Intelligent Fault Diagnosis in Condition—Based Maintenance,Using CART—CBR Model in Oil Analysis

Using developed methods, such as maintenance actions, will be helpful in increasing productivity. Productivity will be improved by using at least corrective and at most Preventive Maintenance (PM) such as Condition Based Maintenance (CBM), which is run by using complex technical systems. CBM has a great advantage over other PM techniques. However, it is not so commonly used in industries. The main focus of the present paper is to use intelligent diagnosis techniques to adopt a different data mining approach to oil condition attributes that fulfills the objective. This was achieved by using classification methods and Case—Based Reasoning (CBR) to create a decision support system model through a case study to illustrate how a CBM system for truck engine fault diagnosis and prognosis can be designed using the artificial intelligence method case-based reasoning and oil analysis The results of the combined CART—CBR method were compared with the results of three methods of CBR, CART and neural network. The results of the comparison show a better performance of the CART—CBR combination method

Keywords:

Classification and Regression Tree (CART) , Case—Based Reasoning (CBR) , Condition Based Maintenance (CBM) , Oil Analysis , Data Mining , Intelligent Diagnosis and Prognosis