Modeling the Effectiveness of Infrastructure and Travel Demand Management Measures to Improve Traffic Congestion during Typhoons

Abstract

Many commuters find themselves stranded during natural disasters like typhoons. In the Tokai region in Japan, many road sections become heavily congested during typhoons, with some commuters reporting homebound trips taking more than four times longer than usual because of road flooding at several locations. Although large typhoons are considered extreme events (in terms of magnitude), they occur frequently (i.e., several times per year), substantiating the need for better preparedness. Nonetheless, it is impossible to predict exactly which roads are going to be flooded during a typhoon. As a result, in this study, a stochastic modeling approach was used that assigns a failure probability to each road segment based on climate model outputs for the region. Using this stochastic model, the travel time reliability between any given origin–destination pair can be determined. By applying this model to the road network of the Tokai region, two major measures were identified that could be implemented to reduce severe congestion during a typhoon. First, targeted infrastructure management measures can be implemented to strengthen heavily used roads, thus reducing the failure probability of major roads. Second, travel demand management measures can be implemented, such as asking commuters to leave their workplace or school one or two hours after their normal departure time. Overall, it was found that strengthening heavily used roads has a bigger impact in relieving congestion than delaying departure time, but that combining both strategies achieves the best results.

On the Potential Impacts of Smart Traffic Control for Delay, Fuel Energy Consumption, and Emissions: An NSGA-II-Based Optimization Case Study from Dhahran, Saudi Arabia

“…Yang and Luo [26] used a GA based optimization, with several inheriting principles of the simulated annealing (SA) process to reduce the overall delay at an isolated intersection for both transit vehicles and general vehicles [71]. Yuri et al and Ceylan et al also explored the applicability of DE for intersection optimization [72,73].…”

“…In another study, Song et al reported that the presence of bike-sharing has great prospects for boosting bicycle accessibility, thus, ensuring green and sustainable urban transportation [23]. Congestion pricing, travel demand management (TDM), proactive traffic prediction, and network optimizations have also proved beneficial in mitigating traffic congestion effectively [24][25][26][27].…”

Attitudes on Autonomous Vehicle Adoption using Interpretable Gradient Boosting Machine

“…Data science methods are increasingly being applied to problems of urban infrastructure analysis and design ( 7 , 9 , 14 – 19 ). In particular, a wide variety of statistical learning approaches have been applied to predictive learning tasks within the realm of urban research ( 7 , 20 ).…”

Modeling the Impact of Spatial Correlations of Road Failures on Travel Times during Adverse Weather Conditions

“…In most works, the simulation is performed by closing each link one by one and measuring the changes in accessibility. On the other hand, the serviceability-based approach analyzes the capability of a transport network to meet certain functional requirements when the network is degraded, such as the works in Sumalee and Kurauchi ( 14 ), Haghighi et al ( 15 ), Asadabadi and Miller-Hooks ( 16 ), and Wisetjindawat et al ( 17, 18 ).…”

“…Among the serviceability approaches, most works apply a constant failure probability ( 14, 15 ) or use a probability distribution ( 16–18 ). In the works of Wisetjindawat et al ( 17, 18 ), their approach used a stochastic method by assuming a probability function to explain the occurrences of road failure because of a natural disaster. The link-failure probability was assumed to follow a negative exponential distribution with a given failure rate derived from a climate data.…”