Freddy Ballé’s Four Points | Richard Kaminski

The following is the translation of an excerpt from from Richard Kaminski’s latest ILF newsletter:

“As Dan Jones wrote in the preface to Lean Management, the leading global Lean Production System in Europe he saw was Valeo’s, developed by Freddy Ballé with coaching Toyota. […] Less known are the methodological battles that led Freddy Ballé to try to correct the failings of the machine he had put in place, battles that have always fascinated Dan and he has encouraged me to describe several items […]:

    1. Quality improvement is the key to innovation. The main industrial problem Freddy has always been improving product quality in production and the implications of these improvements on products under development. Kaizen events help “clean the window” ( and generate immediate savings ) but have meaning only to he extent that eliminating variations reveals quality and process control issues. Quality is much more difficult to achieve than efficiency because it quickly boils down to knowledge — and the need to develop deep skills, which is also one of the true keys to success in Lean.
    2. No improvement without involvement. The first topic for Freddy has always been operator safety. His image is that it takes two feet to move : improvement and involvement go hand in hand. Hop on one foot, and you fall. In the factory, one of his first questions is “How are you organized? ” It’s not about the organization chart but how are the operators are organized. Are the teams stable? With a team leader ? How many operators for one supervisor? How the are the operators’ problems handled?
    3. The pull system is essential not avoid cherry-picking problems. Another common battle is the constant temptation to set priorities for problems. The pull system requires the discipline of solving problems as and when they arise and block the flow. Often the problems faced by operators in the creation of value are very different from the fads that obsess their managers. There are no major or minor problems . There are problems that can be treated quickly and easily and more difficult problems that are a struggle to solve. But the key is to acquire the discipline not to choose problems, but to deal with the reality of flow reveals,  in the order in which it happens (as you produce to Kanbans in the order in which they arrive and not in the order you prefer) .
    4. Hypothesis testing is at the heart of problem solving. Deep learning in problem solving takes place when you test hypotheses . On the one hand , you open minds by asking for several hypotheses beyond the obvious solutions; on the other hand, you learn by testing these hypotheses one by one by practical experimentation rather than rush to a conclusion. Hypothesis test is demanding. However, pursuing ” process improvement ” without individual learning reduces Lean to a mechanical approach that can pick low-hanging fruits but not fundamentally develop the skills of the work force.”
Michel Baudin‘s comments:

What attracted me to this list is (1) that it is short enough to be remembered, and (2) that it is specific enough to be worth discussing.

For example, is it always true that quality improvement is the key to innovation? If your innovation is next year’s new model of a shock absorber, then it stands to reason that the lessons learned from improving quality on the current model can be applied to the design of the new one, so that it can be introduced into production faster and with fewer defects. If “innovation” refers to radically new products like the iPad or Google Glass, however, it is a stretch to link success with quality in production. The skills of the contract manufacturers obviously matter, but they are not the ones who come up with the innovative products, at least until they decide to compete with their customer, as Samsung did with Apple.

Regarding pull systems, I am used to thinking of them as tools for rapid problem detection. Machine breakdowns command immediate attention because they stop production. Defects are detected promptly as parts move swiftly from one operation to the next rather that sit in a WIP warehouse. Using a pull system does force you to address the problems in the order they occur, but only at the level of the immediate countermeasure, not the root cause. At that level, I see no alternative to setting priorities. If you find that the root cause of your forging problems is that your 300 dies don’t have standard dimensions, you have to decide whether and when to address it and it may take a year of sustained effort to fix it. If you simultaneously have a problem with microstoppages caused by parts getting stuck in chutes, you may have to make a choice. Operator involvement broadens the range of problems that can be worked on concurrently, but only to the extent that you can put together teams with the requisite technical knowledge.

Valeo is a well-known success story for Lean, now referred to as “Operational Excellence” and the “Valeo Production System” (VPS).  I couldn’t find any information on line that is not published by Valeo itself. The page on the company web site has quality data from 2013, but the copy is identical to a Reference Document from 2003.  I have had personal communications about VPS from consultants who are Valeo alumni, and I hope to see their comments here.