CAD Print Conflicts with the Customer Supplied Paper Drawing

[Wes Bucey]

Wes,
Ahhh theres the rub - the customer found it at their incomming inspection. There were not many parts or dollars involved so in this particular case the arguements are academic. However, I disagree with your analysis that the customer is not at fault for supplying two different specifications for the same part. The quoting process does not review two sets of drawings, electronic or paper, and the part is expected to be made from the CAD drawing as that is what programs our machines. That's why my company has a niche - if you give me a CAD file I can, in many instances, get you a part the same day. Even complex parts if they do not require outside vendors. Customers generally want part fast especially if there are only 5 to 10 pieces.

However, I agree that we should have caught it as well. Both companies are at fault in their review process. In the end though the customer will win as it is much easier to remove business from a vendor than to turn it away from a customer!

Again good discussion here on this one!

Just because the process doesn't currently review ALL sets of drawings submitted by customer does not mean that it shouldn't. Do your instructions to customers state "only CAD files will be followed in making product!"? I would guess not.

I ran a contract machining business for ten years. We didn't discriminate in requiring a certain type of info from a customer, but we did have a quoting process you might consider adopting into your arsenal.

Instead of sitting down to quote the part, we turned traditional contract machining marketing methods upside down. We picked up the phone and called the customer with questions like:

1. Do you currently use this part?
2. How many do you use a year?
3. Do you currently make it yourself or buy it?
4. Why are you seeking a quote?
   (If the chooch said anything remotely resembling, "Just checking the market." We replied, "Sorry, we only deal with folks who need the product. No quote. Goodbye!")

If it was an existing part, the one answer we wanted to hear was that they were unhappy with their current supply, either for price or quality. We then went into our concurrent engineering mode and asked them to elaborate. If it was a quality problem, we asked them to send us samples of the nonconforming parts with their inspection sheets so we could "diagnose" the problems. Our whole stance was to position ourselves as "partner" first, before WE ever quoted.

If price was the issue, we'd ask, "How much do you need to buy it for?" Then we said, "That might be doable. If not, we'll tell you straight out that it isn't. If we suggested some design changes which would make it less expensive to manufacture, but still have same fit and function, would you be able to work with that? Are there mating parts which would be affected? Can you send us the assembly drawing and the drawings for the mating parts so we can look at that as part of our analysis?"

Usually, by the time we were finished with the introductory give and take, we had several phone calls, talked to design engineers and manufacturing managers. Soon, the problem of "price" was the furthest thing from the mind of anybody at the prospect's organization. We were busy establishing ourselves as the "go to" folks.

If it was a new part, never made before, we asked about potential real use (we didn't quote a blind spread of quantities and price breaks.) We asked about mating parts, end use, etc. Finally we asked the big question, "What do you think you need to buy this for to be able to have a marketable product yourself?" Then we went into our pitch, "That might be doable . . ."

If we spent as much as an hour qualifying a prospect, it was much more cost efficient than sending quotes out on the fax like oysters spawning in the ocean. We had names, extension numbers and "rapport" established. We would call up before we sent out a formal quote with the oral one, telling the recipient to go to his fax machine and pick it up, stay on the line and then tell us if it came through clearly. Our quotes NEVER got lost in the shuffle. We asked for a target date when he would have the answer and the order. We followed up, eager to help if there was a glitch at their end.

By our third year, our regular customers would call us first, ask if we had time to look at a drawing, then send it over by fax while WE waited for it to arrive. Often, they said, (before we went into our interrogation):

1. We currently use this part.
2. We use ____ thousand a year.
3. We've been buying it from John Doe, tell us if we're getting a good deal or not. We currently pay $_____.
4. We're asking you to look at it - if you can come close on the price, we'd rather deal with a known quantity like you. Sometimes these guys act like they don't need us as a customer.
5. What more do you need to know?

Sometimes, just sometimes, the plan works! Our plan was for our customers to think of us as partners with whom they could share confidential information. It worked because we never betrayed the confidence. Many times, we'd tell them they were already getting a good deal. If the quality was a minor problem, we'd offer to consult with their current supplier to help him overcome the obstacle to good quality. Our pitch to the "competitor" would be that we were partners with them in satisfying the customer. A happy, satisfied customer was easier for everyone to deal with. We sometimes used these "competitors" to outsource some of our overflow once we got their quality systems up to snuff. Our point was, "If you have similar machines, you should be able to do similar work. Let us help you tackle the quality and service issues."

 

[Jim Wynne]

If I were the arbiter trying to sort out whether the customer owed you money for the SNAFU, I might have chided the customer for its laxness in configuration management, but I would assign 100% of the cost to your company.

