What's new
What's new

Pricing Custom Machine Build

savman

Plastic
Joined
May 25, 2015
I attempted my first ever custom machine design/build and I am more than twice the estimated time and I am approximately 2/3rds completed. The customer and I are disagreeing on cost (material estimate is inline so far, but my time is very far off) and frankly, although I am embarassed to say this, this looks like it will be the first contract (I've been self employed for 15 years) that I won't be able to finish.

I want to learn as much as I can from this so I don't repeat the situation. To that end I was hoping for at least some ball park feedback on cost re: custom machinery.

Of course I know what my costs are, but I was hoping to get some examples or guiding rules of thumb that I could use as a sanity check. The scope of work on this project was design/engineering; procurement; fabrication; machining; pneumatic circuits; high and low voltage electrical; and programming the logic to control it all.

Instead of trying to describe every element of the machine....I thought it would be more productive if we talked in terms of R&D costs to make a one off prototype of a machine relative to the final sales price of a moderate to low volume piece of industrial machinery. In other words, imagine someone approached you with an idea for novel machine they want made. By novel I mean you can't simply copy an existing design b/c the machine doesn't exist. On the other hand the machine is more or less combining known processes in a unique way. We aren't talking about bleeding edge research here.

If the final sales price of the machine in a low volume production environment is X dollars, would you expect the first one to cost 2(X), 5(X), etc? Say for example something more involved than a band saw mill but (significantly) less involved than a wood gasification plant. A decent analogy is if someone wanted to modify the functionality of a band saw mill (but still build the band saw mill from scratch), add three or four pieces of automated functionality down stream from the milling of wood; say dynamically cutting the wood to length, drilling some holes and sorting the wood. Additionally, they wanted a semi automated feeding system to load the logs. All to be operated by a single 100 lb person of low skill.

An extreme example is pharmaceutical products...the first one costs 200 million dollars and the second one costs 5 cents.

I ask because I have an idea of what I think the machine would cost in a low volume production environment (12-50/year). It's basically a guess mixed with intuition I arrived at by looking a drastically simpler machines and much more complex machines and arriving at a number somewhere in the middle.

Any feedback is appreciated.
 
I don't know what the question is exactly. No one can tell you how much an imaginary machine will cost, other than to say it must be more than you quoted.

We all get caught in this trap from time to time. If the material cost is within budget and it's only costing you time, suck it up and finish the job.

This is why people specialize. I'm sure I could design and build a pretty kick ass stamping die or lunar lander or underwater ROV. But, I can't do it better/faster/cheaper than a company that specializes in that work.
 
The first item is to never assume that there will be subsequent machines or orders. I have seen too many times a request for quotes on a machine based on future orders....that never came.

Second recommendation for one of that you have no experience with is time and material subject to reviews in progress, such as 25% complete, 50%, 75% and 90%. Also expect 5-10% holdback to guarantee performance.

Tom
 
Where is Motion Guru...

We built a machine a few years back at a place I worked, ran fine with the original sample parts, then the production parts came and it all went to hell. There is a lot of money in that line of work, but a lot can go wrong in a hurry.

As far as being twice the estimated time you have to account for the education you received, it might sting but you learned something. What is the reason you won't be able to complete the project?
 
ewlsey,

Let me put it another way: If someone asked you to build a machine that would sell for 100k dollars in quantities of <25/year. What do you think a ball park prototype cost would be? I know it's a super vague question.

As far as sucking it up; I've worked 11 weeks unpaid to try and correct my mistake. 4 weeks ago he made a decision that costs me an additional 13/hrs per week and several hundred more dollars weekly. I can't do it anymore.

I am a responsible for under estimating my time, but there have been changes to the project, not the least of which is the one I mentioned above that costs me dearly each week. My original estimate was 8 weeks and that's how we arrived at a price based on X/hr. I've worked 11 weeks uncompensated to fix my error and things keep changing/getting added. I can't do any more.
 
Any feedback is appreciated.

It sort of started as a "Science" with a Corps of Engineers Major of uncommon understanding.

He could oversee construction of the Pentagon in record time with "Critical Path Method" planning because there wasn't anything NEW to its construction but the shape of it.

His next task, heading up "The Manhattan Engineer District" was harder. Almost NONE of it had been done before.

The technique it takes to sort that cannot ordinarily be accomplished by a lone individual, no matter how brilliant.

Nor WAS it. The "herding of cats" ensued.

One needs differences of opinion.

Enter "PERT" "Program Evaluation and Review Technique"

Now we have a flexible network of nodes with SOFT time budgets. We define their fuzzy boundaries by taking multiple best-guesses of those folks who are "closest available" experts - and now and then a few complete outsiders. An Economist or Psychologist, for example.

