Comparative Study Of The Service Composition Methods For Collaborative Logistics

With the development of the logistics industry model, the world is faced with the strong demand of integration and building a reasonable and efficient composite service to become the urgent request. The performance of the composite services seriously depend on the web service composition algorithms which are selected in the composite process, and because of the limitation of its efficiency and ability a single coordinated logistics service composition algorithm has been unable to strain the uncertain market environment. Such as: service orders trading volume surge in short periods of time, logistics companies cannot meet the needs of users, "Warehouse explosion" phenomenon appears; Or a sharp drop in service orders, logistics enterprises cannot provide fast and good service for end users within its ability. Based on this, this paper constructs an adaptive algorithm mechanism which supports the collaborative logistics service combination based on the analysis of composition algorithms which have excellent performance in the service composition field.(1) This paper describes the related knowledge and operation principle of the existing genetic algorithms, artificial bee colony algorithm and the max-min ant colony algorithm, the process of artificial bee colony algorithm is improved. Finally, the operation steps are determined respectively which suitable for three kinds of algorithms in logistics service combination and comparisons are made.(2) Based on a comparative study of the four aspects of the three above service composition algorithms, this paper combines with the based on case-based reasoning method and put forward the adaptive method of logistics service portfolio. This method can predict the state of the service demand sequence under the conditions of the changing market environment and adjust a service portfolio strategy dynamically, which improves the strain capacity of the system effectively. Finally, this paper demonstrates the effectiveness of the proposed adaptive algorithm through experiment.(3) To make a optimization of adaptive algorithm, this paper in consideration of the longitudinal coordinate and focus on the advantage of horizontal collaboration at the same time which can increase the number of candidate services and improve the ability of a composite service. Experiments show that optimization algorithm can effectively increase the success rate and availability of services.This paper analyzes the needs of most users to explore how to predict the market environment and make a real-time adjust in the uncertainties interference, which has some significance for the development of modern logistics.