Stochastic maintenance models for ceramic claddings

Abstract

Maintenance decision-making involves a series of multiple objectives, some of them contradictory. Usually, stakeholders intend to find the optimal maintenance strategy, to minimize the economic burden, while simultaneously maximizing the buildings’ performance. In this study, a condition-based maintenance model, based on Petri nets, is proposed to evaluate the consequences of alternative maintenance strategies to maintain and improve the performance of ceramic claddings. This maintenance model is a full life-cycle model that integrates the stochastic assessment of the degradation condition of the claddings, and also inspections, maintenance and renewal processes. Three maintenance strategies are considered: (i) major intervention only; (ii) combination of minor and major interventions; and (iii) combination of cleaning operations, minor and major interventions. The uncertainties associated with the degradation process, as well as with the definition of the effects of maintenance actions are considered by modelling the transitions times in Petri nets as random variables. Considering the complexity of Petri nets, the statistical descriptors of the performance of the assets (e.g., mean condition, probability of applying maintenance) were computed using Monte Carlo simulation. The impact of the different maintenance strategies in the claddings’ service life is discussed, comparing the different alternatives also from an economic point of view.