I agree a slow production rate (almost a one-off) does not readily lend itself to SPC.
so that I am not guilty of assuming:
- 30 minutes is for one of several operations on a workpiece OR
- is it the time to finish ALL operations on one piece?
(i.e. only a few pieces per shift or two pieces per hour?)
At this point, it is necessary to revisit the concept of sampling at all.
Beginning way back when Military Standards dictated a lot of quality decisions, it was understood that certain conditions required a decision NOT to sample.
Following is an excerpt I had hanging around from an old document (sorry, no longer have the original or title details, but I suspect it came from an aerospace customer):
1.0 Sampling Decision
1.1
Prior to the creation of an acceptance sampling plan, it should be determined whether sampling is appropriate or not. This determination should be made by Quality Department personnel, with input from other organizations as appropriate. Answer the following questions to determine the viability of acceptance sampling as an inspection option:
a. Is there a need to perform acceptance sampling?
b. Does the drawing or related specifications prohibit sampling?
c. Is the rejection rate low?
d. Is 100% inspection disruptive?
e. Are the costs of escape low?
f. Are the costs of sampling low?
g. Has training been conducted?
h. Are the parts produced in high volume?
i. Will sampling provide a positive return on investment?
1.2
Answering “no” to any of the above questions may indicate that sampling may not be appropriate for the given task. However, corrective actions taken can change this situation to once again make sampling a viable option. If sampling is determined to be a
viable option, a sampling plan must be generated.
Look particularly at Item H and discuss with your customer whether sampling is appropriate for the low volume you describe.
Note that "c=0" means the entire lot fails if one dimension on one piece from the sample is nonconforming. Is it really smart to fail an entire lot because a chip may have stuck in a bore or threaded hole? Typically, failing a lot meant (back in the day) subjecting the entire lot to a 100% inspection, which most quality experts now agree cannot assure 100% conformance of parts which pass inspection.
Before we moved away from massive pieces on multi-axis machines with small quantities of finished pieces per shift, we did our operation pretty much as you describe. Our one overriding concern was that ALL dimensions from one machine setup or tool were conforming BEFORE moving to the next setup or tool so that we NEVER continued work on a piece already out of conformance. When the production rate is low, it makes sense to have an operator check all necessary dimensions on a finished workpiece while the next piece is running.
Elsewhere here in the Cove (see quote box below), I described our method of making operators the quality inspectors and using quality folks as instructors and arbiters between customer and supplier (either direction in the supply chain.)
So if we had two setups on one workpiece, we simply took the time to confirm the first work was in conformance (or met the interim measurements required at that stage of completion) before clamping into the next setup and expending more work on a piece.
We gradually segued into a 90% turning operation and ceased working on any part larger than a human fist with much higher production rates. Just as we were transitioning, we became aware of, but did not install non-contact optical and laser measuring instruments which could check dimensions without removing the piece part from a holding fixture.
We did install such devices on some of our turning machines, which allowed us to run about one shift a day completely "lights out" with the non contact (as well as some contact) probes rigged to stop a machine and alert a standby tech to investigate.
And, about my own operation
Operation:
Operators were all on MRB (material review board) in addition to Quality Manager, Finance/Purchasing, Marketing. MRB meetings were held in their [the operators'] conference rooms. If customers or suppliers were invited to MRB, they met there, too.
All training (in-house, machine tool suppliers, outside experts, cutting tool suppliers, heat treaters, platers, etc.) could be conducted on-site. Customers were encouraged to come and meet with operators running their jobs.
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.
In-process inspection, SPC, etc. was performed by operator in real time. If nonconformance was discovered, production would halt - all operators would collaborate on finding and curing cause, only calling in outside help if solution eluded them. Inspection records, charts, etc. went right to computer where they were available in real time to in-house folk and customers.
Operators had autonomy to bring in experts from our suppliers of material, capital equipment, and expendable tooling to stay up to date on industry innovation. Sometimes, we shut the whole shop down and chartered a bus to take us to the International Machine Tool Show to spend the day.
If an operator wanted to see a customer's operation and how his product was used, we made it happen. Similarly for a supplier's operation.
In my own practice, "suspects" were referred to a Material Review Board. The MRB had power and authority to make unilateral decision on suspects detected in-house. If a suspect arose at a customer, MRB worked jointly with customer to determine true status and devise a remedy acceptable to all parties. Often, a customer would be called in to affirm a "use as is" determination, regardless of where the suspicion arose.
From 1980 on, my MRB (Material Review Board) was always cross-functional, so it had the experience, knowledge, and power to make decisions on the spot (high efficiency - no delays in making decisions about N/C on incoming or outgoing material.)
I did essentially the same thing with the groups which made decisions about Contract Review and plans for new capital expenditures.
It just seemed like good sense to me. I'd be willing to bet lots of other executives independently came to the same conclusion as I.
The point we need to keep in mind is:
For a number of reasons - operator variation, measuring instrument variation, personal interpretation or estimate of an instrument reading, etc. - folks inspecting the same part may come to different conclusions regarding conformance to specifications. It is good company practice to have a process in place to routinely resolve the issue when such instances arise. In my contract machining business, such instances arose frequently enough that we codified the resolution under our Material Review Board, regardless of whether the issue arose in-house or not. This is definitely NOT a matter for discipline or punishment, but for simple, methodical resolution, with NO FINGER POINTING!
ADDED IN EDIT: the simplest agenda follows these steps:
- do we have the expertise here on the MRB to decide if material is conforming or not?
- do we need outside help?
- if we need outsiders, do we have authority to get and pay (if necessary) such experts?
- does our customer have to be involved in the decision?
- once we decide whether material is conforming, do we have the power and authority to ship it/rework it/scrap it?
I've had several people ask me why I spent so much money to provide high quality inspection tools at each machine. My partner and I looked at it this way:
If we spent from $250,000 to $500,000 on a CNC turning center which could turn out a complete, intricate piece from raw stock without changing a setup, why wouldn't we spend a few more thousand to have all necessary inspection equipment at hand to assure ourselves AND the customers that we were operating at the high cpk levels which assured and justified our profitability? (To do anything else would have defied my mom's advice when I was five:
"don't be penny wise and pound foolish.")
Further, if an operator could run an expensive CNC, why couldn't he also inspect his work? If you can trust him to do one, you can trust him to do the other (confirmed by spot checks and observations.)