Maintenance Performance Indicators using Hybrid Models for Food Processing Industries

The performance and competitiveness of manufacturing companies are depending on the reliability and availability of their production equipment. To ensure the equipment always at desired performance, maintenance managers need a good track of the machines operation, maintenance activities and their results. They should always attain all maintenance data, analyze them periodically, and subsequently transform it into useful strategies. Then act on the strategies to increase equipment productivities. In this research, we demonstrate on how to use Maintenance Decision Grid (MDG) and Fuzzy Inference System (FIS) to analyze maintenance performance in food processing industries. The MDG model is used to analyze top ten problematic machines in the production lines. Next, FIS is used to visualize output based on the given rules and input i.e. frequency of failures, downtime and cost. Based on MDG analysis, maintenance strategies are recommended i.e. operate to failure, service level upgrade, condition-based maintenance, design-out maintenance, total productive maintenance and reliability centered maintenance for one of the food processing industries. We collected maintenance data in 2008 and 2009 from the industry. Maintenance strategies given to the technical team and data are collected again in 2010 to observe any improvement. From the observation, we have verified that breakdowns reduced tremendously with the given strategies using MDG and FIS models. The objective of this paper is to use hybrid models i.e. MDG and FIS to analyze failure-based maintenance data. The MDG with FIS model give promising decision analysis where the performance indicators and strategies able to improve machine availabilities