To be VALUABLE the team members must NOT fully agree.

Your costing would probably have been more realistic had you included a "naysayer" and a "synthesist" as well as those with known expertise in whatever the technology included.

Seriously HARD to do that alone.

If you disagree with yourself as often as is needed, you'll risk not leaving the breakfast table to do real work.

You need a "team", and one that is assembled ad hoc, with a different composition of members and expertise, each project.

It need not, and SHOULD not, be a permanent one. There must always be a dynamic lest it stagnate.

The "naysayer" only appears to be a pessimist. His JOB is making the optimists check their premises. WIVES are often very, very good at that. DAMHIKT.

The "synthesist" only APPEARS to be a shit-stirrer. He's actually a facilitator.

Mostly, we are teasing-out the second or fifth-place idea - not for itself, but as feedstock for others on the team to turn into a part of a BETTER idea, and from folks who didn't realize they might have a potentially valuable part of the key, however small.

Not our own ideas. Hardly ever. We find and assess, and "inventory" for re-insertion the ideas of OTHERS.

Synergy ensues. The TEAM prevails, and they can do that even if their concept or product is only second or third best. So long as delivered on-time and on-budget. Or better, whilst a more polished solution is delayed. George S. Patton, Jr. was another who recognised that part:

A good plan, violently executed now, is better than a perfect plan next week.

As said, hard to do as a sole contributor, no matter how brilliant one may be.
 
For the last 10 years or so, I've done nothing but one-off automation and custom machine design. I've been directly responsible for everything from million dollar assembly cells to simple die sets. Even with companies who have been around for 30 years, the exact situation you're describing sometimes still happens. Sometimes worse things happen, and you completely blow your material budget as well as labor, and miss ship dates, etc. But this is a lot more common when you're on the bleeding edge of technology (which is a lot of what I've done.)

The most important thing to quote future work correctly is to have as accurate a history of past performances as possible. Every detail. How long did the mechanical design take? How much of that was modeling, and how much of it was detailing? Same for controls/pneumatics/software. Captured, analyzed, maintained forever. What percentage of a project goes to overhead? Hardware? Tooling? Etc. Capture everything.

Something that is hugely overlooked in this industry is project management, and just how much time that aspect takes. And it is all 'non-productive' work.

Another huge issue is what we call 'scope creep' - when you don't have the targets defined exactly, customers have a tendency to keep moving or redefining them, which translates into redesigns, which translates into huge overruns, typically in engineering labor. A good project manager will capture and control this, and either prevent the scope of the project from changing, or will present the customer with an appropriate bill before proceeding.

If I don't have a project history to draw from for estimating purposes, my personal rule of thumb for estimating time on a project I am 110% comfortable with (a design I could do in my sleep) is to double whatever my initial thought was. (But I probably have history to pull from for an exact number.) If it's something that I think might be a little tougher, or if there is critical information that the customer hasn't handed me and signed off on, I'll go right to 3-5 times my initial thought. If I have serious concerns about either the customer's ability to deliver information, or my own ability to pull off the project, I'll only quote it T&M. I've also told customers that the delivery scheduled is based off of me receiving and signing off on X,Y, and Z documents/information. Screw when they send me a PO - I need information.

History, history, history, history. A simple assembly station with a couple of devices should be a cakewalk. You should know when you quote it if it'll be an extrusion or welded frame. You should know how long it takes to design that from the last one you did. You should know how much it costs from the last one you did. And that same 'technique' applies to every aspect of the machine. Then add 2-4% for inflation for every 6 months that have passed since you last did the job. Or hard quote everything if it's been more than 2 years, or if you just know prices have changed.

Until I became involved in the applications and quoting side of the business, I didn't realize or appreciate just how much time and effort goes into quoting work like this. As an engineer and designer, it's a lot easier for me to do this than a lot of the guys that do the work, because you really need a pretty decent rough design of the machine in order to be able to quote it.

I know you're discouraged right now - I've been where you are, and it's a tough situation to be in. But I would suggest you do everything in your power to finish this one out, especially if it's only costing you your time. Both because that's what I would want someone I was dealing with to do, and because it'll be best for your business moving forward.

FYI - if you want some extra mechanical engineering/machine design horsepower on the next one, feel free to reach out to me through here. I've done some work for members in the past, and would be open to do it again.

Good luck!

