Supplementary material for our paper "Deep Reinforcement Learning for Maintenance Planning of Offshore Vessel Transfer" at RENEW 2020, Lisbon, Portugal.
Initially, SMOTE can be performed using the appropriate script. And as the second step, utilise the fault prediction using XGBoost with SHAP game theory approach to predict fault types. Finally, utilise the simulated RL enviornment in the grid map based on the Taxi Problem OpenAI Gym and train the RL algorithms of your choice. You can utilise Tensorflow for training DQN, SARSA, Expected-SARSA and Q-Learning Models.
Refer to https://youtu.be/ss02F5vwojM for a short video demonstrating text-based simulation of offshore vessel transfer planning and decision making.
ORE Catapult for providing us insights from the LDT's data via Platform for Operational Data https://pod.ore.catapult.org.uk.
This study uses a hypothetical environment for learning process during reinforcement learning's policy updates, so the rewards and the learning agent may perform differently under varying circumstances of fault types predicted by XGBoost and maintenance actions mapped to priorites in the scale of 0-4.
This repo is based on the MIT License, which allows free use of the provided resources, subject to the origina sources being credit/acknowledged appropriately. The software/resources under MIT license is provided as is, without any liability or warranty at the end of the authors.