question

Jyhjeng D2 avatar image
0 Likes"
Jyhjeng D2 asked Jyhjeng D2 commented

Meaning of Apply MTBF to a set of states

Hi:

I set a simple queue in which I use MTBF to trigger the breakdown of a processor. The flowitem comes to the queue1 every 2 seconds. And it takes 1 second to complete the flowitem. The program is attached. sample_1-3A.fsm The first breakdown occurs at 5 seconds; the breakdown takes 3 seconds, then it will continue to work for another 10 seconds, and the setup is as shown.mtbf-01a.jpg


And the result of content vs. time for queue1 is attached.mtbf-01.jpg

The flexsim file is enclosed. sample_1-3A.fsm

I can read the content vs. time without any problem. It says that processor 1 breaks down at 5 s and lasts for 3 s. So from 5-8 s, processor 1 is down. Since the flowitem comes every 2 seconds, at time 6 s, the flowitem coming to the system has to wait at queue1. And at time 8 s, another flowitem comes, by then the number of content of queue should go up to 2. However, processor 1 started to work, and it will take in one flowitem from the queue, so the number of flowitem in the queue remains as one since it takes only one second to complete the processing in processor 1. At 9 s, the processor takes another flowitem from queue 1, rendering there is no flowitem in queue 1. Since the operating time for processor 1 is 10 s after reassuming from breakdown, at 18 s, the processor breaks down again for 3 seconds. Thus the number in the queue goes up from 1 to 2 in this period. And the process repeats.


However, I have difficulty understanding "Apply MTBF to a set of states." It is an option in the Breakdowns tab; I applied it to the processing as shown here.mtbf-02b.jpg


And the result becomes weird to me, as below.

mtbf-02.jpg

The flexsim program is attached. sample_1-3B.fsm

Could anyone explain what happened to the content vs. time in the queue? Why does the number in queue 1 never go up to 2?


FlexSim 22.0.1
apply mtbf to a set of states
mtbf-01a.jpg (71.5 KiB)
mtbf-01.jpg (26.5 KiB)
mtbf-02.jpg (23.2 KiB)
mtbf-02b.jpg (87.3 KiB)
sample-1-3a.fsm (32.1 KiB)
sample-1-3a.fsm (32.1 KiB)
sample-1-3b.fsm (32.3 KiB)
5 |100000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 12 attachments (including images) can be used with a maximum of 23.8 MiB each and 47.7 MiB total.

1 Answer

Felix Möhlmann avatar image
0 Likes"
Felix Möhlmann answered Jyhjeng D2 commented

The setting makes it so that only time spend in the "applied" states counts for the time between failures. In your example only the processing time counts, so the processor always breaks down at the end of the fifth (and later every tenth) process. Because it only breaks down at the end of process, it will become active again right as the second item enters the queue.

· 9
5 |100000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 12 attachments (including images) can be used with a maximum of 23.8 MiB each and 47.7 MiB total.

Jyhjeng D2 avatar image Jyhjeng D2 commented ·

Does it mean it will not consider three seconds breakdown? In this way, how long will be the breakdown time? Could you be more specific on the timetable of processor 1? According to your statement, the timetable for processor 1 would be: time 0-5 s working, then 5-8 s breakdown, then 8-15 s working, then 15-18 s breakdown, then 18-25 s working, then 25-28 breakdown. And so on. It should be different. Because the content vs. time chart does not align with the timetable above, what is the correct timetable?

0 Likes 0 ·
Joerg Vogel avatar image Joerg Vogel Jyhjeng D2 commented ·

If a breakdown considers only processing times, then a breakdown occurs only in processing times, but it can lasts longer than an off-time by a timetable. Time table stop time and breakdown could occur in parallel.

They are totally independent from each other results. A time table is time driven. A breakdown is interval driven. They don't get summarized. To prevent any side effects, please consider different Stop ID's.

Please refer to description of depracated stopobject command or actual stop method on objects.

0 Likes 0 ·
Joerg Vogel avatar image Joerg Vogel Jyhjeng D2 commented ·
After 11s of simulation time there are 5s of processing over, then there is for 3 seconds simulation time a breakdown. In this time interval the queue content rises to 1 at 12s and to 2 at 14s, while there is still an item waiting to get released to sink. At 14s simulation time the breakdown is over and an item, ready for releasing, is leaving. This leaving item is responsible to drop the content of queue to 1 again, because an item is received in processor.
0 Likes 0 ·
Jyhjeng D2 avatar image Jyhjeng D2 Joerg Vogel commented ·

Sorry I could not figure out your answer. If possible, maybe you could draw a timeline (or timetable) of processor 1 to describe what happened during the simulation period of 40 seconds. Then it will be more understandable to me.

0 Likes 0 ·
Jyhjeng D2 avatar image Jyhjeng D2 Jyhjeng D2 commented ·

Sorry I could not figure out your answer. If possible, maybe you could draw a timeline (or timetable) of processor 1 to describe what happened during the simulation period of 40 seconds. Then it will be more understandable to me.

0 Likes 0 ·
Jyhjeng D2 avatar image Jyhjeng D2 commented ·

Sorry I could not figure out your answer. If possible, maybe you could draw a timeline (or timetable) of processor 1 to describe what happened during the simulation period of 40 seconds. Then it will be more understandable to me.

0 Likes 0 ·
Jyhjeng D2 avatar image Jyhjeng D2 commented ·

Hi Felix:

Now I understand the thinking process of your answer. The setting makes it so that only time spent in the "applied" states counts for the time between failures. It means only processing time is counted in this mode in tracking when will the breakdown occurs. Since the first breakdown occurs at 5 seconds, it happens after five processing is completed. Thus the breakdown occurs at a time of 11 seconds. Then breakdown lasts for 3 seconds. The following breakdown occurs after another 10 seconds, meaning another ten processing needs to transpire and is tracked from the 5th processing. So the following breakdown starts at time 31 seconds, and the breakdown last for three seconds. In summary, the timetable is as follows.

time 0-11 processor is working

time 11-14 breakdown

time 14-31 processor is working

time 31-34 breakdown.

And so on, in this way, it matches the content vs. time chart for queue 1.

Am I right in interpreting your answer?

0 Likes 0 ·
Felix Möhlmann avatar image Felix Möhlmann Jyhjeng D2 commented ·

Yes, this is correct.

0 Likes 0 ·
Jyhjeng D2 avatar image Jyhjeng D2 Felix Möhlmann commented ·
Thank you for replying.
0 Likes 0 ·