Page 687 - ExtendSim User Guide
P. 687

Analysis 661
Scenario analysis
• Two responses in its Dialog Responses table: average wait time and throughput This tutorial will keep these dialog factors and responses but will add a fifth factor (the per-
centage of cars wanting Wash or Wash/Wax) and put it in the Database Factors table.
In the model, the proportion of cars wanting either wash only or wash and wax is set in the empirical table of the Random Number block (labeled “Preference Percentage”). The table is shown at right.
For this tutorial the Wash or Wash/Wax factor will have three com- binations of percentages, which represent 3 levels:
1) 75%washand25%wash&wax–thebasecase
2) 90%washand10%wash&wax–themorewashcase 3) 50%washand50%wash&wax–thelesswashcase
As in the earlier tutorial, each scenario will be run 5 times to capture the variation in the results between one run and the next. To evaluate all of the possible combinations now requires 2,025 replications (405 scenarios ((3 x 3 x 3 x 5 x 3 levels)) at 5 runs each).
Create a database structure for the model
As shown at the bottom of the Database menu, the Scenario Manager Final Car Wash model already has two databases:
• TheScenarioDBdatabase.ThiswasautomaticallycreatedwhentheScenarioManager block was added to the model; it is discussed on page 657.
• AModelDatadatabase.Sothatyoucanfocusonthetutorial,therequireddatabasehas already been created for this model. The Model Data database stores the different options and probabilities for the wash or wash/wax factor in 6 tables:
• AParenttable,WashType,thatliststheoptionsforthetwowash types: wash only and wash & wax. Its contents are shown on the right.
• Threefactorsourcetables,eachofwhichhasoneoftheprobability
combinations: Base Case (75% wash/25% wash&wax), Less Wash (50%/50%), and More Wash (90%/10%). These are Child tables that get their wash type information from the Wash Type table.
• AfactortargettablenamedWashPreference.Eachrecordinthefactortargettablehas a wash type field and a probability field associated with that wash type. This table is used as a placeholder for model data. Thus before the scenario runs begin, the table is identical to the Base Case source table.
• Afactorsourceslisttable,namedPreferenceSources,thatisa
list of the three factor source tables. The contents of this table
is shown at right. This table is used when performing DOE, so
that you will know which tables to select from when creating
the design. See “Create a source list (optional)”, below, for more information.
How To

   685   686   687   688   689