Scenario
A scenario represents the specification of a model to run, with some inputs. Each scenario will produce a simulation run. In the context of running multiple scenarios, it is important that the scenario names be unique to permit automated storage within a KSL database. The name of the scenario is used to assign the name of the model's experiment prior to simulating the model. In this manner, each experiment can have a unique name.
Parameters
The model to be simulated
The map of inputs (based on control names) to apply to the model
The name of the scenario. It should be unique within the context of a set of scenario being executed by a ScenarioRunner.
the number of replications for the scenario. By default, this is the current setting of the model.
the length of each replication for the scenario. By default, this is the current setting of the model.
the length of the warmup period for each replication for the scenario. By default, this is the current setting of the model.
Constructors
Properties
The reset next sub stream option This option indicates whether the random variables used during the replication within the experiment will be reset to their next sub-stream after running each replication. The default is TRUE. This ensures that the random variables will jump to the next sub-stream within their current stream at the end of a replication. This will cause the random variables in each subsequent replication to start in the same sub-stream in the underlying random number streams if the replication is repeatedly used and the ResetStartStreamOption is set to false (which is the default) and then jump to the next sub-stream (if this option is on). This option has no effect if there is only 1 replication in an experiment.
Indicates whether antithetic replications should be run. The default is false. If set the user must supply an even number of replications; otherwise an exception will be thrown. The replications will no longer be independent; however, pairs of replications will be independent. Thus, the number of independent samples will be one-half of the specified number of replications
The identifier for the current replication. Each replication in the set of replications for the experiment has a unique identifier. This returns the identifier for the current replication. An identifier of 0 represents that no replications have been run
The current number of replications that have been run for this experiment
Holds values for each controllable parameter of the simulation model.
The identity of the experiment
The name of the experiment
Causes garbage collection System.gc() to be invoked after each replication. The default is false
The specified length of each planned replication for this experiment. The default is Double.POSITIVE_INFINITY.
The length of time from the start of an individual replication to the warm-up event for that replication.
The maximum allowable execution time "wall" clock time for an individual replication to complete processing in nanoseconds. Set the maximum allotted (suggested) execution (real) clock for a replication. This is a proposed value because the execution time requirement is only checked after the completion of each replication After it is discovered that cumulative time for executing the replication has exceeded the maximum time, then the process will be ended (perhaps) not completing other replications.
The number of replications to run for this experiment
Indicates the number of times the streams should be advanced prior to running the experiment
Indicates if the replications associated with this experiment are part of set of experiments with the same name. A chunk is an ordered subset of replications for an experiment. This field indicates the number of chunks within the experiment. There should be at least 1 chunk. If there is 1 chunk then all replications of the experiment are run as one chunk (simulation run)
The range of replication identifiers for the experiment based on the starting replication identifier and the number of replications to be executed.
A flag to indicate whether each replication within the experiment should be re-initialized at the beginning of each replication. True means that it will be re-initialized.
The reset start stream option This option indicates whether the random variables used during the experiment will be reset to their starting stream prior to running the first replication. The default is FALSE. This ensures that the random variable's streams WILL NOT be reset prior to running the experiment. This will cause different experiments or the same experiment run multiple times that use the same random variables (via the same model) to continue within their current stream. Therefore, the experiments will be independent when invoked within the same program execution. To get common random number (CRN), run the experiments in different program executions OR set this option to true prior to running the experiment again within the same program invocation.
Can be used to supply a function that will set up the model prior to being run. This would allow for the assignment properties or invoking of additional logic prior to simulating the model.
returns the last generated simulation run
The starting id for the sequence of identifiers used to label the replications of the experiment
Functions
Changes the experiment run parameters for the experiment. This does not include the current number of replications or the experiment's id. Any property in ExperimentRunParametersIfc may be changed.
Returns a new Experiment based on this experiment.
Extracts a data class representing the experiment's run parameters
Checks if the current number of replications that have been executed is less than the number of replications specified.
Sets the desired number of replications for the experiment