| Contents | Index |
In some cases, the updating of an output signal or the reaction of a block to updates in its input signal can experience a delay:
The update of an output signal in one block might occur after other operations occur at that value of time, in the same block or in other blocks. This latency does not last a positive length of time, but might affect your simulation results. For details and an example, see Interleaving of Block Operations.
The reaction of a block to an update in its input signal might occur after other operations occur at that value of time, in the same block or in other blocks. This latency does not last a positive length of time, but might affect your simulation results. For details, see Choosing How to Resolve Simultaneous Signal Updates.
When the definition of a statistical signal suggests that its value can vary continuously as simulation time elapses, the block increases efficiency by updating the signal value only at key moments during the simulation. As a result, the signal has a somewhat outdated "approximate" value between such key moments, but corrects the value later.
The primary examples of this phenomenon are the signals that represent time averages, such as a server's utilization percentage. The definitions of time averages involve the current time, but simulation performance would suffer drastically if the block recomputed the percentage at each time-based simulation step. Instead, the block recomputes the percentage only under these circumstances:
Upon the arrival or departure of an entity
When the simulation ends
When you pause the simulation using Simulation > Pause or other means
For an example, see the reference page for the Single Server block.
When plotting statistics that, by definition, vary continuously as simulation time elapses, consider using a continuous-style plot. For example, set Plot type to Continuous in the Signal Scope block.
![]() | Common Problems in SimEvents Models | Learning More About SimEvents Software | ![]() |

Model electronic system architectures, process flows, and logistics as queuing systems or agent-based systems.
Get free kit| © 1984-2012- The MathWorks, Inc. - Site Help - Patents - Trademarks - Privacy Policy - Preventing Piracy - RSS |