Significant variation in results for breakdown data in different reps?

Question

question

Ankur A3 asked Dec 07 2024 at 5:29 PM Felix Möhlmann answered Jan 03 2025 at 4:20 PM

Significant variation in results for breakdown data in different reps?

Hi Team,

I am working on simulation where I need to feed the following breakdown data, but it is causing significant variation in end results of mean repair time. Currently, I am running this model for 365 days and it is generating minimum 154 data points in rep 2.

I am worried on accuracy as we are not close to mean time of 84 mins in many reps. How to interpret this data?

Is there any way to improve it further?

Please find the model attached for your reference.

Breakdown_Count_Reps.fsm

Thanks!

Software Version:

FlexSim 23.2.0

breakdown data replications

1733591945120.png (72.1 KiB)

breakdown-count-reps.fsm (94.5 KiB)

· 1

______

Cookie preferences

Your privacy is important to us and so is an optimal experience. To help us customize information and build applications, we collect data about your use of this site.

May we collect and use your data?

Learn more about the Third Party Services we use and our Privacy Statement.

Strictly necessary – required for our site to work and to provide services to you

These cookies allow us to record your preferences or login information, respond to your requests or fulfill items in your shopping cart.

YES

Improve your experience – allows us to show you what is relevant to you

These cookies enable us to provide enhanced functionality and personalization. They may be set by us or by third party providers whose services we use to deliver information and experiences tailored to you. If you do not allow these cookies, some or all of these services may not be available for you.

YES

NO

Customize your advertising – permits us to offer targeted advertising to you

These cookies collect data about you based on your activities and interests in order to show you relevant ads and to track effectiveness. By collecting this data, the ads you see will be more tailored to your interests. If you do not allow these cookies, you will experience less targeted advertising.

YES

NO

Are you sure you want a less customized experience?

We can access your data only if you select "yes" for the categories on the previous screen. This lets us tailor our marketing so that it's more relevant for you. You can change your settings at any time by visiting our privacy statement

Your experience. Your choice.

We care about your privacy. The data we collect helps us understand how you use our products, what information you might be interested in, and what we can improve to make your engagement with Autodesk more rewarding.

May we collect and use your data to tailor your experience?

Explore the benefits of a customized experience by managing your privacy settings for this site or visit our Privacy Statement to learn more about your options.

Answer 1 · 2024-12-09T08:31:22Z

Felix Möhlmann answered Dec 09 2024 at 8:31 AM

By generating a couple million values from the distribution it seems that getting an average of more than 110 with 184 samples has a change of about 0.2%. More than 105 and 100 are aorund 0.9% and 3.2% respectively.

It seems like you got very "lucky" with the results of the replications. You could possibly argue to disregard the replication 7 as an unlikely statistical outlier.

· 6

Ankur A3 commented · Dec 09 2024 at 1:34 PM

Hi @Felix Möhlmann,

Thanks for your answer! it makes sense for me.

Going further, should I exclude the outlier manually as it shows the avg value of all data points in results table?

I think it will be little time consuming to do it manually when there are more scenarios and performance metrics. What would you suggest?

Thank you!

0 ·

1733751161999.png (64.5 KiB)

Felix Möhlmann Ankur A3 commented · Dec 09 2024 at 2:31 PM

From version 24.0 onwards, the experimenter has a filter which can filter out replications based on performance measures or parameters. If you have an active license for that version, upgrading might not be a bad idea.

In 23.2 deleting the respective replication manually would be the only option.

Another thing I just realized is that the distribution you are using is somewhat likely to return negative numbers. Since breakdown times would be 0 in those cases, the average of the down times will be higher than the mean of the distribution. With that taken into account the chances for the results are around 1.5 times as high as I initially tested.

0 ·

Ankur A3 Felix Möhlmann commented · Dec 09 2024 at 2:46 PM

@Felix Möhlmann

I checked, deleting the specific scenarios is working for me to update the results table.

Yes, I am observing the same for -ve distribution values but it is not desired in reality.

Should I try something else to avoid -ve or 0 values?

Thank you!

0 ·

Show more comments

question