Abstract

In the competitive landscape of Industry 4.0, maintaining operational efficiency and minimising downtime is crucial for achieving production targets and reducing costs. Typical maintenance strategies include corrective maintenance (CM), preventative maintenance (PM) and predicted maintenance (PdM). While advanced technologies like the Internet of Things (IoT) and Machine Learning (ML) dominate current maintenance optimisation strategies for PdM, many companies require years to complete digital transformation fully. This research addresses the present needs of companies in the early stages of Industry 4.0 by enhancing maintenance strategies to improve productivity and cost-effectiveness. This project aims to develop an optimised PM schedule using historical data from Mars Petcare’s Bathurst factory. By employing advanced analytics and regression modelling techniques, we seek to identify the optimal PM intervals that balance maintenance costs with machine downtime, ensuring maximum factory availability and operational efficiency. Initial analyses have identified the pack line as the most critical section concerning costs, frequency, and duration. In the regression modelling, the cost of PM and CM are identified to have a close relationship with PM intervals, PM hours and the Machine. The equations are post-processed into plotting, delineating the relationship between PM frequency and total maintenance costs, and pinpointing the optimal intervals. By optimising the maintenance schedule, this project aims to reduce costs and downtime while laying the foundation for future predictive maintenance initiatives. The findings provide Mars PetCare and similar industries at the early level with a strategic tool for cost-effective maintenance management, contributing to enhanced operational efficiency and competitiveness.