Optimizing maintenance strategies through data-driven analysis: case study of a manufacturing company in South Africa Pietermaritzburg.

Abstract

The increasing complexity of industrial systems and pressure on businesses to achieve operational excellence have made maintenance a strategic function in modern manufacturing. This study examined how data extracted from IBM Maximo can be leveraged in the process of evaluating and optimizing maintenance strategies within a South African manufacturing firm. The research focused on eight key production departments within the organization and analysed historical maintenance data spanning the period 2021 to 2025. The study adopted a quantitative, postpositivist research approach and employed structured methods to extract, process, clean and integrate workflows to transform raw maintenance records into analytical datasets. Descriptive statistics, correlation and regression analyses were applied to uncover the relationships between maintenance activities and asset reliability. The findings revealed significant variations across departments in maintenance workload distribution, work type composition, and asset performance. Preventive maintenance was found to increase breakdowns in the month of execution but demonstrated lagged reliability improvements in subsequent months. At the same time, predictive maintenance was underutilized, resulting in statistically insignificant effects. Corrective maintenance exhibited the most significant immediate impact on breakdown frequency, increasing failures during the month of execution. The study concluded that (Computerised Maintenance Management System (CMMS) data holds great potential for driving continuous improvement when converted into actionable insights. The observed interdepartmental differences and maintenance behaviour patterns formed the foundation for recommending a targeted approach to reliability enhancement. A data-driven feedback loop is proposed to support maintenance teams in refining task intervals, focusing attention on high-risk assets, and systematically tracking the long-term impact of maintenance interventions.