ASCP

Airline Scheduling Crew Pairing

Environmental Setting

OptaPlanner Execution Guide

1. Navigate to the ASCP/PairingCreater directory.
2. Run ./gradlew build to generate the crew-pairing.jar file in build/libs/.
3. Move crew-pairing.jar to the ASCP/PairingCreater/ directory.
4. Use the following command to execute OptaPlanner:

java -jar crew-pairing.jar data/ crewpairing/ {datasize} input_{datasize}.xlsx

• If an initial set (output.xlsx) exists, use:

java -jar crew-pairing.jar data/ crewpairing/ {datasize} input_{datasize}.xlsx output.xlsx

5. Ensure that the output folder is added to data/crewpairing/.

Reinforcement Learning Execution Guide

1. Add an input file to the dataset folder within the project root directory.
2. Run REINFORCE.py to train the model.
3. More details on ReinforcementLearning/README.md

Python Packages for Reinforcement Learning

Using requirements.txt

1. Enter your local or virtual environment (Python 3.8 or higher).
2. Run the following command:

pip install -r requirements.txt

PyTorch - CUDA 11.8

Refer to the Download page for PyTorch installation.

License

This project follows the licensing guidelines applicable to OptaPlanner and any relevant dependencies.

GitHub Role

The following conventions should be adhered to when working on this repository.

Commit Convention

• feat: Add a new feature
• fix: Fix a bug
• docs: Modify documentation
• style: Code formatting, missing semi-colons, etc. (no code changes)
• refactor: Refactor code
• test: Add or refactor tests; no production code changes
• chore: Modify build tasks, package manager configurations, etc.

PR Convention

Icon	Code	Description
🎨	:art	Improve structure/format of code
⚡	:zap	Improve performance
🔥	:fire	Remove code or files
🐛	:bug	Fix a bug
🚑	:ambulance	Critical hotfix
✨	:sparkles	Introduce new features
💄	:lipstick	Add/update UI or style files
⏪	:rewind	Revert changes
🔀	:twisted_rightwards_arrows	Merge branches
💡	:bulb	Add/update comments
🗃	:card_file_box	Modify database related files