Quality Vs. Productivity - Blame ISO 9001

[Marc]

> Quality dilemma of the week.
>
> My operations folks have been beating up our quality system (ISO) due to
> the fact that they state their productivity numbers have gone down since
> incorporating ISO-9000. I challenge that.

From: ISO Standards Discussion
Date: Thu, 30 Mar 2000 21:01:41 -0600
Subject: Re: Q: Quality Vs Productivity (Blame ISO-9000) /Hitchcock/Summerfield

From: George Summerfield

Hi Al:

In my experience, Manufacturing is always looking for a scapegoat for any decrease in productivity. Having an office for Quality just gave them one more target to choose from; and its one that the bean counters like to pick on too because of the cost of quality. So, there you are in the middle - "between a rock and a hard place". Have faith, you're neither the first (nor will you be the last) QM to be there.

Others may have a difference of opinion to my following suggestion(s); but here goes - remember, that this is IMHO and totally up to you:

1. Call-backs should not be included in your output. The idea of having a quality system is to standardize procedures and reduce the number of call-backs, warranty work, and legal liability problems. Try to find a way to localize these numbers - a history of how many there were before the implementation of the quality system - and publish an internal report showing the reduction in incidents. This is not production, this is what is commonly referred to as "hidden factory" work. You have to reduce this, without allowing it to be counted against production. Count it separately. It might be taking production workers away from production to perform "re-production", but this should reduce the total hours available.

2. Total hours available has to be clearly defined. Some companies may count two shifts per day as a total of 16 hours per day; but realistically they have to count down-time for maintenance and inspections of manufacturing equipment/machines, production time diverted for warranty or re-production work, training and union or management meeting that personnel have to attend during the work day, shut-downs for family days, holidays, etc. I'm not saying that all of this is not taken into consideration, but you may wish to re-check their figures as to how this was calculated - how exact were they, or did they take an approximation. If theirs' is an approximation, then give a better one of your own, with more detail to substantiate it.

3. An 8 hr shift has a 1/2 hr lunch break and two 15 min. breaks in it. There goes one hour right there. If there is start-up + warm-up time to machinery, than take away another 15-30 minutes for set-up. If there is wind-down and clean-up time after the shift then take away another 15-30 minutes lost production time. In an 8 hour shift, if you can accurately count 6.5 good hours of production per worker, then your company should count itself among the fortunate. If you take into effect human circadian rhythms (we're getting scientific here - and into time:motion studies as well), then between the highs and lows of the human at work during the average work-day; you'd be lucky to get 5 hours of 100% production per worker, and this does not take into effect the historically lower production days of Monday and Friday. However; without slipping back into "Taylorism", you should be able to come up with a truer number than 8 for production hours in a shift. Just adequately justify it when you present the bean counters with a more accurate equation to calculate by. Also add in that there is little they can do to improve on those numbers, especially if your company is unionized.

Hope this helps...

George

------------snippo-----------

From: ISO Standards Discussion
Date: Thu, 30 Mar 2000 21:04:57 -0600
Subject: Re: Q: Quality Vs Productivity (Blame ISO-9000) /Hitchcock/Pfrang

From: Doug Pfrang

A few thoughts to consider:

1. There is a difference between causation and coincidence. Has anyone demonstrated a causal link between the "lower productivity" and the ISO implementation, or have they merely noticed that the two events happen to roughly coincide in time? If the latter, remind them that a great many things can contribute to a drop in productivity.

2. Pointing the finger back at management might be perfectly justified, and might make you feel better if you were the one in charge of the ISO implementation, but it is unlikely to garner support for your views.

3. "Productivity" is a notoriously nebulous measurement, often used by people who oppose some initiative (like ISO) and want to manufacture some "data" to support their cause -- so, they find a measurement that gives data favorable to their beliefs, and they do not care how contrived the measurement is.

4. ISO is a long-term process. In the short-term, writing procedures and revamping processes slows people down and can reduce many measures of "productivity."

5. ISO is also a learning process. Learning takes time to sink in.

6. According to Deming, measurements of "productivity" are antithetical to the entire concept of quality system management.

7. Success with ISO requires management support. By continuing to employ their old measures of "productivity," your managers are effectively denying that support -- so they might (ironically) be causing the unsatisfactory performance results themselves.

-- Doug

[Marc]

From: ISO Standards Discussion
Date: Tue, 4 Apr 2000 11:40:25 -0500
Subject: Re: Q: Quality Vs Productivity (Blame ISO-9000) /Hitchcock/Scott/Nichols

From: "Thomas David Nichols"

> Re: Q: Quality Vs Productivity (Blame ISO-9000),/Hitchcock/Scott
> From: Pfscott2
>
> Al,
>
> I'm not a productivity professional, but it seems to me that the idea
> is to have your people spending as much of their 8 hours doing work
> that is, by contract, billable to the customer, as possible. The goal
> should be100%, but, as you have correctly pointed out, that's a goal
> that will likely never be attained - kinda like batting 1000.
> Efficiencies of 80% -90%, like batting averages of 350-400, are,
> depending on the industry, generally considered excellent. I'm not
> sure I understand how to interpret result of the formula you presented.
> What exactly does that number mean!