Edit: I wrote a lot of this novel while other posts were going up that I hadn't seen. One thing I want to clarify: every custom machine is different, but many have many common features (like the frame I mentioned.) If it's PLC based, there's another 'similar but different' feature that you can establish cost for. Same with guarding, same with a robot, same with a vision system, same with a conveyor, and so on. I know no one likes them, or likes to do them, or likes to talk about them, but a quote that is as insanely detailed as possible is what typically forms the basis for the legal document that will control the entire project. Verbal agreements change (as you learned) and 'that's how we do it' is almost never an acceptable answer, and figuring that out will cost you dearly the first time you have to completely redesign a machine to fit a customer spec, but only get paid for your spec version. Ask me how I know.
 
re: time and materials;


That's the single biggest lesson I've learned. There are simply too many unknowns in this type of project for me to commit to a price. On one component I'm on the fifth iteration. The second was a wholesale redesign, and each subsequent has been tweaks to improve performance.

There is significant functionality he wants that existing commercial machines don't address, hence why he approached me. But I didn't realize all the pitfalls.

Going forward will absolutely be time and materials.
 
re: time and materials;


That's the single biggest lesson I've learned. There are simply too many unknowns in this type of project for me to commit to a price. On one component I'm on the fifth iteration. The second was a wholesale redesign, and each subsequent has been tweaks to improve performance.

There is significant functionality he wants that existing commercial machines don't address, hence why he approached me. But I didn't realize all the pitfalls.

Going forward will absolutely be time and materials.

I respect that decision, but I'll tell you right now that many customers will push you hard to give them a firm quote. If you have a relationship with them, you can sometimes do a budgetary quote, then do the actual job T&M.

If you're going to keep doing custom machine design work, I'm of the opinion that it will cost you business to only do jobs T&M. You'll have to decide if you're OK with that.
 
Johnny,

Thx for the detailed response.

re: you 'multiplier'; A friend of mine who built custom machinery gave me the same advice. I agree wholeheartedly. I don't know what it is, but even though I'm conscious of the fact I tend to underestimate the time it will take me to do things, especially things I don't routinely do, I still manage to underestimate by half or more. If I would have taken your advice of 3 to 5 times on the tough items I would have arrived at a global estimate of 2.5 - 3 times what I did which would have been ok.

As far as finishing...he wasn't satisfied with the progress so he wanted to 'keep a closer eye' on the project and insisted that I move it to his shop. Foolishly, I acquiesced. It has added 650 miles and 13 hours in the car to my week. Additionally, I now no longer have instant access to my heavy machinery and many of my assorted smaller bits. Consequently, it has made it much harder for me to progress on the machine. I've been commuting for 4 weeks but that combined with the scope creep and his unwillingness to revisit the contract in light of the changes is untenable. As I said I've worked 11 weeks unpaid and he hasn't compromised at all. I can't continue.

re: time and materials; you're right and once I've licked my wounds I'll try again, only this time with a better estimate.
 
My take (and we do a lot of "never been done before like this" stuff) is to take what I THINK it will take (on a sound educated deduction based on prior experience) and at a bare minimum, double it....but I usually triple it.

Most of the time, that works out...but sometimes it still bites you in the ass (like this week).
 
Johnny,

Thx for the detailed response.

re: you 'multiplier'; A friend of mine who built custom machinery gave me the same advice. I agree wholeheartedly. I don't know what it is, but even though I'm conscious of the fact I tend to underestimate the time it will take me to do things, especially things I don't routinely do, I still manage to underestimate by half or more.

IF .you have enough history to do stats on that.. you might turn it to advantage.

Real world case,'68-'70. As the"FNG" I analyzed the previous four years of contracts.
"Beltway Bandit" Defense Contractor, never had done the same thing twice. "Pioneers" of exotic technology is what we were.

Discovered the Chief Engineer underestimated labour on every contract by 26%. Not just average. He was within 2% either way, just about 100% of the time. ULM wanted to blame him for 4 years of losses and fire him for that.

I said Hell NO!

Sumbich is as consistent as can be! Leave him in charge of that estimating, do not even TELL him he is off, lest he screw it up trying to correct!

WE will just mark up his numbers AFTER he turns them in, and we are good to go!
And so we were.

Next problem was the obvious one.

It then got harder to win bids.

Mind, we at least made MONEY on those we DID win, so that company got another 35 years or so of life out of that..
 
re: you 'multiplier'; A friend of mine who built custom machinery gave me the same advice. I agree wholeheartedly. I don't know what it is, but even though I'm conscious of the fact I tend to underestimate the time it will take me to do things, especially things I don't routinely do, I still manage to underestimate by half or more.

I still need to touch the stove on this one occasionally to remind myself that it's hot.

