![]() |
Copyright © TIBCO Software Inc. All Rights Reserved |
1. In the Process Editor, select the Activity.
2. In the Properties view, click the Simulation tab.
3.
− Constant - choose this distribution for Activities that never vary in duration. For example, a questionnaire might be designed such that it always takes 10 minutes to complete (or rather the variation is so small that it can be considered constant).
− Normal - choose this distribution for Activities which follow a bell curve (you specify the Mean and Standard Deviation). For example, phone calls in a call center might follow a standard distribution (68% of all phone calls are 4-6 minutes in duration).![]()
4. Choose a Time Unit:
−
−
−
−
−
5. Enter the Value for the duration.
1. In the Project Explorer, select the Participant.
2. In the Properties view, click the Simulation tab.
3. Enter the Number Of People/Machines for the Participant. For example, if the Participant is a field engineer and two are required, enter 2 here.
4.
1. In the Process Editor, highlight the Start event.
2. In the Properties view, click the Simulation tab.
3. Specify the Number of Cases you want to run during simulation. Avoid specifying extremely large numbers as this can make simulation time-consuming.
• Constant - choose this distribution for case starts that never vary such as the closing of accounts (which happens for example, at the end of every month, quarter and year).
• Uniform - choose this distribution for cases that have an equal probability of starting over a given range (for which you can specify a maximum or minimum value). For example, if an audit by a government regulator has an equal chance of happening on any day of the year, select a uniform distribution with 0 and 365 as minimum and maximum values:![]()
• Exponential - use this distribution of for examples such as the following:If there are any loops in your Process (where a Sequence Flow returns to a previous Activity in the Process), you must determine a strategy that allows TIBCO Business Studio to break out of the loop during simulation. For more information, see Simulating Loops in Processes.
1. Ensure that Simulation is selected as the Destination Environment:
b. Go to the Properties view and click the Destinations tab.
c. Select Simulation as the Destination Environment.
− Right-click the problem and select Quick Fix (if enabled for the current problem). This gives you the option of having TIBCO Business Studio automatically correct the problem for you.
− Right-click the problem and select Show in Process Editor. This displays the Process in the Process Editor, allowing you to manually correct the problem.
3. Optional - Set the Simulation Date/Time. By default, this is the current system date/time, but you can set the desired simulation time using the drop-down calendar and by entering a time value.
4. Right-click the Process and select Run As > Run Simulation. If prompted, click Yes to switch to the Simulation Perspective.
6.
• The Simulation Control view shows a running total of the number of minutes in Simulation Time. If the simulation is running too slowly or quickly, you can adjust the Simulation Speed in this view.
• The Simulation Results view continuously updates as the simulation runs, displaying simulation data about the Cases, Participants and Activities.
• The Process displays a visual indication of the number of Observed Cases, Current Queue Size, Activity Delay and Participant Utilization % as the simulation progresses:
The indication of case totals becomes red if the number of cases passing through an Activity exceed the total number of cases (for example because of looping).
• Cases - a summary of the simulation, including the number of cases started and finished as well as case times (average, minimum and maximum).
• Participants - information about Participants. For example:
• Activities - information about the cost of Activities, distribution into queues, wait times, and so on.You can also view this information by exporting the simulation data to an HTML report. See Creating and Viewing a Simulation Report.At this stage, the goal is to ensure that the "as is" process has sufficient detail to adequately reflect reality. One way of checking this is to compare the results of the simulation to any real data that is available. For example, a call center might have data about average call times, or the time it takes to resolve a problem.After simulating the "as is" Process, you must decide whether minor changes to the original Process can achieve the desired results, or if the Process must be re-engineered.Identify bottlenecks/areas for improvement in the Process. For example, long queues or extreme idle times for Participants might be cause for concern.If you decide to re-engineer your process, the "to be" Process represents an alternative to the current process. One approach is to create several "to be" Processes, simulate them and choose the best one. You can either create a new Process or modify a copy of the "as is" Process and prepare it for simulation as described in Prepare the Simulation.By changing properties in the Process such as the number of Participants or by changing the Process itself, you can re-run the simulation and see the effect of your changes.Run the simulation in the same way as you simulated the "as is" Process. The goal of this is to get results that you can compare to the results of the "as is" Process.Compare the "as is" Process to the "to be" or optimized Process. One of the ways you can do this is by using the Simulation Report view (see Comparing Simulations).
When comparing Processes you should keep in mind what has changed between the Processes. For example, if you are comparing two Processes that are identical except for the number of Participants assigned to an Activity, it is easy to explain the different simulation results. However when comparing two radically different Processes it can be difficult to identify what accounts for varying simulation results.Identify any areas where the Process could be improved, re-work the Process and re-run the simulation as needed.As a result of simulations in TIBCO Business Studio, you should be able to either optimize or re-engineer a Process to demonstrate substantial cost-savings. However, there is usually a cost associated changing existing Processes. This cost must be balanced against the case for change.For example, suppose that simulations of the "as is" Processes of a support center show that the problem reporting area of the support web site is under utilized. Furthermore, simulations of "to be" Processes show substantial savings could be made by handling 50 percent more support enquiries via the web site.However, there are costs associated with realizing the promised savings. Customers must be encouraged to use the web site to report problems if any savings are to be made, so an investment must be made in raising customer awareness. Support staff would be under utilized in proportion to the increased usage of the web site, so they must be better utilized, perhaps by taking on outside business from other companies.
![]() |
Copyright © TIBCO Software Inc. All Rights Reserved |