Study On Strategy Optimization Of Crew Scheduling Problem For Urban Rail Transit

The rapid development of urban rail transit in China, with its unique advantages, has become an important solution to traffic congestion. In this situation, the reasonable arrangement and allocation plan for crew members have a great influence on operation order and safety. Crew scheduling is also a key link for operating company. At present, manual scheduling is widely adopted in our country. Because of its disadvantages, heavy workload, much time-consumption and lacks of flexibility, there is a trend to replace manual scheduling with automatic scheduling. Automatic scheduling requires flexibility in preparation process, which will expand the solution range of crew scheduling model and improve the solution utility. Strategy optimization is applied to the crew scheduling problem of urban rail transit system in this paper, specific researches are as follows:Firstly, with the analysis of crew scheduling characteristics, this paper explains the related terms and concepts in scheduling process, establishing basic procedures for crew scheduling. Based on the strategy theory, the concept and optimization model of strategy are put forward. The basic theory of base class constraint model and strategy constraint model are also been studied respectively.Secondly, referring to the correlation theory of crew scheduling, a base class constraint optimization model is established to minimize the total connection time and penalty time, with constraint conditions of shift time, break time, dining time, work time, work efficiency, coverage condition, and work-piece amount. Meanwhile, based on greedy algorithm and probabilistic algorithm, an improved neighborhood search algorithm is developed to obtain the solution of the scheduling model, realized by C#programming language. The model and algorithm are mainly used to solve common crew scheduling problem of urban rail transit.Thirdly, the strategy constraint optimization model is established to solve the specific crew scheduling problem of urban rail transit. Analyzing the advantages of strategy constraint optimization model, this paper develops four steps to solve crew scheduling problem, policy constraint, scene classification, strategy application and result output. During the process of solving scheduling problem, main scheduling methods are concluded and strategy unit is set up, which are realized by C#language. Furthermore, on the basis of classifying common scenes of crew scheduling, this paper puts forward the optimal strategy for two different scenes, using related indicator to evaluate the results of strategy optimization model at the same time.Finally, take the Beijing subway line9as an example. With the analysis of manual scheduling result, this paper uses base class constraint optimization model to solve crew scheduling problem. It uses strategy constraint optimization model to solve the problem aimed at the two scenes. The evaluation indexes are adopted into the results comparison and analysis of base class constraint optimization model, strategy constraint optimization model and manual scheduling. The result shows that strategy optimization thought has the flexibility to changes and it is adaptable to different crew scheduling problems, which is proved to be practical and effective.