My good friend Guillermo Fernandez, the retired General Manager of Colgate-Palmolive Mexico, would always say: "Common sense, the most uncommon of all the senses. Here is a true story to exemplify this very point.
Way back when, I worked for an automotive parts supplier based in Metropolitan Detroit. This supplier was feeling the pressures from the first wave of foreign competition which at that time was the Japanese. Our factories were full of old but still functional equipment. Everyone was scurrying about trying to learn “the secret” that the Japanese were using to beat us at what we thought was our game. Companies were implementing Statistical Quality Control, Quality Circles, and Taguchi Experiments. Executives were mandating the use of these tools willy-nilly, as if just plastering the factory walls with control charts would somehow make a difference.
I even heard a VP level executive say “I want each department to run two Taguchi experiments by the end of the year.” He got his two experiments, because he was serious about it. It was in the objectives of his direct reports. The results? So-so to meaningless. There were control charts everywhere. But without the management mindset and process to properly utilize the tool, the charts really were like wallpaper. After awhile, it was sad to walk through and see charts that had not been updated for weeks or months. Most factories I go through these days are much better managed.
The automotive supplier had a factory in Logansport, Indiana. The factory made springs; all types of coil springs from smaller than you would find in a ball point pen to those hefty ones that you would find on a heavy truck suspension. They had hundreds if not thousands of parts in their catalog. As the flavor of the month was Just-in-Time (JIT) manufacturing, the company, and hence this particular factory, were tasked to implement JIT ASAP. The plant manager of this facility was a go-getter. He implemented what he thought was JIT immediately.
I was in the factory helping solve a quality issue. I had heard of this JIT stuff. You have to remember that it was all new back then. So, I asked about it. I was told they were making product only when the order came in, to the amount of the order. I was thinking, wow this is pretty impressive. Their goal was to have no finished goods inventory except what was on the shipping dock about to be loaded on trucks.
I was working in the factory in an area where they were making smaller springs. I noticed the operators were putting springs in several containers. They sent one container to shipping and put the rest in some nearby shelving. As I thought that was contrary to the JIT mandate, I asked the workers why they were overproducing. The guys were honest and said:
• It takes use three hours to change over the line to run the next scheduled spring
• The run for the JIT order takes only 15 minutes to run and then we are doing another set-up
• It seems stupid so we just run the product for an hour and next time we have an order we just fill it from our stash and have a three hour break.
I then realized these operators understood that you cannot really do JIT when the changeover time was nine times longer than the run time. They also realized that make-to-stock was more appropriate to the Physics of their Supply Chain in which the old machinery was designed for long productions runs and not quick changeovers.
Common sense was indeed uncommon amongst management, but the operators were loaded with it.
Does this sound familiar? I would be interested in hearing your thoughts about this situation or a similar situation. While this is a 25 year old story, the same kinds of things continue to happen today. Do you agree? Do you have a similar story to share?
Last edited by mgavoor; 12/4/2009 at 04:50 PM.
Reason: Take off attachment
You bring up several common issues, specifically with lean.
Quality circles was the first response to "Japanese Management" techiques and like most flavor of the month ideas, the people writing about this and advocating it just didn't understand it. No surprise about it failing.
Quality control fared a bit better, but its failure was that it focused on measuring and tweaking, not on fixing the machines to hold tolerances. Sampling and six sigma fail because they still allow bad parts to be passed on in the process.
This is where JIT is supposed to come in and again it was misundertood by most. One piece flow requires 100% good parts, and machinery and processes have to be configured to produce at this level or shut down...so the problem can be fixed. Another key element of JIT is set-up reduction. The guys on the spring line were correct but the issue here would have been to reduce the set-up to minutes, instead of hours. This takes time, but you adjust your "lot" sizes to your set-up, even in JIT. The goal is to get to a lot size of 1...and that gets back to one of the two key elements of Lean - continuous improvement through small groups. The other is changing the company culture to support continuous improvement - mainly through empowerment.
As you said, most companies lack the common sense to use the power of their people to do these things. They rely on industrial engineers, quality imspectors, six sigma black belts...as if any of these people know half of what is really going on down on the shop floor.
Most managers seek the quick and easy and unfortunately significant change simply doesn't happen overnight. This is why many make the stupid decision to outsource. This creates a host of new supply chain problems and costs, and they normally don't save any real money.
