Facilities planning technologies using graph structured deep reinforcement learning
Uses AI to reduce maintenance and renewal costs for electric power system facilities.
- A reinforcement learning approach that can be applied to capital investment planning and other cost minimization problems. Input data on optimization targets with complex structures into original graph neural networks to learn models using deep reinforcement learning frameworks and resolve reward maximization (cost minimization) problems.
- Create facilities renewal plans for electric power systems.
Benchmarks, strengths, and track record
- In numerical experiments using model systems, confirmed 10% reductions in capital investment costs + maintenance costs compared with Monte Carlo simulations.
Please include the title “Toshiba AI Technology Catalog: Facilities planning technologies using graph structured deep reinforcement learning” or the URL in the inquiry text.
Please note that because this technology is currently the subject of R&D activities, immediate responses to inquiries may not be possible.