A Novel Approach for Analyzing the Behavior of Industrial Systems Using Uncertain Data

In some practical cases, it is not an easy to analyze the behavior of any complex repairable industrial system up to desired degree of accuracy due to vague, imprecise and uncertain data collected from the various resources (historical/present records). If somehow it can be done, then they have a high range of uncertainty. So, in order to reduce this uncertainty, to make a more sound decision for expert/decision makers by utilizing available information, this paper presents a hybridized technique namely artificial bee colony based Lambda-Tau (ABCBLT) for analyzing the behavior of complex repairable system stochastic. To strengthen the analysis, various reliability indices named as systems failure rate, repair time, mean time between failures, reliability, availability and maintainability for a time varying failure rates, instead of constant failure rates, are obtained by using Lambda-Tau methodology and artificial bee colony (ABC) optimization has been used to construct their membership function by using ordinary arithmetic rather than fuzzy arithmetic operations. A case study of the bleaching unit of a paper mill situated in a northern part of India, producing approximately 200 tons of paper per day, has been considered to demonstrate the proposed approach. The behavior analysis results computed by ABCBLT technique have a reduced region of prediction in comparison of existing technique region, i.e., uncertainties involved in the analysis are reduced.