| Frequent engineering changes in construction projects affect the realization of goals,which also leads to the continuous occurrence of engineering disputes.Reducing engineering change disputes and reducing resolution costs of engineering change dispute have become the focus of engineering change research.With the successful application of artificial intelligence technology in other fields,it has achieved good research results in the prediction of occurrence and litigation results about engineering disputes.According to a case base and rule base for engineering change disputes which is built,the method of case-based reasoning(CBR)and rule-based reasoning(RBR)techniques is used to predict the outcome of engineering change disputes,which is of great significance for the scientific decision-making of resolution methods about engineering change dispute.Firstly,the conditional and decisional attributes that affect and represent the judgment result and the judgment rules are extracted based on the analysis of the verdict about the engineering change dispute.The traditional case representation method is selected to construct the case library and the protégé software that is editing tool of ontology is selected to build ontology-based rule base library using SWRL rule language to write rules.The purpose of building base library is that the information of verdict is described as data that is processed by computer.Then,according to the reasoning process of CBR and RBR and their respective advantages and disadvantages,a model of predicting the dispute result using RBR and CBR is proposed and established.The method of similar case retrieval based on rough-set theory and the nearestneighbor algorithm is designed and the Drools as inference engine is selected to achieve the reasoning.Finally,based on the case library,the rule library,the prediction model and method,an empirical study is carried with an example of the engineering change dispute of price.The results show that the accuracy of the lawsuit result prediction of the engineering change price dispute is as high as 90%.It shows that the method used in this paper can effectively realize the prediction of the outcome of engineering change disputes. |
