Role of Machinists in In-Process Inspection in Job Shop

I currently work for a small machine shop that is part of a much larger corporation. Our shop started as a small company and was eventually purchased by the larger corporation, but, even though we are owned by this larger company, we still operate very much like the small machine shop that we started out as.

Our company has very little in regards to a Quality department. We have me - who joined the company in March as the QA Engineer - and one other person who has the role of the final inspector. My background is in automotive in mass production shops, so I'm used to having a large Quality department with several Quality Engineers, Quality Service Technicians, and QC Inspectors. In that environment, QC Inspectors and/or QS Technicians would be checking parts periodically as well as the first piece or each run after a model change, with very little dimensional inspection happening by an operator, but here we have skilled machinists who are primarily doing their in-process inspection on the inspection sheets that are created by our department or Engineering.

Some of the machinists and Manufacturing Engineers are starting to become very adamant that our department, now that I'm on board, should be checking the first piece after each setup and periodically during their run. I'm not opposed to this idea, as I've kicked around this and other similar ideas to try to improve in-process inspection, but our resources are very limited and we've largely been opposed by Management, as they feel that our machinists are highly skilled and paid and should be able to complete their own in-process inspections. With me handling most of the QA stuff on the front end as well as CAPA and customer complaints, and the final inspector wrapped up in verifying everything before it goes out the door, there aren't a lot of resources for checking each job that comes from our manufacturing processes.

For the sake of benchmarking - and to finally summarize this long-winded post - how do you guys handle QC inspections in your small machine shops or job shops? Do you have a quality person to check setups or periodically through your runs, or do you rely on your machinists or operators to do most of your in-process dimensional checks?

 

I think in most cases we want the operators doing the inspections. Your management is right -- for the money they are paying the operators, they can check their own parts. Spot checks and final inspection by a QA inspector are good for verification to keep the operators honest. Good luck.

 

Thanks for the feedback. People here seem to be pretty torn, and the longer I don't do anything to change our current method of inspection the more angry the proponents of Quality confirmation become. Anytime I've kicked around any ideas to try to improve inspection it seems to require additional manpower, which isn't really an option right now.

 

Here I do FPI's when I can, between final inspections and other issues, but I do depend on the operator to a large extent. Sometimes I get to the first piece around the middle of the run, so having the operator check his first piece is critical. Products that are run more often are usually left up to the operator, I will only see these at a final, and that is likely a sampling only. In the morning I "walk the floor" to see what is in process, checking different things as I go, then I know what the day will hold, at least until I read my email...

 

That's pretty similar to where we are now, minus the FPI. We pretty much walk the floor a couple of times a day and double check completed parts and then get an idea of what we need to focus on. Unfortunately a lot of our special product - which is typically very low quantities that may or may not be a repeat job - go straight from Engineering out to the floor, so we can't really make any kind of plan to inspect them. We just have to find out about them as they are being manufactured and many of the jobs slip through the cracks without ever being inspected by anyone other than the operator.

 

The problem with placing confidence in people, is that people leave. New people require a closer eye on them for awhile, or a mentor on the floor. There are people I trust completely, and people I do not trust at all, needless to say I spend my time babysitting the latter.

 

That's a good point regarding auditing the capabilities of the operator. It's also one of the areas where we are a little weak. Our parent company does extensive inspection training with their operators, having them take tests to identify critical features and dimensions on each product line, but we aren't like that at all here. We've been slowly transferring a lot of production from our parent company to here, but we don't have the same holistic approach to the product as it transfers, and we tend to lose a lot in regards to training and inspection. Granted, their facility is a mass production facility, and ours is still a small volume job shop that happens to be making the same product, so there will inevitably be some differences in how we standardize our processes and inspections. I'm not sure if our parent company is aware of the gap, and I've only recently become aware of the differences in the training and evidence of competency that they require, but it's very clear that I need to try to get our facility to the same level of competency moving forward.

 

Hi Steven,

Interesting question. Here are a few points/suggestions.

1) I was a team leader of around 25 machinists. They all do their own first off and in process checks. At the end of the day, they own the quality of their work. And the expectation is to demonstrate that quality with inspection result.

2) The main responsibilities of Quality team, to me, is ...

a) to ensure the appropriate tools/ equipment is qualified (e.g. sufficient resolution, calibrated, acceptable r and r, etc.),
b) engineer quality process (e.g. spc, control plan, pfmea),
c) Perform audits (process, layered) to govern these processes.

Hope this helps...
Matt

 

I've found this thread relevant and helpful to me.

I'm having to juggle dynamics in a small shop that moved into ISO 9001 Certification a bunch of years ago . . . with major disconnects with acceptance of that in the background. This actually isn't an ISO 9001 problem, that's just the Red Herring people sometimes like to pin the tail on.

The issue is a small shop that's always done things as they always did, deciding several years ago to start thinking about Quality Management Systems, not only Checking Stuff, but recording those things beyond pencil and paper and memory.

A result is the open ended belief that any action taken on a Part by a Machinist where a mistake might happen means "The Checker" has to be called too.

In the end, somewhat similar to Steven's situation when I was hired as the Quality Manager.

These are Policy issues unique to where I'm at, and will not be belabored any more here. This should, always, be about objective and practical analysis for the purposes of Quality of course, but also to set the stage for . . . continued improvements and growth in the right direction, as opposed to problematic directions.

Core Observation (in my opinion):

Any manufacturing business must have a quantified METHOD for rationally identifying where two person "first piece" calls need to happen, taking into account all the operational factors that apply, not the least of which is how many Inspectors do you have versus how many Machinists and jobs & job types.

Shops that have this down, have this understood. Some of you out there surely know this, because you are doing it. You INTENTIONALLY evaluate where 2nd Person overhead is required for Inspections, versus what you expect your Machinists to be responsible for and are capable of (e.g. Machinists engaged in precision work, or unskilled labor operating CNC machines).

Some of you, if I understand this correctly, have Machinists doing "1st Piece Inspections" of some sort, even if by Feature. Some of you have Inspectors doing that. Some of you have pieces submitted to a Quality Lab for an FAI (if required), some of us don't do that. Some of you have a mix, where certain part milestones get a 2nd person validation while other feature milestones are the responsibility of the Machinist to Inspect even if a mistake at that point could, technically, scrap the job.

You have an intentionally designed and identified METHOD of evaluation, incorporating true, observable and measurable FACTORS in the shop, that you use to THINK THROUGH where one resource as The Checker is distributed across 15+ Machinists, as example.

My interpretation of MKS77's answer: You either groom people as skilled "Independent Machinists", or you establish dependency-conditioned Operators. At some point there is a fence-line, a limit, to where you throw overhead (Inspectors). Full circle to the idea: You must have a METHOD for accurately identifying where those fence-lines are and what responsibilities each person is expected to own up to and manage. How the mechanism works, and what do people do to maintain traceability for their actions so that continuing improvement of that mechanism can occur, based on operational facts available, not opinion or fiction or memory-in-the-head.

My interpretation of Jennifer Kirley's answer: You must have a METHOD for evaluating and identifying how the responsibility assignments work for 1st Piece, In Process, and other inspections. That METHOD is the filter/lens that tells you if Machinists "do their own Inspections", or where 2nd Person checks occur (Machinist + 1st Piece Inspector Call). It's not an open-ended case of a Checker has to Check as the Industry Standard Default, everything else is just optional. You actually have to get serious about this, sit down and develop the criteria that tells you which approach, or mix of approaches gets codified, based on real metrics in that business.

My apologies for being long winded.

I found this thread useful as I work through dynamics where I am at.