Artificial neural network approach for condition-based maintenance
This work addresses maintenance cost reduction for industries using equipment, but it appears incremental as it applies an existing ANN method to a known CBM framework without novel methodological breakthroughs.
The paper tackles the problem of high maintenance costs by proposing an Artificial Neural Network (ANN) approach for Condition-Based Maintenance (CBM) to analyze equipment condition signals and estimate failure times, aiming to achieve an optimal maintenance policy.
In this research, computerized maintenance management will be investigated. The rise of maintenance cost forced the research community to look for more effective ways to schedule maintenance operations. Using computerized models to come up with optimal maintenance policy has led to better equipment utilization and lower costs. This research adopts Condition-Based Maintenance model where the maintenance decision is generated based on equipment conditions. Artificial Neural Network technique is proposed to capture and analyze equipment condition signals which lead to higher level of knowledge gathering. This knowledge is used to accurately estimate equipment failure time. Based on these estimations, an optimal maintenance management policy can be achieved.