The 5 Universal Laws of Gemba Management

When we say something is “not rocket science” we mean that it is not hard to figure out or to do. If managing people were as easy as rocket science I am sure a rocket scientist would have figured it out by now. Groups of people tend to behave in non-linear and unpredictable ways. We could say “management isn’t non-linear dynamics” instead but in fact management is probably even harder. Involve customers, regulators and the unforseen historical or macrodynamic event and making money today tomorrow and day n+1+1… starts to look quite a bit harder than rocket science.

Fortunately we have learned a lot since the early days of Peter Drucker establishing management as a modern theory and practice. The Japanese were especially avid students of Drucker and created many unique management systems by building on ideas of others such as Drucker and Deming. Not least of these management systems is what we call gemba kanri or gemba management. More broadly it is basic Toyota-style shop floor management. Developing this gemba-focused, team based, high performance learning organization may be the areas that people struggle with the most in the pursuit of lean. Gemba management in layman’s terms is when everyone works together and learns together by solving problems made visible on the front lines. It sounds easy but there are various barriers that we don’t see until we run firmly into them.

Management is an art and a science. The art is in the 20% creativity, experience and panache that we each bring to it and the 80% is in the standard ways in which we operate within the management system, whether we call it best practices or the Acme Business System. These standards enable creativity within a lean enterprise. Even stronger than standards, there are certain universal laws that affect the success or failure of a lean transformation and sustainability of gemba management. Please excuse the inelegant notation in the formulas that follow due to limitations with my text editor.

Here are five laws of gemba management:

Law #1: Go See Inverse Square Law

The speed of response to problems on the gemba is proportional to the inverse square of the distance of the support staff from the gemba.

The variables are

V = speed of response

D = distance between gemba and the respondent

wherein

V = 1 / (D*D)

This is highly intuitive. If an engineer or manager is 500 meters away from the andon light the the response speed will be slower than if they are located 5 meters from the location where the problems are identified. Gemba management, rapid problem solving through root cause analysis and team based performance management require that we rethink the distance we place between one set of brains and another.

Law #2: Wall Law

The frequency of leadership going to the gemba is inversely proportional to the number of walls separating them from the gemba.

If W equals the number of walls, doors and other physical impediments and f represents the daily frequency of visible support from leadership and staff on the gemba:

f = 1 / (1 + W)

This law tells of the folly of walling in our most senior leaders from the realities that affect the organization they lead. Most of us are not even conscious of the number of transactions we must go through in order to go to gemba, such as opening doors, navigating around desks, hallways, coffee machines and other temptations. This law states the obvious; remove barriers to getting to the truth and you will get to the truth more often. In addition this law has the distinction of being spelled with only three letters of the alphabet.

Law #3: The Non-invisibility Law

If need to ask > 0, then visual management = 0.

This is simply and if-then statement to the effect that formalizes the gemba kanri truism “If you have to ask, you don’t have visual management of your operations”. Visual management must leave no doubt. Nothing that is important should be invisible if true management by fact is to be practiced on the gemba.

Law #4: The Y-O Power Law

The effectiveness of root cause problem solving increases by a power of ten for each “why? question asked and decreases by a power of 10 for each “who?” question asked.

If E equals the effectiveness of root cause problem solving and O equals the number of “who” questions asked, while Y equals the number of “why?” questions asked,

E = 1+ (Y*10) / 1 + (O*10)

This does not mean that we never hold people accountable. Sometimes the root cause is simply a person who refuses to follow the standard, and there are standards for how to deal with these people. If there are no standards for how to deal with these people, that is the one of the points of cause. The root cause is still not the person but the fact that policies to enforce standards do not exist. In cases where there are no standards to begin with, people cannot be blamed. Accountability is not a two-way street, it is the three-way intersection of setting, observing and improving standards.

Law #5: Law of Sustainability

The sustainability of gemba management is a function of full engagement in continuous improvement activity over time.

