Custom normal distribution for the arrival of customers

Question

question

Daniel asked Dec 1, '23 Daniel commented Dec 5, '23

Custom normal distribution for the arrival of customers

I'm working on a model that simulates the arrival customers. The number of customers and the arrival time vary between 11.00 am and 02.00 pm. They can be approximated by a normal distribution. There are two distinctive peaks at 11.45 am and 01.00 pm.

I already found my way to the parameters of the normal function in flexscript. Is there a way to change the number stream into the model time so that I can use this function to simulate my customer stream? (The solution would be 2 different sources with separate functions for the peaks).

Summarized: I want to create a normal distribution that doesn't use the inter-arrival-time but rather the actual model time.

Software Version:

FlexSim 23.2.0

flexscript source normal distribution

5 |100000

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

Answer 1 · 2023-12-01T12:07:51Z

Jason Lightfoot answered Dec 1, '23 Daniel commented Dec 5, '23

Sure - you can generate your arrival times using the distribution and then schedule them - which you could for example implement in a process flow with a delay to the arrival time.

You could also try setting up the datetime source activity in a process flow with an Offset Variability that is/are your normal distribution/s.

1701432384278.png (25.7 KiB)

· 1

5 |100000

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

Daniel commented · Dec 05, 2023 at 08:04 PM

Hello Jason,

thank you for the quick response. The date time source solved my problem after I figured out how it worked. Thanks again for the help.

I'll upload an example for anyone that needs the same solution.

example.fsm

0 ·

example.fsm (42.6 KiB)

question