POSICIÓN ABIERTA
AI Generated Penalties in Route Optimization
Master Thesis · Linköping
Description
This position focuses on route optimization challenges where penalties and costs must be defined strategically. The solver aims to find the lowest-cost solution, and penalty selection directly influences the final outcome. The role involves investigating how artificial intelligence can examine historical solutions to refine penalties and scoring mechanisms for the optimizer. A proof-of-concept will be constructed using actual school run data.
Key Responsibilities
- Analyze historic solutions from real-world school run operations
- Develop AI models to fine-tune optimization penalties
- Create a proof-of-concept system trained on live transportation data
- Investigate how machine learning can improve penalty calibration
Technical Requirements
- Machine Learning
- OR-Tools
- Linux
- Optimization