Dec 23 2018
One reader recently asked the Gemba Coach “Our teams have good results with kaizen, but nothing seems to stick for long – any advice?” This begs the question of what you call “good results that don’t stick.” Successfully completing a Kaizen project means reducing a new work method to daily practice. A compelling demo in a management presentation is not the end. Claiming good results on this basis is like declaring victory after running a quarter of the race. By definition, if it doesn’t stick, it’s not a good result! If it happens systematically, then your whole Kaizen activity is a failure.
There is more to Kaizen than 5S
Michael Ballé’s response is all about 5S. Perhaps, it is a reflexion of the near-universal failure of 5S programs. A week after they are launched with fanfare as an “easy first step,” 5S programs appear successful. Two years later, however, you see fading markings on the floor, shadow boards with missing tools, frayed cables, broken lights, and puddles of oil on the floor. Meanwhile, overhead, the sad banners from the launch day are still there. And, during this time, the plant has not undertaken any other kind of improvement project.
Change management practices that prevent results from sticking
Given the ideology of the Lean movement, chances are that, by “Kaizen,” the questioner meant Kaizen events. The first mistake is to equate Kaizen with Kaizen events. If you want results to stick, don’t. Improvement activities range in scope from small changes at one workstation to year-long projects on production lines. It’s what is done that must dictate how an activity is organized, not the other way around. Force-fitting the Kaizen event structure on every Kaizen is a root cause of failure.
The second mistake is to measure progress in terms of the number of Kaizen events. If an organization is measured by how many Kaizen events it runs, then its focus is on running new events. Consequently, nobody gets credit for following up on the past. Instead of making the changes work, the operational groups revert to what they were doing before at the first sign of trouble. If you want results to stick, you need to focus your evaluations on outcomes. Kaizen has the dual purpose of improving performance in operations and of enhancing the skills of the workforce. Both are measurable.
There are other ways for management to send mixed or misleading messages. Launching a new ERP system implementation, for example, is often an effective way to derail Kaizen, by automating the existing system and making it more difficult to change. Improvement efforts that involve more creativity than money are suddenly upstaged by enormous investments in software licenses, contractor fees, as well as employee time in training, data transfer, and testing.
The line organization must own the projects
Most organizations set up a Lean department to select, manage and execute Kaizens. The projects are done to the operations groups they affect instead of by them. You ask a supervisor why a cell is L-shaped rather than U-shaped, and the answer is “I don’t know, ask them,” pointing to the Lean group. The supervisor doesn’t own the design, doesn’t understand the principles behind it, and has no reason to sustain it.
Kaizen projects must be led by the managers directly in charge of the target process. Not only is this necessary for results to stick but there is no way a Lean department can have enough resources to take on the transformation of a plant. For this to be possible, the line management structure must be large enough for its members to have the time required. In production, for example, put first-line managers in charge of 17 operators, not 90.
The Lean department’s job should be to support, not execute. It can provide technical assistance and training, make consultants available, intervene to remove roadblocks, facilitate steering committee meetings, organize project presentations, etc., but it should not assume ownership of projects.
Lock in the gains by using them
The improvements achieved in a Kaizen project are use-it-or-lose-it capabilities. If you reduce set up times, cut the size of production runs to take advantage of the improvement in flexibility. If you don’t, the setup times will creep back up to previous levels in six months. Once you find a way to produce the same output with less WIP, you must actually eliminate the WIP. If you improve a cell to function with two fewer operators, you must actually reassign the two best operators to other, more desirable jobs.
Don’t provide a reversal path
If you implement a Kanban loop, make sure you eliminate the need for routing slips and other paperwork the Kanbans replace. Then remove the old instruments from circulation, so that operators rely on the Kanbans. If the Kanbans are extra paperwork, they are soon abandoned.
When you transition from disposable to returnable containers with a supplier, don’t keep an emergency stash of disposable containers in case you run out of returnables. Design the flow of returnable containers with care, including countermeasures for shortages that do not include reverting to disposables,… If you keep disposables around just in case they are needed, they will be and operations will soon return to the old method.
You design and test your improvements. Then you commit to making them “the way we do this work,” and it includes not providing a reversal path.