Design of multi-component periodic maintenance programs with single-component models

ABSTRACT

Capital assets, such as wind turbines and ships, require maintenance throughout their long lifetimes. Assets usually need to go offline to perform maintenance, and such downs can be either scheduled or unscheduled. Since different components in an asset have different maintenance policies, it is key to have a maintenance program in place that coordinates the maintenance policies of all components, to minimize costs associated with maintenance and downtime. Single-component maintenance policies have been developed for decades, but such policies do not usually allow coordination between different components within an asset. We study a periodic maintenance policy and a condition-based maintenance policy in which the scheduled downs can be coordinated between components. In both policies, we assume that at unscheduled downs, a minimal repair is performed to keep the unscheduled downtime as short as possible. Both policies can be evaluated exactly using renewal theory, and we show how these policies can be used as building blocks to design and optimize maintenance programs for multi-component assets.

KEYWORDS:

• Multi-component systems
• condition-based maintenance
• group maintenance
• renewal theory
• delay time model

Additional information

Funding

The first author acknowledges support from the Netherlands Organisation for Scientific Research through a grant no. VENI 451-16-025.

Notes on contributors

Joachim Arts

Joachim Arts is an associate professor at the University of Luxembourg in the Luxembourg Centre for Logistics and Supply Chain Management (LCL) part of the MIT SCALE network. His research interests are in operations research and its applications in supply chain, logistics, maintenance, and business analytics. Much of his research is practice driven as he works with companies such as ASML, railway operators, Philips, and others. He is particularly interested in research that integrates estimation from data and optimization of decision making. He also studies sustainability issues in multi-tier supply chains.

Rob Basten

Rob Basten is an associate professor at the Eindhoven University of Technology (TU/e) where he is primarily occupied with maintenance and service logistics and its interfaces. Most of his research focuses on after-sales services for capital goods: maintenance policies and maintenance optimization, spare parts inventory control, and design of after-sales service supply chains. He is especially interested in using new technologies to improve after-sales services; for example, 3D printing of spare parts on location and using improved sensoring and communication technology to perform just-in-time maintenance. He is further active in behavioral operations management, trying to understand how people can use decision support systems in such a way that they actually improve decisions and add value. Many of Rob’s research projects are interdisciplinary and in cooperation with (high-tech) industry.