About 20 years ago the place where I worked introduced JIT and Lean. One guy was allowed to set up a small team to reduce change over time and did a great job of it. From 2hours to 20 minutes. At that time I don't think the rest of the organization really understood how this could be used to reduce lot sizes.
10 years later I implemented production cycles where products were made in a repeated and somewhat flexible sequence with minimal (15 minute) change over times. The machine operators loved it because their work was minimised and they knew exactly what each change over coming up was going to need. It wasn't quite JIT but the maximum lead time for any product was 2 weeks and inventory was turned over just about every two weeks. The machine ran at oEE's in the mid 90% range. The math and stragtegy behind all of that really wasn't difficult but it did take time talking to people and getting them on board with working a different way. To my knowledge that was the only site who did that successfully. I think sometimes the not invented here syndrome is quite a powerful one.
I still follow the same logic (with a different company) but have now added a study of margin per hour for each product which helps set a more appropriate cost, especially for smaller, custom orders.
There are some quick successes that can be gained once you have done some groundwork. It took two weeks to introduce cyclic scheduling but two years to lay the groundwork for it. Once you've laid the groundwork and created the right culture successes occur at a faster rate. I think the biggest lack of common sense is that some expect instant and sustainable returns without doing appropriate groundwork.
Corporate culture is usually not what senior executives think it is or would like it to be and it is not driven or created by analytical problem solving tools. It is driven by how people talk to each other, how they work together, the common history they share, the stories they tell. Sustainable improvement comes from a sustainable culture and the common continuous improvement tools in use are only a very small part of what builds that culture.
High levels of poorly considered measurement, data blindness, standard working practices, performance related pay, incentive programs etc all conspire to degrade a 'common sense' culture.
Lean implementation, when is conducted in a blind way, without considering the real situation of the operations can create hilarious situations when we define the lean wheel and lot size, looking only to the expected DOS level.
The best criteria to define those variable should include the reliabity of the set up time. In fact, we should provide training to the operators team to perform the change overs always with the same duration, and it should be measured and followed. This is way that operators can reduce the setup time - reducing the variation, the average time will be moved and the total time will be reduced. This is the time to be used as criteria to define lot size, related to setups. The real loss in the plant floor is related to variations.
The goal to change the process should be delivered to the engineering department, and .this goal would be determined in the Future Value Stream Map.
Yes, now we (some of us) know more about these concepts. But, how many manufacturing businesses have really embraced Lean in real terms?
How many managers are willing to spend the time on "Gemba" to understand the problem first hand?
How many plants have really empowered the operators to identify problems, stop the lines, and solve the problems - for good? (obviously this cannot be implemented overnight)
How many leaders are interested in metrics other than financials?
Your pessimism is not misplaced but neither does it spell failure for everyone who tries to implement Lean. There would be some long term benefits from trying to implement a Lean program although they may be a fraction of what is wholly achievable. Lean is intresting because although one studies and improves processes it also requires and is supported by cultural change. Unfortunately there is nothing in the Lean 'toolbox' that helps executives and managers understand what is required to achieve that change. Much of what is required in that aspect should not be measured (different and lengthy argument to reach that conclusion) but should be communicated, lived and breathed by every member of the organization.
It's odd to think that something as established as Lean is actually incomplete. What do you have to do to 'Sustain' your improvement? What methods are avilable in Lean to help you achieve that? I suspect that they are more inherently part of the Japanese culture and were never embodied in the 'export version' of Lean.
I think that you need to look at "lean" as a process and not a switch that you flip and you are suddenly lean. Once you think that you have accomplished lean - you have failed, because you have then quit looking to improve. So as I see it you will never accomplish being lean, but you can become more lean everyday if you are willing to work at it and if you can get enough of a culture of change in your workforce.
I find that the reluctance to change is most prominent at the top of an organiziation. Of course this is where the most change rhetoric is generated to go with the most resistance. This is often the toughest resistance as it is not a pound the fist on table 'hell no we won't do this' but a much more subtle type of action or even body language that kills change.
I also think that some companies spend so much time and effort trying to measure and justify lean that they end up wasting as much as they save. You need some metrics and I firmly believe that you must measure to manage, but sometimes you can over do it. If you are in tune to your operation and if you really understand your costs you can see the effects of lean and the benefits with your own eyes.
Pretty simple but in my mind it does not have to be hard.
Policy vs. Common Sense
Linear Mode
