Understanding metalwork manufacture -- How are Purchase Quotations prepared?

Question

I have been involved with a number of metalwork design projects lately and I'm increasingly finding myself asking for quotations for small prototype runs (often one-off jobs) to verify the validity of a design.

The kind of items I'm talking about are small items like electronic product enclosures, mounting plates for screwing PCBs onto etc. Items that would probably fit into a shoebox and are made of machined or folded plastic, aluminium, mild steel. Not heavy engineering, so no welding etc.

Question: How would a large custom metalwork manufacturing company (1000+ employees let's say) interpret my design files and what steps would they be likely to take in order to estimate costs for a small run of prototype units?

If I give them a 3D STEP file (output from Solidworks for instance), how would they evaluate the complexity of the design and convert that to a list of costs such as tooling charges, material costs and setup-time costs etc?

What I think might happen:

1. They load the STEP file, find the smallest stock material size that fits the design - this gives the materials cost.
2. An engineer analyses the features in the design and does the CAM toolpath generation and other operations at this level. This becomes NRE tooling costs and production line job-change costs.
3. Shipping costs are estimated based on the final object's size, shape and weight.

But this sounds like it would take a long time to organise and would be expensive in terms of man-hours, which they would want to avoid since I haven't given them any money yet. Quite often the quote comes back to me within a couple of days, which suggests that they might simply be "eyeballing" the project, guessing the price and adding a wavy-hand percentage to cover the guesswork.

How does this kind of thing work in this industry? I have asked one of our smaller manufacturers but they seemed reluctant to answer this sort of question as if it was some kind of closely guarded secret. I want to stay on their good side so I stopped asking :)

I'd like to hear details from those with experience in custom machining, sheet metal fabrication (stamping, folding etc) and other related activities.

Answer 1

As Mahendra Gunawardena said, there's not one universally used method of pricing manufacturing projects, but there are some pretty common facts that result from the economics of the situation.

Firstly, if you're talking about small runs at a large company and the design doesn't include anything exotic, the pricing will probably have less to do with the specifics of your design, and more to do with the overhead costs and benefits of dealing with you as a client. In the kinds of jobs shops that I've worked for and with, the algorithm for bidding small projects goes something like this:

Step 1: Should we quote this work at all or decline to bid?

• This evaluation is usually done very quickly by someone senior enough to understand the business relationship with the particular client and the overall workload of the company. They need very few details about the actual design, often one image and a brief description would be enough if the client is familiar with the jargon of the industry and thus able to explain the task. They'll be considering questions like:
  • Does the company have the spare capacity to complete the work on a schedule the client will accept?
  • Is this the kind of project that the company can produce to an adequate quality level?
  • Is this the kind of project that the company is efficient at doing, so they can do the work for a competitive rate.
  • Is the client competent and clear enough about what their needs are that the product is going to work, or are they likely to confuse a design flaw on their part for a manufacturing defect on our part?
  • Is the client likely to chose us to do this work, or are they likely to use another vendor meaning that our labor to bid the project is wasted.
• If any of the above items are 'maybe's or 'no's then the person will take into account if there's another compelling reason to do the work. For instance even it's s stretch on their quality abilities, if it's a loyal client who brings them other work they may decide to take it on and accept the costs of improved tooling, attention to management, procedures, etc. Similarly, even if it's a kind of work where the company has to give up most of its profits to be competitive, they may still decide to bid on the work if they are very low on work and at risk of having to lay off workers. If the company doesn't need the work and it seems like 'bad' work according to the criteria above, they will likely politely decline to save the cost of investing any further management labor.

Step 2: Assuming the person in step 1 decides that the work is worth going after, the next step is to figure out the pricing they should offer.

