Access Type

Open Access Embargo

Date of Award

January 2025

Degree Type

Dissertation

Degree Name

Ph.D.

Department

Industrial and Manufacturing Engineering

First Advisor

Murat Yildirim

Abstract

This study introduces a new generation of optimization models that inherently capture degradation dynamics and dependencies in multi-asset systems. In both industrial and power system applications, effective operations and maintenance (O&M) decisions require the integration of economic and degradation dependencies, which significantly impact system performance and asset lifetime. We propose robust optimization frameworks that embed sensor-driven degradation models into optimization models, allowing us to monitor and control degradation processes within the optimization model. These models seamlessly integrate predictive degradation models with optimization models, offering a comprehensive approach to optimizing O&M strategies across complex systems. In the first part, we focus on multi-asset industrial systems, where degradation interactions between assets and operational stress affect the lifespan of assets. Our framework models these dependencies, leveraging sensor data to optimize production, maintenance schedules, and failure risk management. We extend this work to power systems, where the integration of renewable energy and distributed generation introduces operational variability, leading to frequent start/stop cycling that accelerates asset degradation. Our extended model incorporates start/stop cycling as a key factor in degradation rates, optimizing unit commitment and maintenance schedules in dynamic power grid environments. Real-world computational experiments validate the advantages of this framework, highlighting its ability to balance lifetime utilization, failure risks, and operational efficiency in both power system applications.

Available for download on Friday, December 18, 2026

Share

COinS