Page 327 - ExtendSim User Guide
P. 327

Statistics and Model Metrics 301
Timing the flow of items in a portion of the model
When using the Mean & Variance block, carefully consider the type of statistics that you want to calculate. Some guidelines for whether or not to select the time weighted statistics option are:
• If the value that you are collecting statistics on has a value at every point in the simulation, enable time weighed statistics. A good example of this is the number of items in a block (reported by the L connector). At any point in the simulation, there are a certain number of items in a block. To determine the average value, weight it over time.
• Ifthevaluethatyouarecollectingstatisticsononlyhasavalueatspecificevents,donotuse time weighted statistics. An example of this is the W or wait time connector. This connector only has a value when an item leaves the block, which is a specific event. In that case, time weighted statistics should not be used.
Time Weighted Statistics model
The Time Weighted Statistics example shows the difference between the two methods of cal- culating statistics and how they are calculated.
Time Weighted Statistics model
Comparing the weighted and un-weighted approaches to the average queue length reported in the Queue block’s Results tab, it is clear that not using time weighted statistics would give an incorrect answer for this model.
Timing the flow of items in a portion of the model
In addition to performance information that is directly available in a model, you may want to determine cycle time – how long it takes an item to go from one part of the model to another. For example, you may want to know how long a customer waits in line to place an order, or how long it takes that customer to get served once the order is placed.
To see how to determine cycle time in a model, see “Cycle timing” on page 312.
Discrete Event

   325   326   327   328   329