Current practice and challenges towards handling uncertainty for effective outcomes in maintenance

The combination of viable heuristic attributes with statistical measurements presents significant challenges in industrial maintenance for complex assets under through-life service contracts. Techniques to obtain and process heuristic attributes raise numerous uncertainties which often go undefined and unmitigated. A holistic view of these uncertainties may improve decision-making capabilities and reduce maintenance costs and turnaround time. It is therefore necessary to identify and rank factors that influence uncertainties originating from challenges in the above context. This, along with an identification of who contributes to such challenges and current practice to handle them, sets the focus for this study.

The influence of 32 categorised factors on uncertainty is assessed through a questionnaire completed by nine experienced maintenance managers from a leading defence company. The pedigree approach is applied to score validity of respondents’ answers according to their experience and job role to normalise scores. Results are discussed in interviews with respondents along with current practice in and ways to improve uncertainty assessment. Scores are weighted through the Analytical Hierarchy Process (AHP) in order to identify the most influential factors on uncertainty in maintenance. The analysis revealed that these include: intellectual property rights (IPR), maintainer performance, quality of information, resistance to change, stakeholder communication and technology integration. These are verified with 40 practitioners from various industrial backgrounds. From the interviews, it is deemed that a holistic view of heuristic and statistical attributes ultimately allows for more accomplished decision-making but requires trade-offs between quality and cost over the asset’s life cycle.