This functionality would be very useful for iterative calculations over the same cube rules, or for Monte Carlo scenario simulations with randomly-generated changes in specified input parameters. It could also be quite useful for AI functionality in running over a number of scenarios, with each iteration # allowing a new set of input parameters to be run through the cube rule model. This is a great use case, by the way, for GPU functionality, since such calculations can be quite hardware-intensive.
A related piece of functionality would be to store the outputs of each interation into a designated "Scenario" or "Iteration" type dimension, where the input parameters of each scenario could also be stored or randomly generated, perhaps in a single dimension attribute (in a text-delimited list) or in multiple attributes (one for each parameter). The number of iterations for the circular calculation could actually be derived from the number of base level members under the Scenario dimension parent member chosen as the source of the Scenario parameters.
