Instances of the STN model can be specified using .yaml
configuration files.
The following parameters can be used:
Parameter | Description | Scripts |
---|---|---|
Ts | Length scheduling horizon | all |
dTs | Time step scheduling horizon | all |
stn | Data file with stn structure | all |
rdir | Path to result directory | all |
solverparams | Dictionary of solver parameters | all |
max | Dictionary with maximum value for each product demand | lhs.py |
min | Dictionary with minimum value for each product demand | lhs.py |
N | Number of LHS samples | lhs.py |
Tp | Length planning period | rolling.py, mc.py |
dTP | Time step planning period | rolling.py, mc.py |
TP | Logistic regression data file | rolling.py, mc.py |
prfx | Prefix for result files | rolling.py, mc.py |
[Product] | List of demands for each planning period for [Product] | rolling.py, mc.py |
alphas | List of alpha values (uncertainty set size) for which to solve | rolling.py, mc.py |
periods | Dictionary: {rolling: # of rolling horizon periods, eval: # of periods to evaluate) | rolling.py |
periods | # of periods to evaluate | mc.py |
freq | True: Use frequency approach (if eval > rolling), False: MC approach | rolling.py, mc.py |
robust | True: solve robust model, False: solve deterministic approximation | rolling.py, bo.py |
ccm | Cost of corrective maintenance for each unit | bo.py |