JaamSim 2024-01 released

[Harry King]

Changes in release 2024-01:

• XYGraph object. This object allows line and bar graphs to be displayed for any set of x-y points. Its primary use is to display the histograms recorded by Statistics, Queue, and Resource objects.
• ‘Statistics and XYGraph Example’. This new example model under Help > Examples demonstrates the use of the new XYGraph to display a histogram of waiting times collected by a Statistics object.
• Cosmetic changes to Graph. The inputs LineColours, LineWidths, SecondaryLineColours, and SecondaryLineWidths have been moved to a new ‘Format’ tab in the Input Editor. New default values in hours have been set for the inputs in the ‘X-Axis’ tab.
• ‘QueueLengthFractions’ and ‘QueueLengthCumulativeFractions’ outputs. These new outputs for Queue return the fraction of time and cumulative fraction of time that the number of entities in the queue is 0, 1, 2, etc.
• ‘UnitsInUseFractions’ and ‘CumulativeUnitsInUseFractions’ outputs. These new outputs for Resource, ResourcePool, and ResourceUnit return the faction of time and the cumulative fraction of time that the number of resource units in use is 0, 1, 2, etc.
• ‘HistogramBinCumulativeFractions’ output for Statistics. This output returns the cumulative distribution for the existing ‘HistogramBinFractions’ output.
• ‘HistorgramBinUpperLimits’ output for Statistics. This output returns the value corresponding to the right-hand side of each histogram bin. It is intended to be used with the ‘HistogramBinCumulativeFractions’ in the inputs to an XYGraph that displays the cumulative distribution for a histogram.
• ‘TargetPercentiles’ input and ‘PercentileValues’ output for Statistics. This input and output allow the user to specify one or more percentiles values to be estimated from the histogram results. For example, entering 90 95 to the TargetPercentiles inputs would cause the PercentileValues output to return the values that are larger than 90% and 95% of the recorded sample values, respectively.
• ‘StandardDeviationOfTheMean’ output for Statistics. This output has been deleted because it is only valid when the recorded values are normally distributed.  This is rarely the case for the statistics collected from a single simulation run, which made this output misleading.
• Automatic updating for SubModel. When a new object is dragged and dropped to a SubModel, any clones of this SubModel are updated automatically to include a clone of the new object.
• Input Conditioning. Single quotation marks are now added automatically to inputs that accept a list of expressions when they are entered in the Input Editor or Input Builder.
• ‘Location’ input for distributions. This input has been added to BetaDistribution, LogLogisticDistribution, and LogNormalDistribution.
• Bug fix for LogLogisticDistribution. This distribution now accepts a Shape input that is less than or equal to 2.
• Bug fix for View windows. The camera position controls now work correctly when the initial camera angle is parallel to the x-y plane.
• Bug fix for Object Selector. Right-clicking on an object the Object Selector now generates the context menu more reliably. Formerly, the context menu appeared only about 50% of the time.
• Bug fix for ExpressionLogger. A log entry scheduled for the end of a simulation run is now recorded correctly. A bug introduced in release 2023-09 had caused this entry to be omitted.
• Bug fix for LogNormalDistribution. The ‘CalculatedStandardDeviation’ output now returns the correct value when the ‘Scale’ input is set.
• Bug fix for multiple threads. Bugs related to the SubModel object and the ‘RandomSeed’ input have been corrected that could prevent a model from executing on multiple threads.

Harry

[Karim Van den broek]

Thanks for the update, Harry. All the best for 2024! 

Op donderdag 4 januari 2024 om 03:40:05 UTC+1 schreef d.harr...@gmail.com:

[Paul René]

Hello! I am having a somewhat surprising result with the Percentiles of the Statistics entity (although maybe I understood them wrong):

My Statistics entity records the Count of the containers passing through (with this.obj.Count), and due to my AddTo rules, this Count cannot be higher than 100.

This is reflected in the bulk results:

However, the PercentileValues look weird:

If I understood correctly, this should mean that a Count of 101.608 would be higher than 90% of the recorded values, which can be restated as "90% of the sampled values are lower than 101.6", which then means that 10% of the values are higher than 101, even though the actual highest value sampled is 100?

Thanks in advance for the clarification, and thanks for the software!

Paul