» » next post - The Lean Edge: How do you define “Respect”?
« « previous post - Mike Rother: Why You Need a Vision
Mike Rother

Mike Rother: How to Set Objectives with Lean

By Mike Rother, - Last updated: lundi, novembre 1, 2010 - Save & Share - Leave a comment

Question:  How do you set objectives with lean?

In some ways the answer is easy. To set an objective with lean you simply go to the process level and answer the question, “Where do we want to be next?”

What’s difficult is not so much setting a lean objective, but putting it in a way that allows it to serve as a useful, workable target condition. Doing that requires you to deeply grasp the current condition of the process. Let me give you an example.

At the plating process in a factory that makes bathroom fixtures, an objective of “100% production reporting accuracy” was set. This metric is important because if the plating process undercounts how many pieces it has processed, it triggers the upstream casting process to overproduce.

So let’s test this objective by starting to ask the 5 Questions:

Q1:  What is the target condition?

A:  100% production reporting accuracy in plating.

Q2:  What is the actual condition now?

A:  84% production reporting accuracy.

Q3:  What obstacles are now preventing you from reaching the target condition?

In this example, you’ve now already reached what Bill Costantino and I call the current knowledge threshold. At this point the respondent will recite a list of perceived problems and issues that may contribute to production reporting inaccuracy. If we ask, “Which one do we need to address now?,” we’ll get opinions, or an old Pareto analysis (all Pareto analyses are old), or some kind of prioritization scheme. Make no mistake about it… you are now off in the weeds and no longer working the lean way. You’ll be attacking problems IN THE HOPE that something will have a positive effect on production reporting accuracy. (Jeff Uitenbroek calls it “whack-a-mole” improvement.)

The problem here is that the objective, “100% production reporting accuracy,” is an outcome metric, and not a lean objective, i.e., not a target condition. This number is influenced by so many variables that systematic, scientific improvement — learning — is not possible.

What is missing is an objective — a target condition — that describes in some detail how we want the plating process to function. That is, what are the steps, sequence and times for the plating process? And to be able to define that you first have to analyze the current condition of the plating process in detail. Grasping the current condition is where a lot of the heavy lifting is in setting objectives with lean.

So let’s try the 5 Questions again, but this time with a target condition.

Q1:  What is the target condition?

A:  We’d like to have 100% production reporting accuracy in plating.  To get that, we want the plating process to function like this (respondent shows a document that describes the steps, sequence and times of how the plating process should work).

Q2:  What is the actual condition now?

A:  We’re currently getting 84% production reporting accuracy.  We observed the plating process in action several times and compared what we saw to the desired pattern of our target condition.  As you can see, at the following specific steps there are repeated deviations from the intended pattern.

Q3:  What obstacles are now preventing you from reaching the target condition?  Which one are you addressing now?

A:  Of the the recurring deviations from the intended pattern, we want to first attack the issue of missed scans due to holding the scanner at an incorrect angle.

Q4: What is your next step?  (Start of the next PDCA cycle)

A:  We have designed a fixture for the scanning gun and will run an experiment to understand the effect of our fixture design.

Q5:  When can we go and see what we have learned from taking that step?

A:  By 4:00 O’Clock this afternoon we should have enough observation data to report on the effect of our change in the scanning operation.

I hope you can see the entirely different way of working in the lean way with a target condition. Notice how much deeper the current knowledge threshold lies. (There is always a knowledge threshold.) If you haven’t already, I suggest you read Toyota Kata soon, because we don’t have a whole lot of time left to fool around with scattershot improvement efforts based on easily-stated but by themselves nearly-useless outcome targets.

Regarding the second part of your question, how does this relate to management by objectives?

The original thinking behind management by objectives (MBO), as outlined by Peter Drucker in his 1954 book The Practice of Management, seems not too distant from what I describe above about target conditions. However, in actual business practice and education, MBO became something more like planning and control from above executed to a large degree via setting of quantitative targets and assessing reports of metrics. Tom Johnson calls this “management by results.” Unfortunately, there are plenty of different ways to achieve a quantitative outcome target that have nothing to do with making real process improvement.



P.S. If you’re wondering how Value Stream Mapping fits with setting lean objectives, an answer to that lies here:

VSM and Lean Objectives

Post to Twitter

Share this post...Tweet about this on Twitter
Share on LinkedIn
Buffer this page
Share on Facebook
Email this to someone
Pin on Pinterest
Share on Tumblr
Posted in Uncategorized • Tags: , , , , Top Of Page

Write a comment