Adventures in APQP

Jim Wynne

Leader
Admin
I'm a quality manager in a very busy (thank goodness) metal stamping and fabrication job shop. In the past few weeks I've been dealing with a situation with a customer--a major North American vehicle OEM--that perfectly illustrates the problems with American manufacturing in general.

It begins with me doing "due diligence" in contract review--going over a drawing for a new job to make sure everything is understood and we have everything we need to be successful in producing the product.


In reviewing the drawing, I see two things that need to be cleared up:
  1. There is reference to a customer document that I don't have.
  2. There is a nonsensical GD&T callout that can't be resolved.
We have been told that the customer SQE is the primary point of contact for these things, so I call him and ask for help. For those who might not be familiar with these sorts of things, customer SQEs are never allowed to answer a ringing telephone. That's what voicemail is for. I leave a message briefly explaining the issues and asking him to call me back ASAP. After three followup calls over the course of three days, he finally calls me back and tells me that for #1 I need to talk to the buyer, and for #2 I need to talk to the design engineer, and he gives me contact information.

So much for the concept of Single Point of Contact.

I call the buyer about #1, and he says he'll send me the document. Excellent.


Now I have to deal with the design engineer, a task I'm not looking forward to. I spent nine years working in a vehicle OEM's design and development center, and I came to understand that beyond the education requirements, there are two basic prerequisites for becoming a successful and prosperous design engineer:
  1. You must be fundamentally incompetent in the process of transforming an abstract concept into cohesive, understandable specifications. "Design intent" is a secret that must never be divulged.
  2. You must never ever, ever admit that you are wrong about anything, despite abundant, irrefutable evidence to the contrary.
I call the DE and explain the problem with the GD&T and ask for help. The acutal problem, it turned out, was not the dimensioning scheme but my own wayward and misguided perception of it. It made perfect sense to him. Understanding early on that I might as well have been discussing the issue with my cat, I asked him if he would send me the "correct" interpretation by email so I would have it for reference when the time came to do the layout. He graciously agreed to do so, and no doubt felt that he was doing a benificent service for poor benighted QA guy.

This customer has an online APQP tracking application, and customers are supposed to input information as the project develops. For example, we are supposed to acknowledge having all of the engineering information we need and that specifications are understood. Of course, the entries are timebound, and the system automatically sends dunning robo-email messages when something falls behind.

After a week or so of not having the document called out on the print, and not having the design engineer's explanation of his GD&T delusions, I received a phone call from the SQE urging me to update the APQP tracker. I politely deferred, citing the lack of important information. He said he would get back to me. He didn't.

Today, about two weeks after this all began, I'm in the same position I was at the beginning, only now I have our marketing project manager (the guy in charge of this customer's account) asking me why I'm behind in updating the APQP tracking. "Just close those things out," he tells me, "and we'll worry about getting the information later."

This ugly and sordid chain of events is, I'm afraid, absolutely typical. We're urged by our customers, on the one hand, to invoke a conscientious APQP process. At the same time we're stymied and stonewalled by the people who should be providing the information needed for both sides to be successful. "You must be sure that you have everything you need," they say, "but good luck getting it from us."
 

Chennaiite

Never-say-die
Trusted Information Resource
Sadly, the situation is not uncommon. I am sure you are also dealing in mails apart from phone calls. God forbids, if something goes wrong, its the evidence that will speak for you, not the SQE or Marketing guy.
Ofcourse, the bigger issue is why such OEMs don't respect the System, while demanding a lot from their supply chain.
 

Pancho

wikineer
Super Moderator
The phone is great for quick resolution of issues when the parties are cooperative, but written communication can't be beat with or without cooperation. Are there comments in APQP tracker that can be used to prompt the elusive replies?