Where S is the sustainability of a continuous improvement culture, a gemba management system or a lean transformation, this law is expressed as:

S = P*i*K*y

where

P = (people engaged in kaizen / total people)*100

i = interval in working days between kaizen ideas implemented by one person

K = number of kaizen ideas implemented per person per interval

y = years of continuity of total engagement in kaizen

If we take benchmark statistics from Toyota for example, we can calculate their chances of sustaining their continuous improvement culture. We will assume that 99% of 250,000 worldwide Toyota employees are engaged in kaizen. We will further assume that at a minimum they are implementing one idea per person per month or every twenty working days, which is in fact a suggestion system benchmark Toyota hits or exceeds at most sites. Toyota has had their suggestion system since 1950. Therefore:

P = (247,500 people engaged / 250,000 total people)*100 = 99%

i = 1 idea / 20 days = .05

K = 1 idea per person per interval

y = 59 years of continuity in full engagement kaizen

This yields the number 292%. In other words Toyota has close to 300% chance of sustaining their continuous improvement culture in the future, not considering adverse conditions that erode these chances. Factors such as leadership changes, a global economic crisis, the dilution of talent through management turnover, overly rapid global expansion or the complete change in business focus are all factors that today threaten the sustainability of the kaizen culture at Toyota. However, thanks to their 59 years of continuity, near full engagement, and high frequency of kaizen idea generation they are well equipped to weather this storm.

In theory, if you sustain full engagement for 20 years with 1 kaizen idea per person per month you will have a sustainable kaizen culture. This intuitively makes sense as 20 years is roughly a generational shift. For most of us is that a near term target to reach 100% sustainability requires the following conditions to be met:

P = 100% engagement

i = weekly or daily kaizens

K = 1 per person per interval

We have very little influence over the y variable on a day to day basis other than to not give up. We can’t speed up time so we can only speed up the interval of kaizen to more frequent than one month, increase the number of small ideas generated, and expand the breadth of engagement to reach everyone.

By quadrupling the combined number and interval frequency you may be able to achieve sustainability in as little as 5 years of full engagement (e.g. S = 100%*0.2*5). However based on my experience length of time and breadth of involvement are the best indicators of success and sustainability. Most organizations focus a small number of people laser beams on improvement activity while treating the rest as dim light bulbs. Just as the sun comes out each day and shines its light on everyone we should all shine our lights on improvement every day.

12 Comments

  1. John Santomer

    November 2, 2009 - 10:41 pm

    I say, yet again – you’ve hit it right on the nose!
    I have a question though on Law No. 3 – The Non-invisibility Law. If we were to follow the Binary system setting 1 = yes and 0 = no;
    Say, If need to ask(1 or yes) > 0, then visual management = 0(no)? There must be something I’m missing here. Shouldn’t it be : If need to ask (1or yes) > 0, then visual management = 1(yes)? Its like saying ; if there is a need to ask then no to visual management? Or probably it is better stated as : If (need to ask = 0), then (visual management = 0). In this case the If – then statement would be analyzed as If (need to ask = no), then (visual management = no). Just inquiring…Thanks.

  2. Jamie Flinchbaugh

    November 3, 2009 - 3:02 am

    Jon,
    Great post and as an engineering geek, great use of equations to explain management phenomenon.
    One addition and one correction, at least from my experience.
    The addition: on the Wall Law, the walls create a physical barrier that make it less likely to go to them gemba. Having worked in a wide variety of environments, I would say this also applies to the amount of PPE (personal productive equipment for those that don’t know) that you must wear. Getting suited and wrapped takes time, and is one more reason people don’t go and see. If it takes you 15 minutes to suit up and 15 to suit down, you can’t really take advantage of a 30 minute gap in your schedule. Of course, it isn’t a lean answer to just eliminate the PPE.
    My correction is that your sustainable law argues for a high frequency of ideas. In my experiences, frequency or number of ideas is easy to measure, but the quality of those ideas has a greater impact on sustainability. That of course is harder to measure, but people are internalizing whether or not the ideas are helping. It is how much it helps that makes people want to do more of it.
    Thanks for sharing. That was a fun post.
    Jamie Flinchbaugh
    http://www.jamieflinchbaugh.com

