| High speed railway line planning (HSRLP) is the critical process to balance the relationship between passenger demand and transport service in the high speed railway transportation planning. It gives a powerful guarantee for the quality of high-speed passenger product and the implementation effect for the follow-up transportation plan. At present, most studies at home and abroad focus on train line planning based on given passenger demand with assuming that passenger demand as a fixed value. However, the passenger demand is dynamic, the railway transport enterprises have to improve their competitiveness by adjusting and updating the line planning as soon as possible. This become an urgent practical problem to be addressed.This paper focuses on the optimization theory and method for the high speed railway line planning based on dynamic passenger demand (HSRLPBDPD). It sets up the theories and methods of the feedback adjustment between passenger demand and line planning. The main work in this paper could be summarized as followed:(1) The mechanism of the HSRLPBDPD is analyzed in this paper. The concept for the HSRLPBDPD is proposed in this paper by analyzing the space distribution features and the time change regulation of the HSRLPBDPD. Moreover, the research content and method is also proposed.(2) This paper develops a two stage approach for the HSRLP. Firstly, a mixed integer programming model based on multi-model is built. Then HSRLP is proved to be NP-hard problem. Therefore, the two stage approach is proposed. The first stage is the line planning for the higher-classified train with the standard multi-commodity model. The second stage is the line planning for the lower-classified train with the multi-commodity model containing some constraints. Furthermore, the benders decomposition algorithm and column generation algorithm integrating the decomposition strategy and heuristic strategy are designed to solve the models for the two stages respectively.(3) This paper proposes a robust optimization model for the HSRLPBDPD. Assuming that the passenger demand lie in a interval, the robust optimization model for the high speed railway line planning is built according to the robust optimization theory. The solution of the model cannot be modified when the passenger demand fluctuate in a interval. Then the model is converted to linear mixed integer programming model which can be solved by the two stage approach, as well as the computational complexities for the formulation is analyzed.(4) This paper proposes a adjustment optimization model for the HSRLPBDPD. When the passenger demand fluctuate beyond the permissible fluctuation interval, the method to adjust the line planning in combination of basic line planning and current line planning is proposed. Firstly, the basic line planning is generated, which contains the trains which have a good operation efficiency and high attendance and the fixed trains. Then the current line planning is generated by in increasing trains in the basic line planning as well as making small adjustments for the basic line planning. The current line planning should have a robustness. By introducing 0-1 variables and deviated cost to control the scale of the adjustment, this paper develops a preplanning model based on the current passenger demand based on the robust optimization theory. Minimizing the operator’s total operating cost, passenger’s total travel time and the deviated cost are the planning objectives of the model. And the computational complexities are analyzed. At last the two stage approach is used to solve the model. |