As far as finishing...he wasn't satisfied with the progress so he wanted to 'keep a closer eye' on the project and insisted that I move it to his shop. Foolishly, I acquiesced. It has added 650 miles and 13 hours in the car to my week. Additionally, I now no longer have instant access to my heavy machinery and many of my assorted smaller bits. Consequently, it has made it much harder for me to progress on the machine. I've been commuting for 4 weeks but that combined with the scope creep and his unwillingness to revisit the contract in light of the changes is untenable. As I said I've worked 11 weeks unpaid and he hasn't compromised at all. I can't continue.

Lesson learned on this one, and it sucks. I use 'The Rule of 10': if it costs $10 to fix it in engineering, it'll cost $100 to fix it on the shop floor, and $1000 (or more) to fix it on the customer's floor. Shipping something before it's ready is always a disaster (and it still happens all too often, especially when there are penalty clauses involved.) In my world, this is a huge source of frustration, especially for the controls engineers who wind up trying to finish a machine on site at the customer's.
 
For all those "addons" you should have upped your price accordingly. Perhaps you did, perhaps you didn't, you didn't say. My dad always said that that is where a contractor makes his money, changes to original contract.

Tom
 
For all those "addons" you should have upped your price accordingly. Perhaps you did, perhaps you didn't, you didn't say. My dad always said that that is where a contractor makes his money, changes to original contract.

Tom

Unfortunately, I fell into the trap of doing a few small things: 'Let's change this thing here', 'why don't we add that', 'Give me your thoughts on this {unrelated project}'....man that last one was a big one.

When I finally tried to push back he essentially said all those changes were part of the process and definitely part of the original meta-scope. (I'm paraphrasing here)

I'm going to have to be more firm re: changes, free advice, etc. in the future.
 
IF .you have enough history to do stats on that.. you might turn it to advantage.

Real world case,'68-'70. As the"FNG" I analyzed the previous four years of contracts.
"Beltway Bandit" Defense Contractor, never had done the same thing twice. "Pioneers" of exotic technology is what we were.

Discovered the Chief Engineer underestimated labour on every contract by 26%. Not just average. He was within 2% either way, just about 100% of the time. ULM wanted to blame him for 4 years of losses and fire him for that.

I said Hell NO!

Sumbich is as consistent as can be! Leave him in charge of that estimating, do not even TELL him he is off, lest he screw it up trying to correct!

WE will just mark up his numbers AFTER he turns them in, and we are good to go!
And so we were.

Yes, just tip him on his side, cordless screwgun in hand,
affix a 1" shim under 1 shoe.... levels him right up.....:D
 
Even though your work is (at least) three orders of magnitude greater than mine, the principles are going to be the same:

Everything takes longer and costs more.
If you get every job you bid, you're bidding too low.
No matter how careful you are, SOMETHING will go wrong.
 
I was told a long time ago by the shop foreman, build to the print thats all you have to do. Do not design anything for anyone make them design it. If you design it you own it and all the issues. I make new machines everyday but I do it for my own business and it takes me months to get everything perfect and ready for production. I cant imagine someone paying me to do that.
 
I cant imagine someone paying me to do that.

Ultimately, your customers must do.

The difference here is that you are taking ALL the up-front risk based on payback from "unit sales" that may or may not ever cover the costs, let alone make a profit. Betting on yourself, as entrepreneurs must do.

The "hired intermediary" in the cases so far herein discussed hopes to be compensated for T&M as he goes, or to "milestones" along the way, and NOT take the specific risk as to eventual sales. nor have to care if there are any.

Now . I mention that because some DO.

Agree a cost plus fixed fee, plus a percentage of future revenue. Co-development, if you will.

The Boeing "Dreamliner" project?

How much of a burden did that place on suppliers and contractors being willing to cover their own share of preparation for new ways of working if they expected to remain participants?

Wouldn't mind such a deal on tooling for an iPhone, perhaps a new model of Tesla electric vehicle.

But... not a lot of other examples most could dare take the risk on.

By the time they are big enough entities to have the "deep pockets" needed to stay in the game, they have become part of the problem for the providers under them as much as part of the solution for those above them.

As usual, the smaller the shop, the more they have to bend towards breaking point to survive. Or claw their way upward. Or just finally break.

They.. are seen as "expendable", y'see. Corn for food more than corn for seed.

One gets "used up", next entity up the chain goes and finds a new one.
 
Damn savman, You just gotta learn to say no.The reason he wanted to move the job to his shop is because he got tired of driving,spending his gas money driving to yours most likely.
Idea men usually can't do shit that's why they hire people who can.They have no clue nor do they care how much time and aggravation or money you have to invest in their "IDEA".

Cut your losses and run,he will bleed you dry and then move on to some one else.
I'm 72 years old and I still occasionally don't know when to say NO.All out of trying to be a nice guy.
Trying to be nice and not saying no when I needed to has caused me more aggravation in life than being belligerent ever has!
 








 
Back
Top