public resources and complicating project delivery. Previous studies have focused on creating parametric cost models, allocating fixed contingencies, and using risk registers to improve estimate accuracy. However, these methods have limitations. For example, parametric models usually lack accuracy once specific design details are set. Fixed contingencies do not fully account for variability across different project types, work classifications, or regions, and few studies have explored how cost estimates change over time. Additionally, risk registers rely heavily on subjective input, which can require extra time and effort and may be less reliable than objective data from historical records. This study introduces a cost adjustment framework to address these issues. Based on a comprehensive analysis of over a decade of project data, the framework integrated statistical modeling to enhance the accuracy of preliminary cost forecasts. It utilized key statistical indicators, such as the mean and standard deviation, to measure variability across work types, program categories, and geographic regions, and to track how costs evolve through different project phases. An Excel-based cost adjustment tool was developed based on this framework, improving cost estimation by providing a repeatable, data-driven approach that accounts for uncertainty. Validation using an independent dataset confirmed the tool’s ability to improve the accuracy of separate estimates. In doing so, it bridges a methodological gap between deterministic estimates and the complex realities of infrastructure delivery. This research contributes to a growing body of knowledge advocating the use of probabilistic methods in public sector project planning and provides a scalable framework adaptable to other agencies and project types.