You would be making that decision without enough information. What we don't know is anything about revision levels of the two sets of specifications. If the purchase order was for some quantity of part 1234, revision A, and both the drawing and the CAD file were identified as revision A, the case could be made that the contract terms were satisfied if the parts produced conformed to either one.

On the other hand, it's possible that the purchase order specified revision A and the CAD file was A and the drawing was B, or vice-versa.

Still another possibility is that the purchase order didn't specify a revision level, in which case the customer should also be liable, because the supplier was obliged to produce part number 1234 and presumably did so.

 

[Wes Bucey]

You would be making that decision without enough information. What we don't know is anything about revision levels of the two sets of specifications. If the purchase order was for some quantity of part 1234, revision A, and both the drawing and the CAD file were identified as revision A, the case could be made that the contract terms were satisfied if the parts produced conformed to either one.

On the other hand, it's possible that the purchase order specified revision A and the CAD file was A and the drawing was B, or vice-versa.

Still another possibility is that the purchase order didn't specify a revision level, in which case the customer should also be liable, because the supplier was obliged to produce part number 1234 and presumably did so.

Jim is correct in one instance, a lot hinges on what accompanied the actual purchase order versus what accompanied the request for quote.

I understood the paper drawing came AFTER the CAD, that is, it accompanied the purchase order.

My company picked up a contract after the previous supplier used the drawing provided for quoting purposes to manufacture a product, ignoring the fact the drawing which actually accompanied the purchase order had some significant changes from the drawing offered for quoting purposes. Neither drawing had a Rev level. The loss suffered by the previous supplier was substantial. To my memory, that supplier had several attorneys refuse to take the case, saying they didn't see a way to win, cautioning the supplier that if he lost in court, he would most certainly be required to also pay the legal fee of the customer in addition to his own attorney fee. There were lots of shortcomings in the deal:

1. No PPAP
2. NO First Article approval before proceeding to production
3. NO agreed Control Plan signed off by customer.

Regardless whether a customer specified PPAP or FAI approval, our shop did the following as a matter of course for EVERY product

We had no quality inspectors (we did have quality trainers and guys who acted as "court of last resort" when a question would arise.) Operators did own first article inspections, based on control plan/inspection plan agreed with customer as part of contract review. Another operator would perform a redundant first article inspection with different inspection instruments. Marked sample with BOTH inspection reports was sent to customer for confirmation before production began.

I would think a prototype or model shop doing one-offs and very short runs would be as obsessive about getting it right the first time, every time, as the high production shop poised to run off millions, billions, and zillions.

We took advantage of the customer, knowing he should have alerted the first supplier to changes, and charged him about 50% premium, since few other suppliers wanted to play roulette with a loose cannon customer. As a sop to the previous supplier, we bought some of his special tooling to allow him to recoup some of his loss. We later recruited that supplier as one of our subs on other jobs.

Let me repeat one of my long held "buceyisms" -
"Every customer is not 'always right,' but the supplier needs to inoculate every deal against a toxic customer BEFORE proceeding."
The proper delivery mechanism for this inoculation is Contract Review. The ideal Contract Review is cross-functional, with at least one member of the team continually asking, "What can go wrong?" until the team is satisfied the possible contingency is covered.

In my years in the contract machining business, we dealt with several customers who had burned bridges with other suppliers. We also refused to do business with a number of others who didn't pass the smell test during our Contract Review. Regretfully, we also had to fire a few customers who proved to be unprofitable, mainly because of the soft costs of doing business with them, not because of any disagreements about the quality of product or service. The single major reason for firing a customer was slow pay.

 

[Jim Wynne]

My company picked up a contract after the previous supplier used the drawing provided for quoting purposes to manufacture a product, ignoring the fact the drawing which actually accompanied the purchase order had some significant changes from the drawing offered for quoting purposes. Neither drawing had a Rev level. The loss suffered by the previous supplier was substantial. To my memory, that supplier had several attorneys refuse to take the case, saying they didn't see a way to win, cautioning the supplier that if he lost in court, he would most certainly be required to also pay the legal fee of the customer in addition to his own attorney fee.

This makes no sense. If there was no evidence of revision control on either drawing, how could the customer claim that one superseded the other? Just saying that one drawing was sent with the PO certainly isn't enough, especially if it was different from the RFQ drawing, and the PO didn't specifically cite the requirements.

 

[Wes Bucey]

This makes no sense. If there was no evidence of revision control on either drawing, how could the customer claim that one superseded the other? Just saying that one drawing was sent with the PO certainly isn't enough, especially if it was different from the RFQ drawing, and the PO didn't specifically cite the requirements.

