Implementation Of Service Automatic Composition Framework In ERP System

With the development of enterprise intelligence,information management system(ERP)has been greatly welcomed by more and more traditional enterprises.How to develop each function of the ERP system faster and more efficiently has become a key issue.This article first uses the service-oriented architecture(SOA)to abstract the various functions implemented in the ERP system into various Web services,which can repeatedly call related services when implementing similar functions,which greatly improves the reuse of the system Sex.Complex functions require multiple services to work together,but in the face of more and more services,how to match and automatically compose calls between services has become a big problem.Based on the SOA architecture,this paper adds the automatic service matching and calling function based on semantics.Transform Web services into semantic Web services so that services have semantic knowledge that computers can understand.When the user's request arrives at the system,it analyzes the demand information and quickly and semantically matches the Web service set.At the same time,when the service is automatically matched and combined,the quality of service(QoS)should be considered.When facing multiple services that implement similar functions,they can assign certain weights to these services based on QoS information and make load balancing calls for the services.This paper presents two algorithms,and combines the two into a service automatic composition calling framework.The first algorithm is an automatic service composition algorithm based on semantics.Compare the input and output semantic information of two services.When the output semantics of a service matches the input of another service,define the two as semantically matched two services.Considering all services as nodes,the two services with semantic matching are connected by a directed edge,and the weight of the edge is the quality information of the service,thus forming a weighted directed graph.Then add two nodes of user request input and output in the figure,and transform the problem of automatic service invocation into the problem of finding the path with the best service quality.In this paper,three algorithms are used to complete the path search,and the simulation experiments verify the efficiency of the algorithm and the advantages and disadvantages of the algorithm under different numbers of services.The other algorithm is the QoS dynamic prediction algorithm.Since QoS data may be missing and usually changes dynamically,in order to realize the dynamic prediction and update of QoS data,this paper draws on the idea of the recommendation system and uses the collaborative filtering algorithm to similar Users and services are aggregated to give a prediction of QoS data when a user invokes a service based on a similar set of users and services.Simulation experiments also prove that the accuracy of the prediction results is above 85%,which shows the reliability of the algorithm.Finally,the two algorithms are integrated into a service automatic call composition framework,and the framework is applied to the ERP system.This framework is used to complete the development of a company canteen dining management sub-module in the ERP system to verify the effectiveness of the framework.