Author:
Ye Zimeng,Zhang Dabin,Liu Li
Abstract
Abstract
Aiming at the problem that fixed speed limit and static regular scheduling of single-type bus cannot meet the actual passenger flow demand with uneven distribution of time and space. By designing a dynamic speed limit rule of full load rate, based on the integrated scheduling of multiple types of bus models, taking the total passenger cost and bus scheduling cost as the overall optimization objective, and considering various constraints comprehensively, OBLESGA (Opposition-based learning elite strategy genetic algorithm) was designed to solve the problem, and the data of Beijing 563 route downlink was taken as an example for simulation experiment. The results show that OBLESGA can improve the convergence rate and solution accuracy of GA. At the same time, the optimized scheduling can reduce the cost of passenger congestion by 86.61 percent and the cost of passenger waiting times by 40.65 percent compared to traditional scheduling, while only increasing operating costs by 26.57 percent.
Subject
Computer Science Applications,History,Education
Reference14 articles.
1. The Identification Method and Analysis of Ground Bus Travel Demand Characteristics Based on Massive Data CICTP 2017 : Transportation Reform and Change—Equity;Di,2017
2. Combination scheduling optimization method of bus cross-line based on passenger flow characteristics identification [J];Jiancheng;Journal of South China University of Technology (Natural Science Edition),2012
3. Deep reinforcement learning based dynamic optimization of bus timetable;Ai;Applied Soft Computing,2022
4. Bus timetabling considering passenger satisfaction: An empirical study in Beijing;Shang;Computers & Industrial Engineering,2019
5. Dynamic Multiline Vehicle Dispatching Strategy in Transit Operations;Delgado;IEEE Transactions on Intelligent Transportation Systems,2022