Stochastic Deterministic Step Type

Units:	-
Default Value:	0
Validation Rule:	In (0,2,3,4)
Description:	The step type for solving the final deterministic pass of the rolling horizon where "None" solves the entire horizon in one step

This setting applies to the final deterministic pass of a stochastic optimization solved with Rolling Horizon. The deterministic pass is performed after the rolling iterations and involves only the full branch samples (the final set of samples reported on). This pass sets storage trajectories based on the optimal stochastic outcome from the rolling iterations. Because this pass only solves the full branch samples by default it solves the entire horizon in a single step, but this option allows you to split the solve into steps for the case where the horizon is very long and solving the entire horizon would be slow or consume too much memory. The option can take the following values:

"None" (value = 0, default)

The deterministic pass solves in a single 'step' covering the entire horizon. The samples in this single step can be split into a number of batches - see Deterministic Batch Count.

"Week" (value = 2)

The deterministic pass solves in steps of Deterministic Step Length weeks at a time.

"Month" (value = 3)

The deterministic pass solves in steps of Deterministic Step Length months at a time.

"Year" (value = 4)

The deterministic pass solves in steps of Deterministic Step Length years at a time.

Note when solving in steps that the step length does not need to match up perfectly with the number of time periods in the horizon.