• This step may be done by the same person who handled step 1, but in most companies, especially larger ones, most of the heavy lifting of this process will be handled by someone else.
• Especially for small projects, this person won't want to spend more time than they have to developing the pricing, so before they do any detailed analysis of tooling or design they will look at the project qualitatively and decide what needs to be penciled out and what can be 'eyeballed.' For smaller projects, lots more can and should be eyeballed. The reason for this is that a small loss (say 10% on a small project represents an amount of money that the company can probably tolerate losing occasionally when there's a mistake. Conversely, a few too many hours of manager or engineer time in the bidding process can quickly become a big enough cost as a percentage of the contract sum that either the company still loses money on it, or their bid is non-competitive. If a company doesn't win the work based on a non-competitive bid, all of the money invested in the bidding process is loss.
• In this evaluation, these kinds of questions will be considered:
  • Have we done a very similar project in the past? If so, often reviewing that project and the actual profit or loss can form a solid basis to bid the new work scaled by quantity and complexity.
  • Do I know what the client is expecting this to cost? If the expectation is comfortably more than actual costs then they can pick a 'safe' number without doing any more work.
  • Are the materials, manufacturing labor, tooling, or management labor are obviously the majority of the costs? If it's clear that one or more of these cost categories was pretty small then they can focus efforts on pricing the ones that matter more and just apply a percentage to cover the others.
• After that, for items where they feel like the the bid warrants more thorough evaluation, they will proceed to estimate the costs:
  • For materials, they would start by checking if the project can be done from a material they ordinarily stock. Even if the yield isn't great, for a small run using common materials the waste will probably be worth the reduced cost of not having to order and receive another stock size. If the material isn't something they keep around, then they will look for prices they have recently paid for similar materials. Failing that, they will need to actually contact a material supplier and get pricing and lead times.
  • For tooling, they would probably start by talking to either an engineer or the folks who actually work on the shop floor. They would start by seeing if any existing tooling can be used or adapted for the project. If not, they would check if there's a tool that can be purchased that accommodates the design. If not, then they would start the process with an engineer or determining what custom tooling (dies, stamps, workholding, etc) would be required and what the design and production effort would take.
  • For fabrication labor they would discuss the project with the folks who run the shop floor and estimate what time it would take to set up tooling, produce, and inspect the whole run.
  • For management labor, they would estimate the amount of time to support the project, This would include things like material procurement, submittal drawings, first article production, functional tests, shipping, etc.

Answer 2

First, I'm not sure if any large custom metalwork manufacturing company of 1000+ employees exist. Most local metalworking companies in the Chicago area are about a dozen employees or less. A laser, press brake, and a few mills and lathes can produce many millions $ of metal parts in an year.

I seen several 20+ years experienced machinist estimate pricing. They look at the print for about 10 seconds and say a price. I'm pretty sure they look at the material, overall dimensions, and machining needed.

Sheet metal cutting is automated. When the dxf or cad file is uploaded, they have software that checks part weight and amount of laser cutting needed. Press brake work is probably # of bends, tooling change, size of part. Haven't been involved in stamping as it is for large quantity production runs.

Answer 3

There is no one size fits all answer. There are many factors that go into during the quotation process. As requester it important to for you know strength and weakness of your supplier.

Following are few of my experiences, though they are electrical engineer based I believe there are similarities.

I use external prototyping houses design and manufacture of electrical/electronic/firmware based systems. I have Non-Disclosure Agreement's (NDA) with all vendors. Therefore I provide as much information as possible to the vendor. There vendors usually references similar project that they have worked in the past. For example if they have worked on MSP430 microcontroller based project in the past with similar components then they have a references with regard to time they spent on the project. They use this a baseline estimate cost to execute the project.

If I give same vendor a project with alternate microcontroller such as a LPC5400 from NXP I usually get a quote that is fairly high and usually takes longer the generate a quotation. If the project is cost sensitive (usually they are) I find a vendor that has experience with NXP product.

Also the if the features requested are generic, then chances are you will get a competitive response fairly quickly. For example the if the PCB board is square and a standard size. For most of the standard features the NRE costs are standard. But if the PCB has features that are custom such as circular or alternate shapes it might take longer and chances a quotation will be higher.

References:

• How to reduce Engineering Bill of Material Cost for electronics products?