A Study on Train Driver Scheduling Problem

• Main Authors: Yen-hsiu, Kuo, 郭彥秀
• Other Authors: Chi-Kang, Lee
• Format: Others
• Language: zh-TW
• Published: 2000
• Online Access: http://ndltd.ncl.edu.tw/handle/68644151623950481924

碩士 === 國立成功大學 === 交通管理學系 === 88 === Train driver scheduling problem is considered to make the work schedule of train driver in order to cover a planned timetable. In practice, railway company uses schedulers who are expert to make schedule by hand and it has many drawbacks. In this research, we propose a new approach to solve train driver scheduling problem more efficiently and systematically. The overall train driver scheduling problem includes two subproblems, CREW SCHEDULING and CREW ROSTERING. We review related literature in this field, give suitable ways to model the two problems and find possible solution methods. In crew scheduling, we decompose the problem into two parts. The major problem is formulated as a multi-objective set covering problem that can be solved using the LINDO software package. The subproblem is programming a generator to produce high-quality duties that increase input data to the major problem to improve the solution in each iteration. In crew rostering, the problem is formulated using network structure. We use the LINDO software package and a heuristic method to solve the crew rostering model. In order to test the applicability of our model, we perform a case study of Taiwan Railway Administration. The results are good.