Accumulated Reward Solver

From Mobius Wiki
Jump to: navigation, search

Accumulated reward solver

The accumulated reward solver (ars) solves for transient interval-of-time variables, i.e., for intervals [t_0, t_1] for which both t_0 and t_1 are finite. The starting and stopping times are specified for each interval or time-averaged interval of time reward variable in the reward model. The ars solver gives the expected accumulated reward, as well as the expected time-averaged accumulated reward over the interval. The results are derived by uniformization. <xr id="fig:arsSolver" /> shows the editor for the accumulated reward solver and its available options and adjustable parameters. The options are similar to the transient reward solver options.


<figure id="fig:arsSolver">

SolverARS.png


<xr id="fig:arsSolver" nolink />: The Accumulated Reward Solver ars and available options and parameters.
</figure>


The output file contains the means, both time-averaged and accumulated, of the performability variables. It also contains additional information similar to that described for the trs solver.


Pitfalls and Hints

  • The ars solver is an extension of the trs solver, and the remarks for trs apply here as well.