Efficient maintenance of feedwater heaters in industrial facilities is essential
for optimizing energy consumption and minimizing operational costs. This
project addresses the logistical and cost challenges associated with managing
inspections, repairs, and preventive maintenance across multiple feedwater
heaters. The objective was to develop a cost-optimization model using a
Mixed Integer Programming (MIP) approach to streamline crew dispatch and
route planning.
The system integrates predictive maintenance techniques to prioritize tasks
based on real-time equipment health, ensuring timely interventions and
reducing downtime. Scenario-based testing demonstrated significant
improvements, including annual cost savings of up to 89.9% during peak
demand periods and emission reductions of 84.4% for new heaters. The
model's adaptability to various scenarios highlights its robustness and
potential for scalability.
The findings suggest that integrating AI for real-time data processing and
scaling the system to larger networks can further enhance operational
efficiency. This work establishes a framework for sustainable and costeffective
maintenance management in complex industrial environments.