The need to utilize maintenance resources both effectively and efficiently has never been greater due to increasing budgetary constraints for maintenance activities within organisations. Maintenance prioritisation becomes essential to ensure that resources are spent on those assets where it does the most good. As a result, one of the vital tenets of maintenance and assets management is the belief that it is important for organisations to identify their “critical” assets. Criticality analysis, a well-recognised method for providing a systematic basis for deciding what assets should have priority within a maintenance management process, is used for this purpose. Evidence from industrial practices reveals that although asset criticality changes over time due to dynamic operating environments, current criticality analysis techniques both from literature and practice cannot deal with this dynamicity. Understanding the dynamic nature of asset criticality is therefore key to ensure maintenance plans are aligned to business needs. Insights from dynamicity of criticality will give decision makers the ability to better withstand, respond and recover from asset performance events, thereby improving the resilience of the asset portfolio. This thesis focusses on 1) understanding the dynamic nature of asset criticality, i.e. what factors influence changes in criticality; how and when criticality changes; how change in criticality is monitored and updated, and 2) the impact of dynamicity on maintenance prioritisation decisions, i.e. when and how does the change in criticality matter to the effectiveness of the maintenance decision. A methodology and its associated models, which can be used to understand and exploit the dynamicity of assets in order to optimise prioritisation decisions, is presented. A Scenario-based Weighted Sum Method (SWSM) model is developed in this thesis to characterize both asset impact measure and criteria weights as dynamic variables that are dependent on the states of influence factors. Multi state importance measure (MSS-IM) is used to describe assets whose performance and impact measures are not simply binary (working; failed). Questionnaires, interview and surveys are developed based on a Soft Systems Methodology (SSM) approach to elicit expert knowledge for the identification of stakeholders’ requirements, factors influencing criticality and development of a system dynamic relationship between factors/criteria. Finally, a modified integer programming optimisation model is used to assess the impact of dynamic criticality on prioritisation decisions.
From a detailed case study of a waste water company, an analysis of the overall methodology identified the benefit of understanding and exploiting the dynamicity of asset criticality for maintenance prioritisation. The methodology combined with the models provides a practical procedure for industrialists to quantitatively assess the dynamics of asset criticality and to respond where appropriate. It enabled a sensitivity and volatility analysis which provide insights as to which assets are most robust and least robust to plausible emergent conditions of the influence factors. Results from the case studies have also demonstrated the contributions made by this thesis in improving criticality-based maintenance decisions for assets with dynamic criticality.