The Research And Implementation Of Information Management And Decision System Based On Drools

The business logic will be directly prepared in application code in traditional road information management and decision system.When the business logic changes, system updates often take a high price. With the changes of business logic accelerating ceaselessly, this highly coupled method becomes more and more difficult to meet the needs of users. The emergence of rule engine changes this situation. Rule engine is a kind of inference engine, and it originated in the rule-based expert system. Rule engine could separate business logic from application code. When the business logic changes, the system can easily be updated.This paper proposes a solution of road information management and decision system based on Drools rule engine, solves the problem of high coupling of business logic and application in traditional information management system by implementing business logic with rule engine, realizes intelligent decision of highway maintenance. Drools is an open source Java rule engine, using improved Rete algorithm.This paper firstly designs the functional architecture, the physical architecture and the technical architecture of the system, then designs the logical structure, the processes and the data models of the background as well as the main pages of the foreground in detail, finally gives the concrete realization of the system, incuding code and algorithms, and the system performance is tested. On this basis, the system has been improved by separating compilation process.This paper has designed the man-machine interface of rules management, and letting business staff to manage business logic brings great convenience to users. As the user’s input data may not be comprehensive, this paper designs four types of rule file and gives different priorities to improve the compatibility, so that the system can handle many types of data. There may be logical conflicts between the rules entered by the user and the rules in the rule base. If not handled, they will cause the system unstable. Based on the principle the set theory, this paper realizes the conflict detection of rules with string processing algorithm.Since the rules file needs to go through the processes of compilation and pattern maching, compared to traditional application, the rule engine will certainly spend more time. After testing and comparison, this paper has found that the performance of the system with rule engine is indeed decreased to some extent. Through analyzing the operation process of rule engine, this paper has realized that the rule engine performs many times after compiling once by separating compilation process. Through testing, the performance of improved implementation of rule engine has been greatly improved, and can meet the needs of users.