Note
This package is under development. Expect things to break significantly in versions v0.0.*
.
PyJobShop is a Python library for solving scheduling problems with constraint programming. It currently supports the following scheduling problems:
- Machine environments: single machine, parallel machines, hybrid flow shops, open shops, job shops, and flexible job shops.
- Constraints: release dates, deadlines, due dates, sequence-dependent setup times, no-wait, blocking, and precedence constraints.
- Objective functions: minimizing makespan, total completion time, number of tardy jobs, and total tardiness.
You can find PyJobShop on the Python Package Index under the name pyjobshop
.
To install it, simply run:
pip install pyjobshop
The documentation is available here.
Tip
If you are new to scheduling or constraint programming, you might benefit from first reading the introduction to scheduling and introduction to CP pages.
PyJobShop uses OR-Tools' CP-SAT solver as its default constraint programming solver. CP-SAT is an open-source constraint programming solver and encompasses other technologies such as satisfiability and mixed-integer linear programming solvers. CP-SAT has consistently won the annual MiniZinc Challenge, demonstrating its superior performance on constraint programming models.
Besides OR-Tools' CP-SAT solver, PyJobShop also implements the IBM ILOG CP Optimizer, which is a commercial constraint programming solver.
To try out CP Optimizer, you can install PyJobShop with the optional cpoptimizer
dependency as follows:
pip install pyjobshop[cpoptimizer]
This installation comes with the free community edition of CP Optimizer, which only solves models with up to 1000 variables and 1000 constraints. Models beyond that size require the paid version or academic version. See our documentation for instructions on how to obtain the academic version of CP Optimizer.
We provide example notebooks that show how PyJobShop may be used to solve scheduling problems.
- TODO
We are very grateful for any contributions you are willing to make. Please have a look here to get started. If you aim to make a large change, it is helpful to discuss the change first in a new GitHub issue. Feel free to open one!
Feel free to open an issue or a new discussion thread here on GitHub. Please do not e-mail us with questions, modelling issues, or code examples. Those are much easier to discuss via GitHub than over e-mail. When writing your issue or discussion, please follow the instructions here.