Hi,
I am writing to request guidance on the best approach for a logistics optimization problem I am currently addressing using OR-Tools.
I need to schedule delivery and pickup services for a fleet of vans over a multi-day horizon (approximately 3 days). My scenario involves about 100 services per day, and I need to account for the following constraints and requirements:
- Multi-day Planning: Efficiently managing services across a 72-hour window.
- Capacity & Time Windows: Respecting the maximum weight capacity of the vans and specific time windows for each location, including lunch breaks/closures of the sites being served.
- Service Deadlines: Each service has a specific expiration date that must be respected.
- Cost Minimization: The primary objective is to minimize overall operational costs.
- Determinism: I am looking for a way to achieve deterministic results, ensuring that the same input consistently produces the same output.
- Explainability: I require a solution that is "explainable" or easy to interpret for stakeholders.
Given these parameters, could you recommend whether the VRPTW (Vehicle Routing Problem with Time Windows) library is the most suitable tool, or if a different modeling approach within OR-Tools would be more effective for this scale and complexity?
Thank you for your time and for any insights or documentation you can provide.
Best regards,
Simon