I’d like to report an issue I’ve encountered with the optimization engine. I’ve noticed that the optimization results differ significantly when processing multiple orders for multiple vehicles compared to when processing a single vehicle with a list of tasks.
Here’s an example:
I input 30 vehicles and 700 orders. The engine generates several routes, including one with a single vehicle and a sequence of checkpoints. The estimated route duration is 18 hours.
I then submit only that specific route back into the engine with the exact same parameters. The result? An estimated time of 12 hours with a reordered sequence of checkpoints.
This discrepancy is concerning. Ideally, I’d like to avoid having to reprocess each individual route and trust the initial optimization results.
Is this a known issue? Are there any settings or configurations I can adjust to ensure consistency between the two optimization scenarios?
Any insights or assistance would be greatly appreciated.
Thanks!