See other templatesSee other templates

Industrial economists talk about how the USA is tremendously productive. Our efficiency and labor cost per products produced continues to improve, and we are truly a world-class economy from that standpoint. Unfortunately, you may be feeling too tired to wave the stars and stripes in the next parade, as the increases in productivity sometimes come at the expense of the folks working in those factories. We are doing a lot more with a lot less people, the folks who are employed work more hours, with a lot of time pressure to process a lot of work. So how might this affect someone running a CMM, and what strategies can you employ to help get through the daily crunch? Some of the issues covered do not apply evenly from company to company, nor do they happen all the time, but may be cyclic.

Dealing with management by crisis

The increased level of workload may often be found in administrative staff as well as production and skilled trades. When people spend the clear majority of their time just dealing with the daily crunch of tasks at hand, they get further behind, make more mistakes, don’t take time to plan their work, until - along comes a crisis. Pick your crisis, name your crisis, we see them all the time. We have an ISO certified CMM inspection lab, and a significant percentage of our work comes from companies in crisis. For the company in trouble, each crisis costs extra man hours to solve, putting people further behind, until in some cases management by crisis practically becomes the standard operating mode, as staff spend most of their time putting out one fire after another.  It can be painful for us too, as the companies that are in crisis are always “hot” for measurement data so they can solve a process gone awry.

When the crisis hits the inspection room, the CMM programmer will often become a focal point as management looks for answers. Sometimes the crisis even involves the actual inspection results from the CMM or other measuring devices. A classic case is when the CMM report says a part is good/bad but a check fixture on the floor says the opposite. We will return to this scenario. Prepare yourself for these situations by applying strategic planning as part of your daily routine.  

The basic concepts of good strategy

Not being a rocket scientist myself, it’s helpful to pay attention to the advice of those smarter than myself, or even better – an actual rocket scientist! A couple of years ago, I read a book by Richard Rumelt called “Good Strategy/Bad Strategy”. Mr. Rumelt was indeed a rocket scientist, working as an engineer at NASA jet propulsion laboratories. For several decades now, he has been a consultant to very large corporations and the US government helping them to apply strategic concepts. They paid him a lot of $$$ for his advice, I bought his book for $10.99, and will cover a of his few key concepts that you get for free. BTW the book is a pretty good read, as most all the ideas he shares are illustrated with real world examples, both good and bad. It’s kind of gratifying to read about corporate CEOs who are paid 50+ times a year more than me, who made very bad decisions. Better yet, some of the concepts in the book can be applied to your own career and life to help make better decisions.

The kernel of good strategy

I find it easier to think of the kernel as the framework or a map for good strategy.

1) Diagnosis – it all starts here, and for folks in the inspection side of things this is the main activity we provide to the companies we work for. Specifically we measure a part and provide feedback about the “what”. The part complies to dimensional specifications or it doesn’t, and if not, exactly how does this part fail to meet the specifications. Most of the discussion here will relate to a few topics on the diagnosis portion of good strategy.

2) Guiding policy – this is the overall approach to resolving the problems discovered in the diagnosis phase.

3) Action items – these are the specific set of actions taken to implement the guiding policy

A clear example is the doctor-patient. You visit the doctor, they check your blood pressure.

Diagnosis: he/she discovers that you have high blood pressure

Guiding policy: reduce the blood pressure, get it back to normal

Action items: lose weight, exercise more, watch the salt and take a medication

All three of the phases for developing good strategy are important, but many bad strategies (probably most of them), are caused by a poor initial diagnosis. The metrologist is often the primary first responder in a crisis, doctor measurement, the CMM taking the readings that help with the diagnosis. This can be frustrating, as the diagnosis role in a crisis is about remediation – helping to fix a problem likely caused by somebody else’s bad decision or lack of planning. But remember that a crisis may devolve into a “blame game”, so be thorough in your approach, and review your inspection methods thoroughly to make sure that your process is not part of the problem.