  3. Rob

    November 3, 2009 - 6:10 am

    I recognise the truth behind these equations although I’ve never seen them expressed in this way before. I think another law could be related to the guideline, “make only what you can sell”. This obviously leads you onto the well-known three equations, which look the same at first glance but actually aren’t:
    1. Price – Cost = Profit
    2. Profit = Price – Cost
    3. Price = Cost + Profit
    – Price: The customer decides how much he is willing to pay for the product – not the manufacturer.
    – Cost: Manufacturers have control over the costs by removing muda, decreasing inventory, etc.
    Profit: Is what you have left.
    Ask:
    – what does the customer what?
    – how much he is the customer willing to pay for it?
    – how can we remove muda to lower costs?
    – how much should price be?
    In other words, costs exist to be reduced, not to be calculated.

  4. Tim McMahon

    November 3, 2009 - 7:38 am

    Great post Jon. I like how you focus on the process and not the results. Simple, fun way to explain management at the source. While we don’t report sustainability in this exact formula we do measure that data. In our case it is the number and frequency of kaizen which reduce our sustainability.

  5. Wal Kazarin

    November 3, 2009 - 8:15 am

    Jon,
    thank you for this best message. Many russian managers ask me and my colleagues (consultants) about how long time need to support continuous improvement effort to change business culture. Here is a answer. 8)
    And thank you for all 5 laws.

  6. Jon Miller

    November 3, 2009 - 8:59 am

    Thanks all for reading and for your comments.
    John – the point of the Non-invisibility Law equation is that if you MUST ask to find out the condition of your shop floor, you do NOT have visual management. Visual management should make abnormalities jump out. If no red lights, etc. then all is OK. If a manager really has to ask, “How are we doing at this moment?” and can’t gain that information visually from an hour by hour board for example, they fail at visual management. If need to ask = 0, then visual management is not zero, that is the corollary.
    Jamie – you are absolutely right. There are lots of non-wall barriers like PPE. Great point. In the law of sustainability I have made several assumptions that go to your point. One is that the ideas are small. They would have to be to be so frequent. Two is that they are local so that the person with the idea can implement them. Three is that they are actual implemented kaizens, not ideas. Therefore the quality of these ideas is assumed to be good. We are in fact measuring kaizens, not ideas. Kaizens are good ideas, a subset of all ideas which include bad or non-feasible ideas. Thanks for pointing that out.
    José – I would be honored to have the post translated into Spanish on your site.

  7. Daniel Markovitz

    November 3, 2009 - 3:05 pm

    BRAVO! I’m going to poach a piece of this post for my next one. Visual management is important whether you’re dealing with seatbelt assemblies or marketing plans.

  8. RAMMOAHAN

    November 5, 2009 - 2:44 pm

    Dear MR.Jon Miller
    very good presentation. EVERY INDUSTRY ,ALL THE ENGINEERS MUST READ IT & FOLLOW TO IMPROVE THE EFFICIENCY
    I AM FROM INDIA, STUDIED IN USA& UK,PRESENTLY IN USA, I WILL EXPLAIN TO ALL MY CEO FRIENDS IN INDIA, YOUR GOOD EXPLANATION
    Regards
    M.K.RAMMOHAN

  9. Jon Miller

    November 6, 2009 - 6:32 pm

    Hello Ram, thanks for your kind comment. I hope your friends can use these laws to their benefit. Jon

  10. Tonis Kilp

    December 18, 2009 - 11:01 am

    Brilliant! This all applies to service and government areas also, thanks for the ideas, the walls can also be thought of vertically like in an organizational structure or silos between departments.

  11. Alexander Karbainov

    November 24, 2010 - 10:04 pm

    I think there is a mistake in the Law #5.
    The Law must be: S = P / i * K * y,
    because i – interval in working days. Yours note of the Law can be right if “i” is frequency.