I presume you mean is creating  3d computer models of your cables worthwhile ?

The proponents of the 'digital twin' design method will assure you it is. Personally I think that the situation is rather more nuanced. Firstly if cables are to be flexed as part of the machine operation,  then while you may use CAD to check things are a good fit, the actual design process is a long way from being automatic, and needs considerable expert input to avoid the very careful design of something that is impossible to build or worse, can be built but fails in service. Cables can snake in ways that the simple rolling and unrolling models do not really capture, and can snag and chafe on things the model suggested they would not touch. However a basically sound design can be verified quite well by CAD, although even then things like best cable lengths can be a bit off.

For static wiring it is less important, as there is presumably some deliberate slack in the harness to allow a re-make, and everything can be tied down so it does not flap too much.

As you allude far from all connector makers have got as far as providing STP files or similar for their products, and a lot of data-sheet drawings lack clarity about the reference planes for cable cutting, and some lack clarity about the exact  shape of the connector as well ;-)  So you end up making the first cable set, and then noting where it is too long or too short and then amending the drawings anyway, and much of the 'design once' benefit is lost.

Then it depends how much design re-use you do. If each product development uses different cables, connectors and fixing methods, then creating accurate models  is scarcely worth the hassle. But if you intend to make the same or similar things for many years to come, then it almost certainly is.

We sometimes do, but but not often, and not always when we later think we should have either...

Mike.

We would essentially be re-using our cables across our machines. We don't have a large portfolio of products and the service lives of them are several years long. The concepts of each are relatively similar and certain functions are copied across all. So in that case, it would seem that it would be beneficial. I also think that if I definitely go down this route, as cables are developed in future they would become more standardised as I build up a library of preferred cable and connector manufacturers.

I fully understand that a "digital twin" (I like that idea) will be helpful but will not capture every snag and fault that could potentially occur.