Data Driven Maintenance Efficiency

Abstract

Significant opportunities for improvement were identified within the maintenance engineering department of a leading biopharmaceutical manufacturing firm. The primary focus is on addressing a known issue where preventive maintenance execution is unevenly distributed throughout the month. During the initial weeks, engineering technician teams tend to execute tasks at an arbitrary pace, only to hastily expedite their efforts towards the end of the month, leading to an accumulation of work. The project aimed to streamline and optimize preventive maintenance practices by enhancing reliability and balancing the workload among technicians. Historical preventive maintenance data was retrieved from the firm's computerized maintenance management system (Maximo 7). This data was then analyzed and pre-processed to uncover key correlations that could be leveraged to better distribute the workload across the month, thereby ensuring optimal processing of tasks. After identifying improvement opportunities, graphical aids were created to represent the analyzed data and presented to management, enabling the identification of teams that were underperforming compared to their peers. These visual tools were also used to continuously measure the outcomes of the implemented interventions. Although further improvement is still needed, positive progress was achieved, leading to an overall enhancement in preventive maintenance practices. Key Terms ⎯ Computerized Maintenance Management System; PM Hours; Python for Data Analysis; Seaborn.