Aris Simulation Capabilities

Hello Laurent,

Q - Simulation of process and their attached subprocesses (with transfer of data in-between them)

A - yes simulation can go through processes from high to detailed levels. You just have to create a flow with same events objects

Q - Flowing data objects across tasks and tag some of these objects after they have passed through a task (for example to avoid that these objects flow twice through the same task when this is a possible outcome of the path structure).

A - don't understand. Are you talking of loops ?

Q -  Declare global variables and increment them with value during task execution.

A - this is called simulation experiment. You can declare variables (number of ressources, probabilities... and their values (1 to 999xxx, 0 to 1) and lauch your experiment to analyse the results

Q - Use of event timer to block temporarily the flow of execution.

A - i'm not sure to understand the question, but you have to use waiting time attribute in function to declare the waiting time (1 hour, 1 day, 1 month...) before the activity can be executed.

waiting time is not declarated on events.

Q - Create "template" processes where information will be passed at runtime.

don't understand

regards

Laurent,

I work at Lanner, we provide the simulation component within ARIS Business Simulator. I am interested in what you are looking to do with simulation, if I can be of help please let me know.

We have some general simulation information available on the ARIS Community - posted by myself. we also have more material and case studies on the Lanner website.

thanks

Geoff

Hi Geoff,

Sorry for sneaking in on this thread but wanted to pose a question to you. Are there any "templates" (or questionnaires) that could be found regarding simulation attributes in ARIS? For example, if we want to uncover bottlenecks in a process, which simulation attributes are required? What if the goal is resource utilization, then which simulation attributes are required? If no specific goal is in sight, which attributes would give us a "leg up" if we wanted to simulate some time after the process model has been created? (Looking for a mechanism by which simulation data can be gathered during the modeling workshops.)

Thanks,

Bob

Robert,

I don't have anything that is 'oven ready', however as a general guide you can add data gradually, the more data you provide the more insight the simulation model can provide.

Simply by adding the probability attributes at gateways you can then simulate a number of process instances to predict how often each function is executed.

If you then add timings to each function, you can check the overall processing time and the variability which can naturally occur, especially depending on the route taken through the model.

A next step would be to include the human resources, varying the availability of resources and seeing the effect on utilisation, total throughput and end to end processing time.

Schedules can be used to vary the arrival patterns and resources available to more closely simulate real periods, although often peole focus on peak periods and dont vary the data over time.

I have posted some articles previously which may be of help. I also have some slides that I can share if you contact me directly....