We use RFIs and a magical contractual provision that we always get accepted and then has a beautiful way of smoothing the flow of info: "Notwithstanding anything to the contrary elsewhere in this contract, Vendor shall be entitled to an extension of time at least equal to the delays by, or attributable to, Purchaser in meeting its obligations under the Contract, including delays in timely payments, responses to RFIs, and approval of submittals."
 

Marc

Fully vaccinated are you?
Leader
Something to think about...

Contract Review, especially changes after the Original contract is signed (such as a print revision), is always a problem that I have seen pop up. The initial contract for a job is sometimes poor, but when changes occur many companies simply do not do an adequate *new* contract review.

I was thinking of this because of recent thread where there was a discussion of when APQP ends: https://elsmar.com/Forums/showthread.php?t=13149&page=3
 

somashekar

Leader
Admin
Aaaah, you are talking about OEM and subcontractor. What about within the organization.
I have seen specifications released from which even the lord from above could not buy parts or assemble them to make the product, let alone testing them. When one happens to head an unchecked design and development and the whole design and development plan is nothing but 'Yes, we can do that soon and thats easy' you only have one smart alec whose followers also develop into a smart alec. It hurts the organization and false belief prevails over true judgement.
 

Jim Wynne

Leader
Admin
Pancho: I wish it were that simple. In reality when dealing with automotive customers, not only are they oblivious to their own requirements, but they are often blissfully ignorant of the details of the contract. This is not to say that the contract isn't there and can't be relied upon if things get nasty, but the way these things generally run is that you waste valuable development time in just trying to get the basic necessary information, and before you know it's time to start producing parts, and things get done by the seat of the pants. In most cases there are no ill effects outside of the wasted time, but there are times when things go wrong that could have been prevented if suppliers were just given the information in time to avert the problems.
 

Jim Wynne

Leader
Admin
Marc: I agree. I've just had a situation where several months ago, with a large non-automotive customer, we had to ask for a deviation for a specification that proved impossible to hold. This customer is in the habit of not revising prints in these situations--they just issue a deviation "for the life of the tool." Just last week, after a tolerance had been changed from ±.010" to ±.020, we received a revised print where that particular characteristic was newly identified as critical, meaning that we were now expected to hold Cpk at 1.67 for the original specification. The person who had initiated the change was apparently completely unaware of the deviation.
 

Jeff Putman

Starting to get Involved
Jim,

I know it isn't really funny, but your story made me laugh because I have walked in you shoes too many times. Good luck.

Jeff
 

michellemmm

Quest For Quality
Marc;bt963 said:
Something to think about...

Contract Review, especially changes after the Original contract is signed (such as a print revision), is always a problem that I have seen pop up. The initial contract for a job is sometimes poor, but when changes occur many companies simply do not do an adequate *new* contract review.

I was thinking of this because of recent thread where there was a discussion of when APQP ends: https://elsmar.com/Forums/showthread.php?t=13149&page=3

I agree. In a normal world, a well-designed contract review process works. If the company's product development follows a waterfall model as recommended by ISO, the process becomes fairly straight forward. A few changes here and there are not significant.
Recently, I have been involved with software development. Customer prescribed the "Iterative" development process. In my opinion, this model is developed for those who make requirements up as they go along to "Refine" after never ending verification and validation!! This translates to taking one step forward and ten steps backwards. It gets worse when the customer is in pursuit of "perfection".... In the software world that I am getting to know, APQP is not used. Most of the customer engineers are non-practicing six sigma black belt certified.
I am not sure if a contract review process can be designed to refine the process for software development. After three years of investment in time and money on a simple product, the company found out that the 400 requirements are ambiguous. Some are not testable/ verifiable. Others conflict one another... Back to the drawing board!
 
D

DarrellH

Welcome to the world of AGILE, the software philosiphy for "design on the hoof",
I have been fighting with this for 12 months against a background where "management" are "flabbergasted at the number of bugs we release".
I share your frustration.

My advice is invoke the power of the customer "what are the customer deliverables", measure against them, all else is superfluous.
 
Top Bottom