Preventative maintenance optimisation in a capital-constrained environment

Abstract

Due to the recognition of the importance of maintenance from an organisational perspective, a number of different maintenance-related approaches have been developed. These approaches include reliability centred maintenance, business-centred maintenance, total productive maintenance and life cycle costing. They consider maintenance from specific different viewpoints and no single approach can be applied to all circumstances. Common to all these approaches are techniques to optimise the maintenance strategies using mathematical models. A variety of mathematical approaches are described in the literature, all of which involve the minimisation of the total costs incurred in relation to the required maintenance activities. This study focuses on data-driven optimisation models that consider costs and the reliability performance of equipment. The practical implementation of these optimising maintenance models presents two main challenges. First, the decision on when to use which model would depend on the type of system/equipment under consideration, as well as on available data. Different models based on analysing the historical failure data of the system or component are considered in order to optimise the maintenance strategies to be applied to these two types of individual systems. In the case of having a number of identical components or systems in series, where a shutdown of one of the systems results in the shutdown of the entire series, models are considered to allow for analysis with the correct maintenance technique of components or systems showing these trends. A major limitation of these maintenance optimisation models is that they all require failure data for their implementation, which is not always obtainable. Historical maintenance cost data, however, is mostly available, therefore forecasting techniques and life cycle cost modelling are also considered. Second, the successful implementation of optimised maintenance strategies will be dependent on informed budgetary decisions being made. Therefore, the challenge of integrating the outputs from the variety of optimisation models utilised into a cohesive compilation and sensible presentation of an overall maintenance budget for a complex plant needs to be addressed. This study presents an integrated maintenance optimisation model that uses the appropriate sub-models described individually in the literature to enable the integrated compilation and sound presentation of an overall maintenance budget for a complex plant for appropriate decision-making. iii The use of the case study validates this methodology. It illustrates that a concise, integrated overall budgetary maintenance decision model is highly beneficial in communicating the budgetary requirements for an organisation. It was found that the outcome resulted in an effective decision-making tool with significant potential for implementation in a variety of organisations in search of optimal maintenance planning and budgetary requirements.