production scheduling.
Assign jobs to machines in sequence. Keep due dates, respect setup times, shift patterns and maintenance windows, and keep machines busy.
Which job runs on which machine, in which order, so the due dates hold with the fewest changeovers?
Job shop and flow shop scheduling are among the hardest combinatorial problems. kint models them as constraint programs or mixed-integer programs, depending on the structure. A CP solver handles discrete scheduling, MIP handles continuous-time models, and large instances combine both.
where production scheduling gets hard.
Switching from product A to B can take longer than switching from B to A. That asymmetry makes sequencing much harder than plain assignment.
A job may need machine 3, then machine 7, then machine 3 again. Blocking and waiting spread across the whole shop floor.
Short makespan, high throughput, few changeovers and met due dates pull in different directions. Improving one usually costs another.
from the rules to the bound.
kint's decision agent writes the model from the planners' description. In the pilot, kint's forward deployed engineers review it with the experts. Then the solver does this.
model jobs and machines
Jobs, machines, time slots, setup matrices and shift patterns become variables and constraints in one formal model.
search with constraint programming
The CP solver removes impossible assignments early through propagation. The search then looks for the best sequence among those that remain.
combine methods for large plants
For very large instances kint combines constraint programming with MIP relaxations to get tight bounds faster.
deliver the schedule with its gap
The result is a Gantt schedule with a stated quality: how close it is to the best schedule possible under the rules.
a production scheduling plan, with its reasons.
Below are the parts of a production scheduling result. Planners check them before the plan is released.
the anatomy of an answer- planA sequence per machine and shift, for release by the planners.
- binding limitFor example: the changeover on one bottleneck machine decides the makespan.
- sensitivityWhat an extra shift, or a shorter setup on that machine, would change.
- bound and gapBound and gap for the schedule, stored with the run.
- recordData version, model version, who reviewed the model and who released the plan.
bring us this decision.
Bring one recurring production scheduling decision with its rules and data. Your planners get a first model they can read.
- Results
- Plan, binding limit, sensitivity, bound and gap, record
- Data centres
- Operated in Germany, or in yours
- all problem types
- routes, schedules, stock and price
- Contact
- hello@kint.pro