Quantifying Preventive Maintenance Efficacy: A Baltimore City Use Case
Existing preventive maintenance efficacy research heavily focuses on quantifying system degradation in deterministic, probabilistic, and policy-based models, yet in a data centric age they inadequately address data requirements, the linchpin for improving preventive maintenance and graduating to predictive maintenance. Using Baltimore City’s public facility maintenance work orders, this study demonstrates the impact of data requirements on frequency, time and cost key performance indicators (KPI) and addresses omitted variable bias introduced by lack of condition-based data. Overall results show that maintenance cost has annually increased by $6,520 despite a sharp drop in 2018. Facilities in poor condition with persistently high repair needs present an opportunity for Baltimore to tailor its preventive maintenance strategy using condition-based data and separating money for a narrower definition of functional maintenance, potentially making better use of up to $14 million. Data requirements such as tracking corrective and preventive maintenance work for the same system and parts, combining facilities condition index (FCI) scores with work order frequency, prioritizing which facilities get preventive maintenance using return on investment thresholds, and enforcing data quality discipline compose the road map to these insights.