Genetic Algorithms Applied In Fleet Maintenance Analysis: Case Studies with Seven Underground Scoop Trams

"While genetic algorithms have been widely adopted in ore grade estimation and optimization, mine design, and production scheduling in the mining engineering field, reliability and maintainability analysis of mining equipment using genetic algorithms has been a relatively new research area in recent years. This paper presents an application of genetic algorithms using case studies with seven underground scoop trams whose maintenance records have been collected over a period of 12 months from an underground mine in the Sudbury mining area. Based on the assumption that various factors that contribute to machine failures resemble the biological evolution process, the authors have developed an in-house PC software application called GenRel which adopts genetic algorithms in its programming. In this study, the authors feed the historical maintenance data into GenRel in order to predict future maintenance characteristics of mining equipment in a time period of three months. Then the predicted maintenance characteristics are compared to the actual maintenance records in the same time period. A statistical study is then used to examine the similarity between the predicted data and the recorded data.INTRODUCTION Today maintenance costs of sophisticated engineering systems are so high that maintainability draws great attention from scientific researchers to operations managers. For instance, a study by [1] shows that American manufacturers spend more than 300 billion U.S. dollars on plant maintenance and operations. Therefore it is understandable that the main objectives of applying maintainability principles to engineering systems are to reduce projected maintenance costs, to use maintainability data to estimate system or equipment availability/unavailability and to determine labor-hours and other related resources needed to perform the projected maintenance. A system with better maintainability would inherently provide the benefit of lower maintenance costs, less time to recover with lower breakdown frequency (design for simplicity), less complexity of maintenance tasks and relatively reduced man-hours [2]. Most maintainability functions use the Time To Repair (TTR) as the independent variable. It is common to use probabilistic or statistical concepts to define a maintainability function, for example, the probability density function. Let t denote the time. Assuming a repair starts at t=0 and completes at time T, the maintainability can be mathematically defined [3] as in Equation (1)."