Lot Sizing Problem

This repository contains solving approaches, examples and benchmarking tools for the discrete lot sizing problem. This is material for the Satalia-internal optimisation challenge.

The Lot sizing problem is described in detail as problem #58 in the CSPlib: http://www.csplib.org/Problems/prob058/

Solving Approaches

Constraint Programming approach with Gecode

In the gecode/ folder you can find the C++ source code for a CP approach using the CP library solver Gecode. It includes an exact approach, as well as a large neighbourhood search (LNS) approach, and it takes as input the examples in text format found in the examples/txt/ folder.

For more details, check out the README.md in the gecode/ folder.

MIP and CP MiniZinc models

In the minizinc/ folder there are two different problem models:

(a) a Constraint Programming model in minizinc/cp/lot_sizing_cp.mzn

(b) a Mixed Integer Programming model in minizinc/mip/lot_sizing_mip.mzn

You can use them to solve the problem with open-source MiniZinc and one of the CP or MIP solvers that are shipped with it. The models take as input the .dzn files in the folder examples/dzn/. The CP model also comes with a solution validator.

Problem examples

Problem instances can be found in the examples/ folder.

Running benchmarks

For running benchmarks on the CP approach with Gecode, check out the benchmarking/ folder.