Model with Symbolic Parameters

As mentioned previously, the tool supports the use of symbolic parameters. Symbolic parameters can be used in expressions that specify a distribution, expressions that specify the probabilities of an action, and expressions that define the value of a reward. For example, the rate of an exponential distribution (EXP) can be specified using an expression that contains a parameter. In the same way, the probability of an action can be specified using an expression that contains a parameter. Also, when specifying rate or s, the value of the reward can be specified using an expression that includes a parameter. Symbolic parameters cannot be used in conditions or inside actions.

Parameters are useful for sensitivity analysis . Suppose that we want to study the influence of different event rate values. Then, if we define the rate of events as a parameter, the chain will be generated only once. The transition probabilities of the chain are defined as functions of the parameters. The user can choose a value for these parameters and the transition probability of the chain is easily calculated by the tool without recomputing the Markov chain. Parameters are set in the interface of the Analytical Model Solution. In the proper option of the interface, the user can define the numerical values for the parameters and generate the chain. The rewards are also calculated after the specification of the parameters. It is necessary to specify all parameters before solving the model.

If simulation is used, no symbolic parameters can be specified. When simulating a model, the state space is not generated and the simulator calculates the samples for the events and the probabilities of the action based on the numerical values given.
IMPORTANT: A model with parameters can not be simulated!