Optimize worker-role allocation using mixed integer programming to model objective and constraints.
Workers have specific workstation based skills. In absence of that specific skill, two workers are needed to work on a workstation. Workers could be absent unpredictably. Objective is to maximize company output based on given initial worker skills. If possible, worker preference should be taken into account with allocation. In addition, worker training is available to upskill workers for other workstations. Solution is devised to upskill workers efficiently in order to prevent factory stalling and maximize output.
Mixed integer programs are formulated and solved using the library PuLP.
All code and modeling is done in MOR_project_test.ipynb
.
Team skills go up because of the cross-training policy.
Crosstraining reduces the employee cost over time.
More teams are able to find an optimum solution when crosstraining policy is implemented.