第四方物流服务是新经济形式下的新型物流产业形态,它能够在区域范围内最大限度地整合具有互补性的资源,过滤出高服务水平的服务组合方案,为客户提供高水准、高服务质量、低经济成本的专业化服务.然而,如何高效地、动态地构建和组合出客户满意的、及时可靠的一体化服务,提高服务发现与组合的满意度是第四方物流产业运作模式具有挑战性的问题.针对该问题,基于遗传算法中的交叉变异思想,通过对社会认知算法中模仿学习和观察学习过程改进,使其可以用于求解离散型的物流Web服务优化组合问题.实验结果显示,该算法与遗传算法、最大最小蚁群算法相比,具有更高的寻优能力和收敛速度,能够为现代物流服务提供有效的支撑.
The fourth party logistics service is a new kind of logistics industry under the new economic form,it can maximize the integration of complementary service resources within the region,filter out the high level of service composition plans,provide clients with professional services which have high standard,high quality and low economic cost.However,how to efficiently and dynamically build,integrate and compose the integrated services which is timely,reliable and keep the customers satisfied,improve the satisfaction of service discovery and composition is a challenging problem under the operation mode of the fourth party logistics industry.Based on the crossover and mutation thought of genetic algorithm,this paper improved imitation learning and observational learning of the social cognitive optimization,so that it could be used to solve the discrete logistics Web service composition problem.The experimental results show that the algorithm has higher search ability and convergence speed which is compared with genetic algorithm and the maximum minimum ant colony algorithm,it really can provide effective support for the modern logistics service.