We have been working on a new capability in Nitro-CELL to help enable our customers to perform a "Feature-Based Cost Estimation" of their model(s). This uses a few new capabilities and does require a CREO model - specifically not a "rock" -- but a feature-created model. The results of these commands export to Excel for building your own cost models around (e.g. computing feature cost, raw material cost, etc.)
There are two (2) approaches we are implementing thus far:
Feature Counts by TYPE - e.g. 10-hole features = 10 holes
Feature Counts by NAME Matches - e.g. a wall with 2 holes in it is descriptively named for detection/reporting purposes
We are also looking into another capability specific to sheet metal, but we need to confirm some CREO back end functions before we share that concept/capability.
The above combined with other Nitro-CELL functions (like extracting the bounding box/volume and batch processing) have proven to be very handy for cost estimation.
We know that there have been several *very expensive* tools out there that companies have used for cost estimation. Most of them feel like "heavy" and overly complicated. We think (and a couple of our customers seem to agree) that feature based cost estimation is a valuable thing BECAUSE it reinforces the need for good modeling standards and design practices. And this is the key.... Poor modeling (regardless of where it is done) will not produce good results (period) - not just for modeling - but also for costing in this context.
So I am looking for feedback on the above concepts...
We have a few customers that have had this as a goal, but never had a way to easily implement it. Others have not thought about it at all, and still others have spent a LOT of money for software and dedicated resources to make it work for them.
How do you do cost estimation now?
Is there a need for automating this process if so?
What areas/applications do you feel would benefit the most?
How would you do it differently than from above?
Looking forward to your feedback.
P.S. - If you want to directly test what we are working on, please reach out.
Just as an update on this.... Here are a couple of videos that show off how we implemented this capability.
This is very high level - but you can see that it does have some value -- not for all scenarios - but certainly for quite a few. There are more functions available than we are showing here.
Look love to hear your feedback on this.