详情

原文传递 ARTIFICIAL INTELLIGENCE FOR TRANSIT RAILCAR DIAGNOSTICS.

题名：	ARTIFICIAL INTELLIGENCE FOR TRANSIT RAILCAR DIAGNOSTICS.
作者：	Mulholland-IP; Oren-RA
关键词：	TRANSIT-RAILCARS; DIAGNOSTICS-; ARTIFICIAL-INTELLIGENCE; EXPERT-SYSTEMS; KNOWLEDGE-BASED-SYSTEMS; SITE-SURVEYS; VEHICLE-MAINTENANCE; MAINTENANCE-PRACTICES; MAINTENANCE-COSTS; ECONOMIC-ANALYSIS; COST-SAVING; TIME-SAVING-METHODS
摘要：	This report documents and presents the results of a study to determine the feasibility of applying Artificial Intelligence (AI) techniques to the diagnosis of transit railcars. The AI techniques investigated were expert systems, case-based reasoning, model-based reasoning, artificial neural networks, computer vision, fuzzy logic, and a procedural knowledge-based system. Site surveys were conducted at transit railcar maintenance facilities and at railcar subsystem suppliers. The site surveys gathered information about current and future diagnostic and maintenance practices, possible barriers to implementing advanced AI technology, and maintenance cost data. An economic analysis was performed to provide an estimate of cost savings expected by reducing the diagnostic effort. An initial AI diagnostic program, using a commercial AI shell, was estimated to cost $172,000. The economic analysis indicated that if improvement in diagnosis leads to an improvement of 7.2% in the mean time to repair of the propulsion subsystem, then the diagnostic program will have paid for itself within 1 year. The AI diagnostic program recommended for initial implementation consisted of a hybrid of model-based reasoning and expert system approaches.
总页数：	TCRP REPORT. 1994. (1) pp62 (8 Fig., 15 Tab., Refs., 5 App.)
报告类型：	科技报告