Hello, for some time I have been looking for an objective index of the complexity of the assemblies...
Let me explain:
My customers provide me with assembly files (solidworks), which are often quite heavy, on which I have to work, then make drawings (my services consist of various design checks).
Most of the time, I have to give them a quote before I can look at the assemblies, since they haven't yet designed them when they're preparing their budget
--> And, depending on the customer, their way of designing, and the complexity of the machines, the files are more or less heavy: opening time between 20 seconds and 10 minutes or more...
As a result, either I can work efficiently, or I waste a lot of time with files that are too large, which directly impacts the profitability of my services.
=> As it is difficult to envisage invoicing my clients by the time spent (because it does not only depend on the heaviness of the files, but also on my own efficiency, the quality of my equipment, etc.), the solution seems to me to be to offer them, each time, a quote indexed on the degree of complexity of the files they provide me. In this way, they would be able to understand and accept that if their files are very large, it is normal that the cost of the service is impacted accordingly, and everyone controls their risks (they could redesign, or lighten what they provide if necessary)...
There are many clues already, but:
- The opening time is a clue, but it really depends on the power of the PC,
- The number of parts is far too imprecise an indicator because it also depends a lot on the constraints, the complexity of the parts and sub-assemblies...
"> I need an index that is as objective as possible, that my customers can control as well as I do, even if we don't have the same machines...
I'm sure I'm not the only one with this problem...
Do you know if there is an "objective" (or as much as possible) index of the complexity of assemblies and/or parts?
but... This link shows how to design better to save time, but what I'm looking for here is not to do my designs better, but to evaluate quite objectively the complexity of a product that has already been built. I can't redesign assemblies of 4000 parts every time, it would be ultra inefficient...
There is a function (in the evaluate tab) that gives "statistics" in the assemblies: number of levels (of sub-assemblies), number of unique parts, number of total parts, number of bodies.
@ hubert G If I understood Louis correctly has two distinct worries
To make the estimate based on elements of complexity since his mission is related to controls of various designs
Then when he has the contract to carry out the mission.
That's why dead volumes don't necessarily seem to me to be the right solution because in this case we lose a lot of information (including constraints and a lot of links in the creation tree if I'm not mistaken).
@ louis.molline
I suggest working in three stages because there are three very different issues as far as I understand.
What I would describe as upstream work for the estimate since in some cases the parts are not made or the machines are not finished.
Validation of the quote upon receipt of the quasi-final files and update of the quote; if the client accepts a certain % of uncertainty and the revaluation that goes with it (this is done in all contracts and in private or worse public markets)
The period of analysis work itself.
For part 1 I lack a bit of elements to suggest something.
On the other hand, for part 2 it should be interesting to work from macros drawing information from the nomenclature and other available date subsets: in order to obtain a composite indicator. A properly sorted nomenclature in Excel already gives a lot of information. With a little VBA, you can also improve sorting and analysis.
For part 3 , macros should also save a lot of time by automating a lot of tasks such as collision detection, alignment and play checks.