Value-at-Risk Models for Hazardous Material Transportation.

详细信息   

• 作者：Kang ; Yingying.
• 学历：Doctor
• 年：2011
• 导师：Batta,Rajan,eadvisorKwon,Changhyun,eadvisorBatta,Rajanecommittee memberKwon,Changhyunecommittee memberRogerson,Peterecommittee memberVerter,Vedatecommittee member
• 毕业院校：State University of New York
• Department：Industrial Engineering.
• ISBN：9781124730769
• CBH：3460764
• Country：USA
• 语种：English
• FileSize：1416694
• Pages：148

文摘

This dissertation develops Value-at-Risk (VaR) risk models for hazardous material (hazmat) transportation and decision support models for single-trip and multiple-trip routing design,traffic assignment with risk equity consideration,and with road banning regulation. The first part concentrates on the introduction of the Value-at-Risk (VaR) model for hamzat transport and applies it to the case of routing a single hazmat trip. Several properties are analyzed regarding the model and a two-stage exact solution method is developed based on it. A case study is presented to provide route choices according to decision makers' risk preferences for hazmat shipments in Albany,NY. The second part applies the VaR concept to a more realistic multi-trip multi-hazmat type framework which aims at determining routes that minimize the global VaR value while satisfying equity constraints. A solution algorithm is embedded for the single hazmat trip problem into a Lagrangian relaxation framework to obtain an efficient solution method for this general case. The computational experience was tested based on a real-life hazmat routing scenario in the Albany district of New York State. The third part addresses the problem of balancing both system optimum and user equilibrium for hazmat shipments under road banning constraints. A bi-objective model is built with two solution methods developed. An exact solution is built for small-sized network by converting the bi-level model into single-level integer programming model. Later on,a Genetic Algorithm-based heuristic is developed to solve problem in large-sized network. A case study,with highway network in Albany,NY,is presented to illustrate the results.