process flow error message

[a9080109]

@moehlmann_fe

i try to use flexsim 2025 version to run the traffic control! but something wrong happening!!! i dont know haw can i fix 

[a9080109]

@moehlmann_fe

And after I removed the traffic control of the model and ran it, similar messages still appeared, but as far as I know, this has little to do with my process flow. Is it some internal bug?

[a9080109]

@moehlmann_fe

sir ! can u know how to  resolve this topic?

[moehlmann_fe]

The issue seems to be caused by a specific chain of events: The AGV is travelling to a destination CP in it's "idle-loop". It arrives at a traffic control zone and is preempted and made to stop at the same CP is was travelling to previously. If the AGV is now resumed, the original travel task finishes instantly. This also triggers the On Arrival event that a third token somewhere else was waiting for to start giving out a Task Sequence to the AGV. This and the new "idle-loop" task seems to 'collide' in some way and cause the error. I was able to replicate this in a simple example model. I have attached it if you interested in checking it out.

Making sure that the "idle-loop" does not include CPs that represent an entry to a traffic control zone seems to be a possible workaround. Specifically I replaced CP5 with CP3 in the first travel activity of the OHT2 idle loop.

[a9080109]

By the way, if I need to collect the time of space circulation in the dashboard, how can I do it?

[moehlmann_fe]

I don't understand what you mean by "time of space circulation". Could you describe it in other terms?

[a9080109]

i mean idle loop  i want to check this stat in dashboard !!!

[a9080109]

@moehlmann_fe 

Bonus problem 

I want to know that I need to unload to q now. It will take 5 minutes to create a token and continue the process to the next step. But I found a very strange phenomenon that the minimum waiting time is more than 100 seconds. I think this value is incredible. How can I solve it?Is my delay not working? Or is it caused by other problems? Finally, I don't know if it's my illusion. There seems to be no difference in the results between waiting for 300 seconds and not waiting.

[a9080109]

@moehlmann_fe 

I also want to ask that I moved the position of DELAY but it has a great impact on my results. I don't understand why there is such an impact. One is to end TS first and then create the token. The other is to create the token and then end it.

[moehlmann_fe]

At time 2117s an item is teleported out of the "Foup STK Step3" queue to the "APA_Foup1_OUT" queue. The processor there still has a pending Process Finish event for that item, meaning it was removed from the processor prematurely.

[a9080109]

Why did this happen? How can I solve it?

[a9080109]

By the way moehlmann_fe

i want to know idle loop   this state in dashboard !!! i think it different state in my model

Accet Solution

eply

[a9080109]

I found that there seem to be many such cases in my model. How can I solve them? I currently need to monitor the time I execute some processes, for example (the time it takes for me to leave the process of pulling agv and unload it) but I found that this seems to be a problemIs it because there are two tokens at the same time that the confusion occurs?

[moehlmann_fe]

I traced one of the teleporting items back to its cause. At P2_3 there is a queue that serves as an input (P2_3_INOUT1) but is part of the output group. An item being unloaded to that queue thus generates a token that sends an AGV to pick it up again while it's still processing.

To track the idle travel time I would suggest to add a second state profile to the AGVs. You can then set/reset the state of that profile whenever an AGV gets pulled/pushed.

https://forums.autodesk.com/t5/flexsim-forum/add-custom-state-to-a-state-gantt-chart/m-p/13511752

[a9080109]

https://forums.autodesk.com/t5/flexsim-forum/custom-agv-state-be-coverd-when-it-s-allocated/m-p/1360...

 i want to do like this thing!! can u have some idea about this?

[moehlmann_fe]

I provided the solution the linked post refers to, so yes. But I think the method is explained and demonstrated well there. Since you are not using an Object Process Flow for the AGVs, it might be easier to use a combination of the OnStateChange and OnMessage triggers. If the state changes to something that is not the 'forcedState' send a message with a delay of 0s and change the state back in the OnMessage trigger.

[a9080109]

i try to change my process but its didnt work  i dont know what mean is forcedState ??  can u do a example??

[moehlmann_fe]

The post you linked to links to a different post with an example. That example uses a label called 'forcedState' to swap the state of the AGV whenever it changes away from said state.

[a9080109]

I looked at his files but didn't see the code he mentioned in the picture! Can you give me an example of a small model to explain it?

[moehlmann_fe]

Here's an example model that uses triggers. I set the "Forced State" label when a task sequence with priority > 0 is received and reset it when the TE finishes a task sequences and the next tasks priority is 0. In your model you would set the label after pulling an AGV and before giving it tasks. And reset the label before finishing the task sequence.