金梦宇,何胜学,张思潮. 基于改进遗传算法并考虑尾气排放的公交组合调度[J]. 交通运输研究,2021,7(02):55-65.

摘要：为在提升公交运行效率和服务水平的同时减小其对环境的负面影响，针对公交单线路单向客
流差异较大以及双向客流不均衡的现象，建立以乘客时间总成本、公交车运行总成本与尾气排放
成本之和最小为目标的跳站与区间车组合调度模型。考虑公交停靠方案与发车频率在优化过程中
的重要性差异，提出一种概率随迭代次数变化的动态概率遗传算法，对最佳停靠方案与发车频率
进行求解。通过算例分析得出：该公交组合调度方案的总成本比单一全程车调度方案节省4.49%，
虽然乘客时间总成本上升了5.04%，但公交运行总成本下降了7.12%，尾气排放成本下降了8.22%；
提出的动态概率遗传算法的求解时间小于2min，表明此算法适用于求解有明显主次影响关系的多
要素优化问题。

关键词：公交调度；跳站；区间车；尾气排放；遗传算法

Bus Combinational Scheduling Based on Improved Genetic Algorithm and Considering Exhaust Emission

Abstract： In order to improve the operation efficiency and service level of public transport, and reduce its negative impact on environment at the same time, this paper proposed a combinational scheduling model of bus stop-skipping strategy and short-turning strategy considering the phenomenon of great difference of one-way passenger flow and unbalanced two-way passenger flow on a single bus line. This model integrated to minimize the sum of the time cost of passengers, the total operating cost of buses and the exhaust emission cost. According to the difference of importance between bus stop scheme and departure frequency in the optimization process, this paper proposed a dynamic probabilistic genetic algorithm with probability varying with iteration times to solve the optimal bus stop scheme and departure frequency. The calculation analysis of one example shows that the combinational scheduling scheme can save 4.49% of the total cost compared with the scheduling scheme having single bus running whole process. Although the total time cost of passengers increases by 5.04%, the total operating cost of buses decreases by 7.12% and the exhaust emission cost reduces by 8.22%. The solution time of the dynamic probabilistic genetic algorithm is less than 2 minutes, which demonstrates that the algorithm is suitable for solving the multi-element optimization problems with obvious primary and secondary effects quickly.

Key words：bus scheduling; stop-skipping; short-turning; exhaust emission; genetic algorithm

原文PDF如下：

基于改进遗传算法并考虑尾气排放的公交组合调度_金梦宇.pdf