Case study · India
Production scheduling for a textile manufacturer
A constraint-based scheduler built on Google OR-Tools that turns open orders and machine capacity into a production plan the floor can follow.
- Open orders and machine capacity
- Constraints and objectives
- OR-Tools CP-SAT solver
- Plan the floor can follow
The problem
Planning production at a textile manufacturer means fitting open orders onto a limited set of machines, each with its own capacity, against delivery dates customers expect to be met. Built by hand or in a spreadsheet, a plan like that takes a long time to make, is hard to check, and goes stale the moment an order changes.
What we built
A constraint-based scheduler. It takes open orders and machine capacity and produces a production plan the floor can follow: which order runs on which machine, in what sequence, and when.
How it works
In scheduling, a plan that looks reasonable and the best plan that is actually possible can be far apart. So instead of rules of thumb, the problem is written down as a model.
- Constraints are things that must be true: a machine runs one order at a time, and an order only runs on a machine that can make it.
- Objectives are things to make as good as possible, such as finishing orders by their due dates.
- The solver, Google OR-Tools CP-SAT, searches for the best plan that satisfies every constraint.
When something changes, the plan is solved again from the current state instead of being patched by hand.
Outcome
The plan comes from the same rules every time, so it no longer depends on who builds it. When orders change, a new plan is a re-solve away instead of an afternoon in a spreadsheet.
Stack
| Part | Tools |
|---|---|
| Solver | Google OR-Tools CP-SAT |
| Model | Machine capacity and order constraints, with due-date objectives |
Have a problem like this one?
We reply within one business day. If AI isn't the right fix, we'll say so.