考虑三方利益的车货匹配优化

为研究平台模式下考虑车主、货主及平台三方异质化需求的车货匹配问题，在既往研究考虑车货双方利益的基础上，引入了平台方需求. 首先，在分析车货匹配活动参与方需求的基础上，构建了最大化送达时效满意度、最小化货运成本和最大化平台收益的多目标优化模型；其次，在模型求解方面，改进了带精英保留策略的快速非支配排序遗传算法（non-dominated sorting genetic algorithm Ⅱ，NSGA Ⅱ），一方面在子代种群更新过程中引入精英选择系数，提升种群的多样性，另一方面结合自适应的思想，在算法迭代过程中调整交叉变异的概率；最后，利用成渝区域间的车源和货源数据进行仿真实验. 结果表明:改进的NSGAⅡ在中小型算例上的准确率均超过91%，与传统的NSGAⅡ相比，平均收敛速度提升了45%左右；在算法稳定性方面，所提出的算法受随机初始化影响较低，多次实验的相对标准偏差值小于1%. 

Optimization of Vehicle–Cargo Matching Regarding Interests of Three Parties

To study the vehicle–cargo matching problem regarding the heterogeneous needs of the vehicle owner, cargo owner and platform under the platform mode, the platform demand is introduced given that the previous studies only consider the interests of both vehicle and cargo parties. Firstly, based on the analysis of the participant needs in the vehicle?cargo matching activity, a multi-objective optimization model is built to maximize the satisfaction of the delivery timeliness, minimize the freight cost and maximize the platform revenue. Secondly, in terms of model solution, the non-dominated sorting genetic algorithm Ⅱ (NSGA Ⅱ) with elite retention strategy is improved. On the one hand, elite selection coefficient is introduced in the process of updating the offspring population to improve the diversity of the population, and on the other hand, the adaptive idea is combined to adjust the probability of cross mutation in the algorithm iterations. Finally, the simulation experiments are carried out using the data of vehicles and freights in the areas of Chengdu and Chongqing. The results show that the accuracy of the improved algorithm proposed is more than 91% on small and medium-sized examples, and the average convergence speed is increased by about 45%, compared with the conventional NSGA Ⅱ algorithm. In terms of algorithm stability, the proposed algorithm is less affected by random initialization, and the relative standard deviation of multiple experiments is less than 1%.