Using Dynamic Vehicle Routing Model to Dispatch Emergency Response Teams on Freeways

碩士 === 國立臺灣大學 === 土木工程學研究所 === 106 === According to historic data, more than half of freeway congestion is caused by incidents, and enhancing efficiency of incident response can reduce the delay caused by congestion. For most of freeway management systems, they depend on incident response teams to c...

Full description

Bibliographic Details
Main Authors: Chih-Yu Tai, 戴至佑
Other Authors: 許聿廷
Format: Others
Language:en_US
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/jqa5pn
Description
Summary:碩士 === 國立臺灣大學 === 土木工程學研究所 === 106 === According to historic data, more than half of freeway congestion is caused by incidents, and enhancing efficiency of incident response can reduce the delay caused by congestion. For most of freeway management systems, they depend on incident response teams to clean up and remove incidents, and optimizing decision of response team dispatch is an effective way to improve the efficiency of incident response. The goal of this study is to apply an optimization method to enhance the quality of the incident assignment strategy to reduce the response time for the response teams to arrive at each incident. We propose the framework of a dynamic dispatch system, clarify the parameters that need to be input or obtained, and develop a model based on the vehicle routing problem for incident response on the freeways. In addition to minimizing the total incident response time, the model also incorporates the considerations of reassignment and prioritizing incidents in dynamic dispatch, making the dispatch model more flexible and consistent with realism of operational needs. In this study, the roadway network of the Northern District of Taiwan Area Freeway is used for the case study to compare the dynamic dispatch model with the current dispatch approach. It is found that the dynamic dispatch model can provide an incident assignment strategy that minimizes the total response time after a new incident occurs so as to enhance the efficiency of freeway incident response, particularly for the peak period where multiple incidents are likely to occur concurrently. Additionally, it also allows the flexible adjustment of the reassignment frequency or incident priority by changing the weight of the related parameters. Our research can provide the insights agency for the construction of a dynamic dispatch system for freeway management units and help dispatchers make more efficient dispatch decisions.