Process Details

Process run-times

The first section of the report (after the model description) deals with the general statistics of the process(es). The information gives you an overview of the process(es)' lead time(s), which are visually enhanced by histograms or box-plots.

Histogram

The first figure in the process details depicts a histogram for the processing time of one of the processes (in this case each pool is a process). It simply shows the amount of process instances (Y-axis), which finished in a certain interval (X-axis).

In our example the length of the process lead time concentrates in the range 22 until 27 hours. (See the histogram in Figure 1)
So, this first glance at the report tells us, that we need to find out, where the process instances out of range originate from.
The process instances with a lead time less than 22 hours are not our problem, but they might be the key to shift our cluster (22-27 hours) to the left.
- Let us just assume that the shorter the lead times of the process the better.
The process instances on the right side of our cluster are the problem that need to be dealt with. So let’s go further into it!

Figure 1. Report: Process details - Histogram for the processing time

Figure 2 below shows the statistics for the process(es)' lead time(s).

See the extra table for more information on the statistical elements.

Figure 2. Report: Process details - Process run-times (in table representation)

Process runtimes (see table above and the box-plot below):

Title: Name of the process.
Obs: The number of finished process instances.
Mean: The mean value of the process' lead time.
Std.Dv: The standard deviation of the process' lead time.
Min: The minimum value of the process' lead time.
Max: The maximum value of the process' lead time.
Lower Quartile: The lower quartile of the process' lead time.
Median: The median of the process' lead time.
Upper Quartile: The upper quartile of the process' lead time.

Box-Plot for the process run-times

Additionally to the histogram we provide a box plot that represents the table above (Process runtimes). Notice that the cluster we discovered in the histogram is also illustrated within this box plot (the range from lower quartile to upper quartile).

Figure 3. Report: Process details - Process run-times (in Box-Plot presentation)

Concurrently existing processes

Finally, there is a further statistic table, that reveals the concurrently existing instances of a specific process.

We learn that there approximately 12 process instances of our model have been active at the same time.

Report Process Concurrently Existing Processe

Figure 4. Report: Process details - Concurrently existing processes

Title: Name of the process
Obs: The number of started activity instances.
Mean: The mean value of the concurrently existing process instances. In this case there were 11.6338 process instances active on average.
Std.Dv: The standard deviation of the concurrently existing process instances.
Min: The minimum value of the concurrently existing process instances. In this case there was at least one process instance active throughout the simulation.
Max: The maximum value of the process' lead time. In this case there were at most 15 process instances active throughout the simulation.
Period: The period in which the process instances occurred. Note that we set the simulation stop time to 240 hours, which are ten days actually.
Hint: The difference of two hours between that and the period information originates in the fact that we set the process' inter-arrival time (start event) to constantly two hours.

Going deeper

In a lot of cases the histogram and the box-plot can already provide a starting point for improvements.
Nevertheless we need to go deeper into the process' details to fully understand its dynamics.
That’s why the further sections of the report deliver information on the process' elements to comprehend the process' local dynamics for the benefit of the entire process efficiency / structure, etc.
Advance to the report’s Activity Details