question

Pinakin K avatar image
0 Likes"
Pinakin K asked Jordan Johnson answered

Stats do not stabilize after warm up is triggered

In the new update, the stats in dashboard which are average values and time dependent, like throughput/hr, do not stabalize right away as it did in the previous versions. Is this intentional?

FlexSim 20.1.0
flexsim 20.1.0
· 5
5 |100000

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

tannerp avatar image tannerp commented ·

@Pinakin K,

I just tested this and it still worked for me. Could you upload a model in which the warm up time is not recognized so that we can take a look at it?

0 Likes 0 ·
Pinakin K avatar image Pinakin K commented ·

@tanner.p

Flexsim Recognizes the warmup, even the throughput/hour value resets, but the reset value takes a long time to stabilize. It should stop at a value if there are no variations in the model. But it does not.

0 Likes 0 ·
Pinakin K avatar image Pinakin K commented ·

FlexSim2020_Dashboard.fsm

This file is the FlexSim 2020, you can see how the throughput stabilizes after the warmup is triggered.


FlexSim2020_Update1_Dashboard.fsm

This file is the FlexSim 2020 Update 1, you can see it takes a large amount of time to stabilize the throughput value in the dashboard.

This is the comparison i'm talking about.

0 Likes 0 ·
tannerp avatar image tannerp Pinakin K commented ·

I misunderstood; I thought that the warm up time wasn't recognized in your stats collector, but I see now what you mean. I'm not sure what is causing that to happen, although both models experience some increase followed by a stabilization period. My guess is that it's a behavior like in calculus where the limit for (1/x) as x approaches 0 is infinity. In other words, if the throughput is very high and the time is very low, it becomes a (VERY LARGE / very small) equation (see image below).

I created my own Statistics Collector for the Processor's Throughput per Minute and it does the opposite in that after the warm up time, it goes to 0 rather than infinity. You can see that in the updated model I've attached. This makes me believe it's just how the data is collected and throughput per minute calculated.

throughput per minute homemade.fsm

@Jordan Johnson, do you know what changed in 20.1 that would cause the sharp spike in throughput per hour immediately following the warm up time? It is more exaggerated than in prior versions.

0 Likes 0 ·
Pinakin K avatar image Pinakin K tannerp commented ·

Thanks. Yes I get what you are saying, but in your statistic collector, you did not subtract the warmup time from the Model.time. So it is calculating the throughput from time 0 instead of time 500 (Warmup time). Thats why your model would go from 0 to infinity.

0 Likes 0 ·

1 Answer

Jordan Johnson avatar image
0 Likes"
Jordan Johnson answered
5 |100000

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