基于偏好矩阵遗传算法求解长期车辆合乘问题

发布时间：2018-01-06 06:05

本文关键词：基于偏好矩阵遗传算法求解长期车辆合乘问题　出处：《计算机应用》2017年02期 　论文类型：期刊论文

【摘要】：针对长期车辆合乘问题(LTCPP),提出带有偏好矩阵的遗传算法(PMGA),将拥有私家车且目的地相同的用户群体分配到产生总花费最少的合乘小组。首先,建立计算基于全体用户费用成本的目标函数,构建以用户时间窗和车容量为约束的长期车辆合乘模型;然后,结合模型特点,在传统遗传算法(GA)的基础上,通过在交叉算子与变异算子中添加偏好矩阵记录并更新用户间的偏好信息来提高可行解的数量和质量。实验结果表明,在相同计算环境下,当用户数量小于200时,通过PMGA所获得的20个解中的最优解的值与最优化算法相同;而处理大规模的实例时,PMGA可以获得更高质量的解。所提算法可以明显提高长期车辆合乘问题的求解质量,在降低汽车尾气污染和减少交通拥挤等方面具有重要作用。
[Abstract]:In this paper, a genetic algorithm with preference matrix (PMGA) is proposed to solve the problem of long term vehicle ride-sharing. Assign the group of users with the same destination and own private car to the group with the least total cost. First, set up the objective function to calculate the cost based on the cost of all users. A long-term vehicle co-ride model with user time window and vehicle capacity as constraints is constructed. Then, combined with the characteristics of the model, based on the traditional genetic algorithm (GA). The number and quality of feasible solutions are improved by adding preference matrix to crossover operator and mutation operator and updating the preference information between users. The experimental results show that in the same computing environment. When the number of users is less than 200, the optimal solution of 20 solutions obtained by PMGA is the same as the optimization algorithm. When dealing with large scale cases, PMGA can get higher quality solution. The proposed algorithm can improve the quality of solving long-term vehicle ride problem obviously. It plays an important role in reducing vehicle exhaust pollution and traffic congestion.
【作者单位】：辽宁工程技术大学软件学院;
【基金】：辽宁省自然科学基金资助项目(2015020095)~~
【分类号】：U491;TP18
【正文快照】： 0引言随着社会经济持续快速发展,我国城市私家车保有量持续大幅增长。私家车的大量使用所引发的交通拥堵、环境污染等问题日益严重[1-3],因此车辆合乘问题成为近年来研究的热点。目前车辆合乘问题的研究多是单次车辆合乘问题,具有随机性和临时性。本文所提出的长期车辆合乘是

【参考文献】