DEVELOPMENT OF AN EXPERT SYSTEM AND EMPIRICAL PROBABILITY FOR DIAGNOSING AND PREDICTING CAUSES OF OVERHEATING IN BULLDOZER (MODEL D60S-5 KOMATSU PRODUCT

Abstract

Overheating is the condition of an automotive engine in which the operating temperature of the engine is higher than the normal range. Overheating is a common problem of automotive engine cooling system. Causes of overheating include defective radiator, water pump, local climate and others. Overheating can cause economic loss and down time of equipment, hence there is need to attend to it promptly. This research work focused on diagnosis of causes of overheating in bulldozer engine of KOMATSU product MODEL D60S-5 and provided tools for diagnosing the causes and proffering solutions to the identified cause(s) of overheating. Automotive engine cooling systems and components were studied extensively. Manufacturer’s maintenance manuals and maintenance job card of case study were consulted to obtain information about the equipment and past overheating occurrences. Probabilities of the past overheating occurrences were determined and tree diagram used in decision making and for prediction. There were four major classes of the causes of overheating. A software program (expert system) was developed using C# (pronounced cee sharp) programming language. Diagnosis was carried out accordingly following the causes of overheating in the tree diagram from the highest possible cause to the least possible cause. The causes of the overheating were identified. The causes of overheating in case study 1 and 2 were defective pump (ET 3) and insufficient oil flow rate. The probability model and the software (expert system) performed better and faster than the bulldozer manufacturer’s maintenance manual in the diagnosis of the overheating problem