Develop a check list of best practices (hopefully I’m preaching to the choir here), for your systematic approach to part inspection. A great concept for process improvement is the Pareto principle: 20% of the problems cause 80% of the cost. Going back to the doctor -high blood pressure example, experience and research have already established the guiding principle and actions that will solve 80% of the problem. Your practical experience or that of other experienced staff may get to the root of the problem quicker, if anyone takes to time to ask the right questions.

Another paradigm is to make sure you don’t overlook the simple and/or obvious causes of measurement error, and that is the best place to begin looking for fixes. You may want to break your list down into categories. For instance, a random error will likely have different cause and quicker fix than a systematic error. Don’t get embarrassed by overlooking an obvious problem like a dirty part or loose styli. It happens – one time a statistical whiz kid type called me in to look at the gage I just sold them, because it wasn’t repeatable. I drove ninety minutes one way, went on the floor, started taking some measurements repeatedly on the same part. The non-repeatability was not random – the part kept growing larger ever time! So I tightened the tip on the indicator. The down-side was, the whiz kid was so embarrassed that I never got another call. Fixed a problem – lost a customer, bummer.

Just like the doctor, the diagnosis is not always so easy or apparent. Going back to the example of a non-correlation issue between the CMM measurements and the fixture gage on the shop floor, that can be a very frustrating experience. A checklist here would start with the obvious questions. Is the part on the floor dirty when they put it in the gage? If not, then is the gage fixture constraining the part differently than the CMM holding fixture?

So, you checked all that, and the problem is still not resolved. Before you pay to re-calibrate the CMM or rework the fixture, consider that it could be the way a CMM measures the part. it’s pretty much a given that the CMM is measuring the part slightly different than the gage. There are many variables that could cause misleading results, it would take a book to cover all that ground, not an article. But here is a common example -  going past the simple stuff like the CMM takes measurement points in slightly different places that the fixture -  the CMM may be measuring features like holes by averaging the points, using a least- squares best fit approach. If the hole is out of round, especially if the form error is not symmetrical - the hole is oval etc., then the CMM could evaluate the position of the hole differently than a fixture pin that fits to the maximum material condition of the hole.

Another error that we often see, is the case where a surface profile is over-constrained to one or more datums thus potentially reporting a good part as bad.

At Inspection Technologies, lately we have been using a software from QVI called Smart Profile. The software provides better analysis of 3D profiles with a lot of options for fitting and graphical reports that are more intuitive. The Smart Profile software also has an excellent GD&T analysis functionality. This alone makes it worth the price. You can run the part prints requirements through Smart Profile, and it will flag a GD&T mistake in the design. This can save companies a lot of time, and preempt some arguments. The feature control frame proofreads the tolerances and issues suggestions and warnings if it suspects incorrect entries for GD&T. This is very good software, and can take measurement data from any CMM or vision system. We have even been using it to take data from CT scanners as it is able to process large polygon files.

Given that you may be up to your eyeballs in workload, the thought of learning to become proficient in yet another data analysis software might not be attractive. This type of analysis can be outsourced. You may have a local source you can tap, if not Inspection Technologies will perform this type of analysis on a contract basis for your company. Even legacy data sets from any brand of CMM can be analyzed if a company needs to go back and look at some historical data.

A recent study showed that most people use less than 15% of the features available in Microsoft Word, or Excel. It may be the case that your CMM software already has some functionality built-in that would resolve a problem, but you haven’t tapped into yet due to lack of training and time. This is another case where bringing in some outside talent may solve a problem quickly, always a good thing in crisis mode. The host of this website (Mark Boucher) and contributor Ray Xing are knowledgeable resources who can help with a variety of CMM software programming issues.

By providing accurate inspection of part geometry, the metrologist is a key player in the diagnosis phase of good strategy development for solving those “the parts don’t fit” type of crisis. Many of these problems could be prevented before they occur if the measurement data from CMMs and gages were better managed, instead of sitting in a file somewhere that is ignored by overworked staff. Taking time up front to develop and implement good strategy saves time and money compared to running in crisis mode. Fix the roof or the plumbing before you have major water damage.

The author:

Dan Smith

Inspection Technologies Inc.

This email address is being protected from spambots. You need JavaScript enabled to view it.







JSN Megazine template designed by