Using Time Tables with Process Flow

Question

question

William Proctor asked Sep 13, '19 Ben Wilson edited Dec 19, '19

Using Time Tables with Process Flow

I see some previous answers about how time tables can be used to control the flow of tokens in Process Flow from previous years. At this time, what is the most straight forward way of doing this? I have two situations:

#1 - Limit the activities of a process flow to a specific period of the week.

#2 - Limit the activities of a process flow to when a resource is scheduled to be running (resource linked to a time table "shift schedule".

Thanks!

Software Version:

FlexSim 19.1.2

process flow time table

· 2

Jeff Nordgren commented · Sep 13, 2019 at 05:00 PM

@William Proctor,

A little bit of a broad question. Do you have a model or sample model you could send us showing what your concerns are? It would be easier to answer specific questions based on your model needs. Not all needs are the same and there is always more than one way to solve a problem.

But in general, I would use the regular FlexSim objects to do what they do best and Process Flow to do what it does best. For example, it is much easier to just let the FlexSim objects handle process times, down times, break times, time schedules, etc. Process flow is best at giving you a good understanding of the flow of your model. Take a processor for example, I would maybe pull from a list when it is ready to receive a flowitem and maybe send it to another list when it is finished in that processor. But that's hard to give a definitive answer when you don't know the flow of the particular model. Without a specific model to look at, it's hard to give you a good, hard answer to what would be best for that model. Probably didn't help to answer your question much, did it.

0 ·

William Proctor commented · Sep 13, 2019 at 07:50 PM

Jeff - Thanks for your feedback. Sorry that I was a bit vague. What is happening is that I am referencing a 3D task executor in the process flow logic (as a resource) but the process flow logic does not seem to consider that the 3D task executor may not be available via a schedule of availability (time table). Process flow just seems to acquire and use a resource even if it is on break. Past posts have talked about this issue and some logic for changing the token label as a means of controlling the process flow logic (via the time table down logic). I just wanted to understand if this is still the case and, if so, if there is a simpler way to accomplish the bridge between process flow and the time table operations. @Jeff Nordgren

0 ·

Answer 1 · 2019-10-03T19:49:48Z

tannerp answered Oct 3, '19 Ben Wilson edited Dec 19, '19

Hi @William Proctor,

I apologize for the delayed reply. If the Time Table is set up correctly, it will cause the Task Executor to be down for the duration of the down function. During this time, Process Flow can't utilize the resource, even if it is acquired. In most cases, the token is effectively "stopped" because the TE can't complete its tasks. However, the token may still Acquire the TE. If that's causing problems in your model (maybe you have multiple Acquire activities using the same Resource), we can look into that, but the Process Flow shouldn't be able to continue if the Resource is stopped. Would you be able to upload your model or perhaps a smaller example?

I created an example model that shows that even if the token acquires the TE, the TE is stopped and the token cannot continue.

te-acquire.fsm

te-acquire.fsm (32.5 KiB)

te-acquire-example.png (133.8 KiB)

· 2

Craig DIckson commented · Dec 11, 2019 at 08:37 PM

Your sample model does work as advertised, but I think the token is actually stopping at the "Load" activity, not the "Acquire". I changed the model to acquire a different TE, which was attached to the Time Table, but still Loaded and Unloaded the original TE. The tokens didn't stop. (See picture) Unfortunately this is what I really would like to do, use a time table to stop tokens (the reason for using time table is that it also does stop activity elsewhere in the (3d) model).capture.jpg

I'm open to suggestions, but I'd really prefer not to have to replicate the time table functionality in PF

0 ·

capture.jpg (179.0 KiB)

tannerp Craig DIckson commented · Dec 19, 2019 at 04:28 PM

Hi @Craig DIckson,

I apologize that I haven't seen this comment until now. Generally, it's best practice to post a new question and reference the old one so that it's seen by more people.

To try and provide an answer here, though, I will mention that the Time Table can't cause a Process Flow activity to stop, but you could set up a similar trigger without using Process Flow, I think. For example, you could maybe try putting a trigger in the Task Executer "On State Change" that listens for when the TE is down. If it goes down while in the middle of an activity (like what you mentioned above where the token got stuck at the Load activity), it could pass it's active task sequence to a dispatcher that would move it to a free TE.

Just an idea. Maybe you could post this as a new question and see what other ideas are available? If you post your model, I can also take a look and see if I can come up with something.

1 ·

question