考虑随机需求和硬时间窗的多目标车辆路径优化方法

A multi-objective vehicle routing optimization method based on stochastic demand and hard time window

市场的波动性导致客户货物需求量的随机性,使车辆路径优化问题更加复杂。考虑客户需求量的随机性,基于车辆运输满足需求可靠性要求,构建与货物需求期望和方差相关的约束条件,建立最小化线路里程和均衡度的多目标车辆路径模型。针对模型特点,设计基于非支配排序的精英蚁群算法。主要设计3点策略:1)采用贪心策略构建初始蚁群;2)考虑等待时间、时间窗宽度对蚂蚁概率转移的影响;3)通过非支配排序选择精英蚂蚁释放信息素。以Solomon中C101类部分客户为例进行实证分析,在考虑不同需求和时间窗情况下,对参数进行灵敏度分析,证明模型和算法的有效性和可靠性。研究结果表明:在考虑随机需求和硬时间窗下的多目标车辆路径能较好地提高运输效率。

The volatility of the market leads to the randomness of customer demand volume for cargo, which makes the problem of vehicle routing optimization more complicated. This paper established a multi-objective vehicle routing model to minimize route mileage and equilibrium. By considering the randomness of customer demand, the constraints related to the expectation and variance of cargo demand were constructed based on the condition that vehicle transportation meets requirements of reliability. According to the characteristics of the model, an elite ant colony algorithm was developed based on non-dominated sorting, and mainly three strategies were designed: 1) Construct initial ant colony by greedy strategy;2) Consider the influence of waiting time and window width on the probability transition of an ant;3) Select elite ants to release information pheromones by non-dominated sorting. Take some customers of the C101 category in Solomon as an example of empirical analysis. By considering different requirements and time windows, the sensitivity analysis of the parameters was carried out to prove the validity and reliability of the model and algorithm, and the Pareto solution is solved.Considering random demand and multi-target vehicle paths under hard time windows can better improve transportation efficiency.