Just because it makes "no sense" to you does not mean the attorneys looking at the case in possession of the pertinent documents were fools. The fact remains the drawing with the actual purchase order was deemed by them the one which controlled the contract. Thus, they didn't see a way to win.

Granted the customer was a nasty piece of [insert expletive of choice], but a lot of folks I know say they'd rather deal with a thief than a fool because you can put in safeguards against thieves since you can predict possible things they might do. Few of us can claim to predict the actions of a fool.

I was willing to deal with the customer, but I charged enough to offset the extra cautions. And, more significantly, I had the advantage of learning the customer was slipshod in its internal practices - fertile ground for our "brand" of being "go to guys."

 

[Jim Wynne]

Just because it makes "no sense" to you does not mean the attorneys looking at the case in possession of the pertinent documents were fools. The fact remains the drawing with the actual purchase order was deemed by them the one which controlled the contract. Thus, they didn't see a way to win.

Then what you're saying is that we have no way of knowing what the (presumably astute) attorneys were looking at in giving their advice. Once again, there's not enough information from this vantage point to be able to understand the point of the story. We need to do contract review conscientiously and understand that when there's more than one copy of a drawing, they may be different.

One other thing that's been overlooked is the nature of the CAD file. It could be a 3D model, or it could be a simple digital representation of the drawing. In the case of the latter, it's usually a simple matter of opening the file and looking at the revision block to confirm the revision level, but with CAD models, it may not be so easy, and model revisions may be tracked differently from part revisions. This is why some companies will print the solid model revision information on the 2D drawing.

 

[Wes Bucey]

Then what you're saying is that we have no way of knowing what the (presumably astute) attorneys were looking at in giving their advice. Once again, there's not enough information from this vantage point to be able to understand the point of the story. We need to do contract review conscientiously and understand that when there's more than one copy of a drawing, they may be different.

One other thing that's been overlooked is the nature of the CAD file. It could be a 3D model, or it could be a simple digital representation of the drawing. In the case of the latter, it's usually a simple matter of opening the file and looking at the revision block to confirm the revision level, but with CAD models, it may not be so easy, and model revisions may be tracked differently from part revisions. This is why some companies will print the solid model revision information on the 2D drawing.

What you say has merit, but I get a sense the customer referred to by OP did not have a system as sophisticated. If they had, they would have probably made much more effort to disavow the paper version or vice versa. You would have. Almost certainly, I think you might agree that any supplier making the entire production of a product to customer specs without confirming details with the customer is less than "GMP."

There might be plenty of blame to spread around, but I'm pretty sure the legal liability is all in the supplier's lap.

You and I both agree the situation posed by the OP is an example of a bad consequence from an incomplete Contract Review. Even if by some miracle the supplier could win a legal battle to get paid for product unusable by the customer, he certainly loses any future business with the customer and suffers embarrassment and loss of reputation for falling into the situation which might have been avoided with an extra fifteen minutes of reviewing and comparing available documents before producing the product.

Seems like PDCA at the OP's organization left out some steps.

Worst of all, if the OP's organization was part of an automotive supply chain, some martinet in a higher tier learning of the incident might find a way to tar its reputation even more, eliminating intervening tiers from including it in their supply chains because the automotive supply chains reek with FEAR of alienating primes or first tiers.

I was obsessive/compulsive enough that I don't think I might have fallen into such a trap, but if I had, my subsequent risk assessment would most likely have resulted in me sucking up, taking the loss, and filing it under "lessons learned."

 

[SteelMaiden]

Re: CAD Print Conflicts

Marc, Jim and Steel - Good answers. In a small shop like I am in we don't have the resources to review things twice. I will see if putting language in the quote will pass muster with management.

Thanks again!

While I certainly understand what you are saying, I must ask you the same question that I have asked many times in my career with QA and QMS: "If you don't have the time to do it right, where are you going to find the time to do it over?"

 

[Jim Wynne]

Re: CAD Print Conflicts

While I certainly understand what you are saying, I must ask you the same question that I have asked many times in my career with QA and QMS: "If you don't have the time to do it right, where are you going to find the time to do it over?"

That old bromide makes sense if there are repeated errors that could be averted by spending more money and taking more time. It doesn't make sense, however, if you're dealing with outliers--unusual issues that might happen even if you do spend some extra time and money. Whether it's prudent to spend time and resources on any given activity is a function of cost and looking at the big picture.

 

[Wes Bucey]

Re: CAD Print Conflicts

While I certainly understand what you are saying, I must ask you the same question that I have asked many times in my career with QA and QMS: "If you don't have the time to do it right, where are you going to find the time to do it over?"

The carpenter's motto:
"Measure twice, cut once!"
has merit for most businesses.