Some functions cannot return normal values when Python is called

[1830101468]

[ FlexSim 22.2.1 ]

Hello, as you can see in this video, when I call Python, some functions do not return values properly. Would you please help me see what the problem is, and I would appreciate it.

Relevant attachments have been uploaded（PythonFunction）

[1830101468]

I'm trying to write an Astar algorithm myself, but I found that the model was slow when there was a lot of kiva and the map was large. So I want to see if calling Python scripts could evaluate paths faster and thus speed up the model. Could you give me some suggestions? Thanks again!

[kavika_faleumu · Accepted solution]

Hey @shaon S, I think I found the problem. When using "range" in a for loop, the value passed in must be an int. num1 is a double, and so when trying to calculate range, it was working incorrectly. I tried casting num1 to an int in the range function and my FlexSim function works fine now.

(I was messing around with some values for testing, so I'm appending a random value and then the value num1 that many times. It should work fine with our code)

[1830101468]

Thank you for your help. I have another question.When Flexsim needs to call Python frequently, can Python enable multithreading for processing and return values？Thanks again.

[kavika_faleumu]

As long as FlexSim accessors/returners are only being used on the main thread and the multithreaded Python code all waits to return a value on the main thread, then you can do whatever you want with multithreading.

On another note, I don't think that a custom Python AStar algorithm that FlexSim calls repeatedly will be any faster on a large model with many Kivas, especially since the original AStar algorithm is built in C++. I suggest finding other ways to minimize resource allocation in the model. For example, if you have racks with many flowitems, you can turn on virtualization.