For as long as I have been advising manufacturing companies I have been telling them that the single most telling indicator of whether they are doing any good or not is their ability to set a plan for the next week and execute it – schedule attainment. Rarely, if ever, does a plant actually execute its schedule exactly. In most plants whatever happens is not even remotely close to the schedule. Full time positions with ‘Expediter’ in the job title exist for a reason.
I am re-reading Tom Johnson’s “Profit Beyond Measure” for the umpteenth time because it is a an enormously valuable resource for understanding lean and lean management; and because it is an ‘onion book’ – one of those books that clicks a new light bulb or two on each time you read it.
Tom cites Deming’s “Out of the Crisis” with a quote that gets right to the heart of my schedule attainment principle:
“If you have a stable system then there is no use to specify a goal. You will get whatever the system will deliver. A goal beyond the capability of the system will not be reached. If you have not a stable system, then there is again no point in setting a goal. There is no way to know what the system will produce: It has no capability.”
The reason the factory cannot attain schedule is that it is operating with what Deming described as an unstable system – processes out of control. Setting goals for productivity, delivery performance, quality or just about anything else are meaningless and a waste of time. Even less meaningful and a greater waste of time are production folks’ efforts to explain their variances from those goals. As Deming pointed out, when the processes are all over the map there is no way of knowing what it will do. It is like trying to explain why the drunk weaved to the left rather than the right.
Deming goes on to conclude, “Focus on outcome is not an effective way to improve a process or activity.”
So what does this most obvious truth have to say about folks who think the most important part of managing manufacturing is agonizing over metrics? I suspect Messrs. Deming and Johnson would say that metrics like those in this photo from a real live gemba board are no cause for congratulations. Rather, they are neither good nor bad but instead, a reflection of a process that is very unstable. Who knows what its output will be when and if it ever stabilizes and generates consistent output?
Paying a lot of attention to the outcomes – especially wildly fluctuating outcomes – instead of focusing on the performance and the stability of the processes is what Deming calls ‘short circuiting’ the need to learn and “management by fear”. Johnson says that such management by fear is inevitably higher costs and lower long-term profitability. That is true because there is a direct relationship between variation in processes and cost – or as the Toyota folks would out it, a direct relationship between variation and waste.
Measure anything you want on the factory floor and, if it is not linear – a nice flat line – there is waste. Ask ‘Why?’ enough times and the odds are overwhelming you will find a root cause somewhere other than the place where the non-linear outcome occurred. There is the big clinker in managing by local metrics. Dig deep enough into the productivity numbers and you are more apt to find parts shortages, lousy product design, unreliable machines, last minute schedule changes … none of which have anything to do with the process being measured or the folks held accountable for the results of the metrics … just about all due more to poor management or poor execution in some silo other than the one being measured.
When Henry Ford said, “Profit is the inevitable conclusion of work well done,” he meant exactly the same thing. Leading and managing are all about getting immersed in the details of the processes, and the first order of business has to be stabilization – getting consistent output and consistent results. In this regard, standard deviation is probably the most important manufacturing metric – how much variation are we getting around the mean?
Only when the processes are stabilized can we have a fighting chance of improving them and thereby improving the results. When we do that we have a baseline for improvement; when we don’t do it we are just fighting fires and shooting craps.