Performance, cost, and risk
What are the performance, cost, and risk impacts of implementing this product?
- Performance: Improved maintenance schedules to enable just-in-time maintenance for a wide set of asset types, accounting for dependence between components.
- Cost: While operating and replacement costs should be decreased, the replacement decisions may still not be optimal and substantial costs may be involved with collecting the required data.
- Risk: Requires high-quality data to produce accurate advice, and solving the large-scale models associated with these problems is challenging.
Implementation requirements
What capabilities would a business/organization/institution need to have to implement this product?
- Processes: Preventive maintenance planning procedures should be mature when moving towards dynamic planning policies.
- Resources: Historic maintenance schedules and asset health data, computational infrastructure, maintenance planners.
- Competences: Knowledge of DRL, knowledge of dynamic planning procedures.
- Technologies: Graph convolutional networks, Markov decision process tools, asset management platform.
Related works
- Bukhsh et al. (2023). A Maintenance Planning Framework using Online and Offline Deep Reinforcement Learning.
- Jimenez-Roa et al. (2024). Maintenance Strategies for Sewer Pipes with Multi-State Deterioration and Deep Reinforcement Learning.
- Kerkkamp et al. (2022). Grouping of Maintenance Actions with Deep Reinforcement Learning and Graph Convolutional Networks.
- Eggertsson, Basten, et al. (2023). Maintenance optimization for capital goods when information is incomplete and environment-dependent.
- Eggertsson, Eruguz, et al. (2024). Maintenance optimization for multi-component systems with a single sensor.
- Kerkkamp et al. (2022). Grouping of Maintenance Actions with Deep Reinforcement Learning and Graph Convolutional Networks (Fact Sheet).
- Eggertsson (2024). Advances in Asset Management (Thesis).
Contact information
For further inquiries regarding this product, feel free to get in touch with:
- Rob Basten, Eindhoven University of Technology. r [dot] j [dot] i [dot] basten [at] tue [dot] nl
- Zahara Bukhsh, Eindhoven University of Technology. z [dot] bukhsh [at] tue [dot] nl
- Nils Jansen, Radboud Universiteit. nils [dot] jansen [at] ru [dot] nl
- Ahmadreza Marandi, Eindhoven University of Technology, a [dot] marandi [at] tue [dot] nl
