Optimized Maintenance Decision-Making – A Simulation-supported Prescriptive Analytics Approach based on Probabilistic Cost-Benefit Analysis

##plugins.themes.bootstrap3.article.main##

##plugins.themes.bootstrap3.article.sidebar##

Published Jul 22, 2020
Lily Geraldine Koops

Abstract

Prescriptive Maintenance strategies are emerging as potential next level of reliability and maintenance best practice. Likely outcomes of maintenance alternatives and their effects on e.g. cost and safety are comparatively evaluated by exploiting various sources of data, knowledge and models. By this means, optimized courses of actions are recommended to quickly resolve problems and to automate Maintenance, Repair and Overhaul (MRO) decisions. In this work, the key question is pursued as to how their dependability and potential business advantage can be assessed and improved in the presence of uncertainty and variability of various decision-influencing factors such as degradation and maintenance model parameters and cost sources. For this purpose, a step-by-step procedure to optimal solution prescription and potential / risk assessment is developed based on a probabilistic approach to cost-benefit analysis and on the definition of relevant metrics. By the help of a Wiener process degradation model capable of implementing random effects of imperfect repairs and a Monte Carlo simulation, its value is illustrated by a use case example – repair / replacement decision support in the aeronautical context. The probabilistic approach not only allows to determine, which decision option promises the higher profit and is thus preferred, but also with which risk and potential cost disadvantage it is associated. Furthermore, it uncovers, where higher-quality data or information, can gainfully reduce result uncertainty and hence be assigned a monetary value. It is argued that the presented approach could give industry practitioners directions for identifying and optimizing business cases for Prescriptive Maintenance, by pointing at which sources of data or information are particularly valuable and hence justify dedicated investments for acquiring it. The relevance of the results is discussed specifically with reference to emerging digitized and automated repair processes as well as more generally in the context of future data-trading schemes.

How to Cite

Koops, L. G. (2020). Optimized Maintenance Decision-Making – A Simulation-supported Prescriptive Analytics Approach based on Probabilistic Cost-Benefit Analysis. PHM Society European Conference, 5(1), 14. https://doi.org/10.36001/phme.2020.v5i1.1269
Abstract 2393 | PDF Downloads 339

##plugins.themes.bootstrap3.article.details##

Keywords

Predictive and Prescriptive Maintenance, Optimized Maintenance Decision-Making, Potential and Risk Analysis, Wiener Degradation Process, Imperfect Repair, Repair / Replacement Decision Support

Section
Technical Papers

Similar Articles

1 2 3 4 5 6 7 8 9 10 > >> 

You may also start an advanced similarity search for this article.