A Dynamic Fuzzy Expert System Based on Maintenance Indicators for Service Type Selection of Machinery

Abstract

Due to the multiplicity of standards and complex rules; maintenance, repair and servicing of machinery could be done only by the fully qualified and proficient experts. Since the knowledge of such experts is not available all times, using expert systems can help to improve the maintenance process. To address this need and the uncertainty of the maintenance process indicators, this research proposed a Fuzzy Expert Systems (FES) for decision making on the type of service. Since all indicators identified in the literature aren’t important adequately, more influential indicators in the service type selection are chosen using inferential statistical analysis firstly. Then, the fuzzy rules of the knowledge based were designed by these selected indicators. Finally, Inference engine has been designed based on Mamdani model to detect the service type of equipment. This research selected Shemsh Sazan Zanjan Company as a case study to implement the proposed expert system. According to our experiments, the proposed system increases the reliability by suggesting effective ideas that lead to decrease production line breakdowns. The main contribution of this paper is providing a new approach for designing maintenance dynamic FES based on Maintenance Indicators for service type selection that can decrease production line breakdowns.

Keywords

• Expert systems
• Artificial intelligent
• Fuzzy logic
• Preventive repair and maintenance
• conveyors belt
• Maintenance Indicators