非满载车辆调度问题是车辆调度问题中的一个基本问题 ,由于它是一个典型的NP难题 ,传统方法的求解结果往往不能令人满意。本文采用最大保留交叉、交叉率和变异率自适应变换等技术 ,设计了基于自然数编码的遗传算法 ,用于求解非满载车辆调度问题 ,在实验分析中获得了较好的结果。 Vehicle scheduling problem is one of the basic problems in vehicle scheduling problem. Because it is a typical NP problem, the solution of traditional methods is often unsatisfactory. In this paper, genetic algorithm based on natural number coding is designed to solve the scheduling problem of non-full-loaded vehicles by using techniques such as maximum reserved intersection, crossover rate and adaptive mutation rate. The better results are obtained in the experimental analysis.