Are you sure the goal should be 100%? Goldratt's book, "The Goal," convinced me that trimming your resources to keep everyone busy all the time is highly inefficient. If everybody has to work at peak performance all the time just to handle the planned flow of work, what happens when the flow has fluctuations? With no reserve to call on, an unplanned delay can only grow at each stage of production until it becomes a crisis somewhere down the line.

I also have a question about the productivity formula in the original post. Suppose it used to take five hours to produce one unit of your product, but now it takes only four hours. You can now produce 25% more units at the same cost, but only if sales increase by 25%. Now suppose that sales (and profits) have stayed the same. Does the formula indicate that productivity is the same as before? If sales are off by 10%, does that count as a decrease in productivity?

I am assuming you have taken George Summerfield's advice and removed re-work from the formula. If you formerly spent half your time fixing production mistakes and now have 100% reliability, it is silly to say that productivity has fallen.

Thomas David Nichols

[Marc]

From: ISO Standards Discussion
Date: Tue, 25 Apr 2000 16:05:53 -0500
Subject: Re: Q: Quality Versus Productivity (Blame ISO-9000) /Hitchcock/Tribaldos/Paten

From: Mike Paten

Productivity is always some measure of output over input. However, corporate level measures can be deceiving - that's why they should be as broad as possible - such as output units / total labor hours. In fact, I wonder if they have any value at all (except to board members looking to hang someone or receive a bonus).

However, the most important feature of any measure should be that it serves as a sound baseline for improvement. If "productivity" or "quality" goes up because of "gamesmanship" or internal numbers crunching you've got a problem.

I was once responsible for a Productivity "Gain Sharing" program for a major Navy industrial facility - where essentially, "profits" are returned to customers in the form of lower billing rates the next contract year. The overall productivity indicator was direct labor hours / total operating costs, with direct labor hours strictly limited to hours billable to a Navy "customer". The increase in that ratio (from quarter to quarter) was considered a productivity "gain". That factor was applied to quarterly "profits" (revenues-costs) and split 50/50 between the industrial facility and the Navy program office for ultimate "return" to the customer. At any rate, the gain could only be met if certain "quality standards" were also maintained -i.e. rejects, returns, complaints, etc. all had to be at or below predetermined acceptable levels.

I once asked Mr. W. Edwards Deming what he thought about productivity and quality indicators, profit/gain sharing and the like. His response was data always needs to be analyzed over time (we tend to think short term) and that it should never be used to penalize anyone - rather only for process or system improvement. Profit or gain sharing he said was generally a bad idea - if people are worth more compensation, give it to them. Profit/gain sharing (if it must be done) should be shared equally with everyone in the company because executives, sales personnel, and production personnel (people that generally get credit for productivity gains) can not do their job without support from everyone in the company.

This rather long dissertation has only one conclusion - data/statistics are only useful if they are understood and accepted by those who NEED them to manage the business. Bean counters (and many consultants) usually come up with numbers that cannot be used to identify action - and therefore, such numbers have little or no importance to the workforce or any legitimate improvement activities.

The best "numbers" I've seen come from mapping out a "real" process, identifying points in the process that can be measured, identify measurement points that represent "bad" outcomes (rework, waste, complaints, etc.) and track them over time to identify/prioritize improvement opportunities.

Global measures of productivity rarely mean anything to anyone. When is the last time that customer satisfaction, first time quality, on-time delivery indicators, etc. were used to improve a process, a system or a company?

Best of luck.

Mike Paten

[Marc]

From: ISO Standards Discussion
Date: Thu, 27 Apr 2000 11:47:34 -0500
Subject: Re: Quality Versus Productivity (Blame ISO-9000) /Hitchcock/Tribaldos/Hankwitz

From: "Hankwitz, John"

I'm not sure which of Deming's points would be most appropriate
here.

10. "Eliminate slogans, exhortations, and targets for the workforce", or more likely

11. "Eliminate numerical quotas for the workforce" and "Eliminate numerical goals for people in management".

You provide perfect examples to illustrate Demings's points. When you create something to represent what you're trying to accomplish, the representation becomes the driving force rather than the accomplishment itself. That's why "management by objective" crashed and burned long, long ago. (or at least should have) It usually ends up being counter productive. (Will management ever learn?)

Focus on eliminating your system constraints or root causes preventing optimal performance. Your results will take care of themselves.

John Hankwitz