Functional resilience evaluation of road tunnels through stochastic event simulation and data analysis

Abstract

Resilience of tunnels can have significant impact on the efficiency of the entire transportation network. The ability to assess resilience of tunnels accurately is important for tunnel owners and stakeholders when they evaluate the cost-benefit of the investment made and the monetary value of future maintenance and upgrade activities. In this thesis, a simple and direct measurement metric for tunnel functionality was proposed with the focus on the usage of road tunnels. An ideal data collection framework for tunnels was proposed to support the calculation of tunnel functionality, as well as additional data-driven analysis that can be conducted to seek correlation between tunnel design and operation parameters with its resilience. As an example, existing tunnel operational data collection practice in a large tunnel in Colorado was summarized and compared with the proposed framework. Data analysis was performed for Eisenhower Johnson Memorial Tunnel (EJMT), Colorado. Since the data for the tunnels was found to be insufficient and incomplete to perform a completely data-driven analysis, a stochastic simulation model to predict tunnel resilience over time was developed by simulation of individual disruptive events. The model was a combination of modules representing disruptive events namely, accident, vehicle fire, hazmat platooning, maintenance and operations. This model was validated to the extent that is realistic using limited data from EJMT. Further, a parametric sensitivity analysis was performed to identify the impact of tunnel parameters, associated with disruptive events, on the tunnel functionality loss and resilience. The parametric study was also expanded to conduct a preliminary correlation assessment of tunnel key parameters and their performances using 22 major road tunnels in the United States.