Skip to main content

Long computation time

C
Hello,



I am performing an Vehicle Routing Problem with Time Windows in AIMMS.

The problem consists of 164 locations. I already expected that using all these locations would take a very long time. Therefore, I decided to divide this area into sub-areas.



I tested the model with 8 locations and for those locations I got a result. However, I am now doing the same for another area which consists of 18 locations and the computation time is very long. It stays at a gap of 38% and once in a while it decreases with 0,01%.



Now is my question is there something I can do about this as it seems that the model does work for smaller areas? If you need my model for this, please let me know to which address I can send it to as the data is confidential.



Kind regards,

Claire

    Jess Es
    Forum|alt.badge.img+6
    • Jess Es
    • Grand Master
    • May 16, 2019
    Hi Claire, thanks for your question.

    This article has some efficiency tips that may help: https://how-to.aimms.com/Articles/125/125-execution-efficiency.html

    @mohansx perhaps you can offer some further advice specific to this situation?
    Jessica E. | Community manager
    mohansx
    Forum|alt.badge.img+5
    • mohansx
    • Former AIMMSian
    • May 17, 2019
    Hello @Claire ,

    Did you use proper domain conditions on your variables and constraints so that the model is better equipped for scaling up in size ?

    It would be helpful to take a look at your model before we can offer any proper advice. Can you send the project folder zipped along with the data case to support@aimms.com ?
    Mohan

    Reply


    Terms & Conditions