The bin packing problem is an optimization problem in which items of different sizes must be packed into a finite number of bins or containers, each having a fixed given capacity, in a way that optimizes the problem depending on the desired outcome.
The goal in this case:
- An “item” is packed to at most 1 “bin”;
- Each “bin” must meet the requirements of loaded quantities: a lower bound & an upper bound;
- The profit gained from “packed items” is maximal.
Data provided:
- K number of vehicles/bins
- N number of items/orders. Each item has a weight (w) and a profit (p)
- Lower bound (c1) and upper bound (c2) of each vehicle
For more explanation, problem modeling and implementation, please refer to our report
We have implemented a solution for the bin packing problem using the following algorithms:
- Integer Linear Programming (ILP)
- Constraint Programming (CP)
- "Greedy" assigning method
- Local search algorithm for completion