Maintenance optimization and spare parts management in data-integrated environments

Capital goods have many components that require maintenance. Usually, a maintenance service provider is responsible for the maintenance activities of complex capital goods for which a high level of skill and knowledge is needed.

These responsibilities involve planning maintenance activities and providing spare parts. The high cost of downtime for capital goods infrastructure requires preventive maintenance before a failure or immediate repair after a failure.

Most of the time, the service provider must make decisions regarding maintenance activities and spare parts shipments to the customer in real-time. Real-time decision-making requires real-time coordination between different parties and data flows.

Some of these parties are multiple customers at different locations, service engineers, and spare part warehouses. It is important to estimate when a critical component will fail, which parts caused the system failure in the first place, and which spare parts might be needed. Historical maintenance data and data received from the system can provide better estimates for these decision variables.

Data-integrated environments

Thanks to Industry 4.0 and internet of things technologies, it is now easier to collect, store, and share data among different parties (i.e., customers, maintenance service providers, and service engineers). Complex capital goods often function in data-integrated environments where data regarding the health status of components, historical data on maintenance activities, and possible causes of failures can be easily collected.

There are opportunities (cost benefits) for maintenance service providers to integrate these technologies into their decision-making. For her PhD thesis, Ipek Dursun sought to create

knowledge on decision-making in data-integrated environments for maintenance optimization and spare parts management. Her research provides insights to practitioners on the (cost) benefits of adopting these new technologies.

With efficient maintenance operations, businesses prevent unnecessary costs of downtime, maintenance costs, shipment cost of spare parts, and inventory holding costs. Additionally, better-planned maintenance activities have a positive impact on the environment in terms of reducing the waste of useful lifecycle of parts, and reducing CO2 emissions caused by the shipment of unnecessary spare parts and unnecessary engineer visits.

Main Findings

The first part of Dursun’s research looks at maintenance optimization policies under the so-called failure model uncertainty, and the second part of her research is on spare parts management by using advance demand information from the point of view of a service provider of capital goods.

The research shows the potential benefits of integrating Bayesian learning and the pooling of data from multiple systems into optimal decision-making for maintenance. The research also shows the potential benefits of using advance demand information for the required spare parts.

The knowledge derived from Dursun’s research can be used in practical business problems concerning maintenance optimization and spare parts management in data-integrated environments. The models developed as part of her work aim to provide a basis for the decision-support tools that are utilized by maintenance service providers. 

Title of PhD-thesis: Maintenance optimization and spare parts management in data-integrated environments. Supervisors: Geert-Jan van Houtum and Alp Akcay. Other main parties involved: Daytime.