Home > The Lean Post> Follow-up Q&A for Jim Lancaster's Work of Management Webinar

Follow-up Q&A for Jim Lancaster's Work of Management Webinar

by Jim Lancaster
September 21, 2017

Follow-up Q&A for Jim Lancaster's Work of Management Webinar

by Jim Lancaster
September 21, 2017 | Comments (8)

“The Real Work of Management” webinar drew very engaged attendees who submitted hundreds of questions. That was many more than we could address during the hour-long live session in which Lantech CEO Jim Lancaster described the daily management system the company implemented when lean continuous improvement efforts stopped delivering great financial results.

So, we selected the questions that represented the major topics you wanted to know more about for a follow-up Q&A with Jim. Here are some of his answers:

Q: Could you tell us a little more about the right-hand side of the action slide [slide 29 below]? One side has “maintain the current condition.” the other side has “improve the current condition.”

Jim Lancaster: The left-hand side is managing the standardized work of an area and the process. The right-hand side is where we manage either improvement projects in the area or projects in the area that are part of a broader, company-wide strategic project. 

An improvement project A3 describes clearly what our objectives are, why we’re doing the project, and what the general plan -- the A3 storyline -- is. Then that breaks down into a series of key task monitors (KTMs) that break the work down into weekly chunks. By breaking work down, we know whether we’re on track or not, just like you know with the takt-time clock whether you’re on track or not.

Q: What’s the significance of the 30-day and the 90-day columns?

Jim: Level one KTM (L1 in above slide) is 12 months. The chunks of work that you’re trying to do are in deliverables by month. That’s where you’d plan out the whole project for 12 months. When you go to a level two (L2), it’s 12 weeks. You take the first three months of the level one KTM and you break it down into weekly chunks. You don’t do the ones farther out, just the next 12 weeks.

Then the level three (L3) is daily, and it’s the next 10 workdays. You don’t try to do the daily plan beyond the next ten workdays. You don’t try to do the weekly plan beyond three months.

Q: How do you switch from managing crises to tackling small variations in processes when major issues still happen? 

Jim: There are two kinds of things when you’re managing problems. The first is managing to survive today and take care of the customer today. You have to do that on big crises, little crises -- everything. When you’re clearly just trying to take care of today, that leaves you some time to do the improvement on root cause. What we try to do is do the root cause and solution in the area where we can get the biggest bang for the buck.

You’ve got to make sure we’re making the right thing for today, but in terms of the problem we’re going to chase back and make sure we never do it again, that’s going to be the problem right in front of me, so it’s easier for me to fix it.

Then tomorrow I fix the next one that’s sitting right in front of me. That’s just a very different view than trying to figure out the most important, biggest, hairiest problem and taking three months or six months to fix it. Meanwhile I produce 10 more big problems and then you never get out of the gravy yard spiral, so to speak.

Q: Do you teach structured root-cause problem solving to everybody in your plant?

Jim: No. We just teach it to the team leaders and the folks that are doing the improvements. Operators generally are not doing problem solving because they’re too busy assembling or welding or whatever.

Q: Are you trying to spread lean thinking to your supply chain? 

Jim: I have tried in the past and utterly failed. I have not been successful nor have I seen anybody else who has been particularly successful with it. That’s why I’m in-sourcing.

Q: It seems that this system would be hard to implement if the whole business was out of control. Would you suggest any stabilization activities before going into your system of management. 

Jim: If the managers doing the walk-around do not understand the work, the value of maintaining work, and how difficult it is for an operator to maintain work, then putting the system in place is likely to increase people’s frustration. You don’t have to necessarily install a pull system but you have to implement standardized work and problem solving to start.

If you don’t know how to problem solve and your management folks don’t understand how things get made and don’t understand the world that operators live in, then you’re not going to understand how to support then when you walk around. Walking isn’t going to do anything.

“If the managers doing the walk-around do not understand the work, the value of maintaining work, and how difficult it is for an operator to maintain work, then putting the system in place is likely to increase people’s frustration.”

Generally, you get your operation under control as managers and leaders actually learn the business. Once you have a critical mass of managers and leaders who know the business, meaning they understand how work gets done and the world their operators live in, then the management system can be really helpful.

If you put the daily management system in too soon, you get top managers who don’t know what’s happening on the floor. They just walk around and guilt trip the people who have red [abnormal] signals in their processes. They’ll pontificate about why they think the problem might be occurring.

The operators and the team leaders just roll their eyes because management isn’t really helping them. This system is set up to help, but if you’re not really helping then all you did was put a bureaucratic step that wastes everybody’s time. 

Q: Does your line management use computer tools to update their own metrics or do you have a dedicated Excel expert or resource.

Jim: Neither. We use grease boards and dry-erase markers. Sometimes we get fancy and we use wet-erase markers.

We don’t use computers and Excel [to track metrics] because data becomes invisible inside a box. Pick metrics in the process where it’s easy, fast, and natural to do so.

The engineering types will try to use software to make everything look pretty and calculate out to the third decimal point and have a sheet of paper color-coded up on the board. There are some areas where the team leader does that because they’re doing it themselves that way and it’s quicker for them because that’s how they do it. I prefer grease pen because it is instantaneous and accessible. 

Q: The book states that if an indicator exceeded a previously defined run-rate tolerance, the area manager would turn over a laminated card from green to red. What is the tolerance that is typically used to initiate action?

Jim: If there’s a problem, any problem, there needs to be a countermeasure to ensure that everything’s okay for today and no customers are affected. When the card flips --  in most of our cases -- it’s a red grease mark instead of a green grease mark. When that happens, then there's an expectation that we go find out why we performed less well than average yesterday. There is no tolerance. Once you get a red, then you go look. The only reason you wouldn’t do that is if your quality loop board’s already full. If you’ve already got five quality problems you’re chasing, then we just ensure we’re good for today and we don’t do anything.

It’s not a tolerance issue. Once again, we’re not blaming or holding someone accountable or judging someone’s performance. That’s not what we’re doing.

Next Steps:

Benefit from Jim Lancaster’s lean journey. Learn how to detect and correct the “silent enemy” of lean continuous improvement – deterioration – in your processes.

Save

The views expressed in this post do not necessarily represent the views or policies of The Lean Enterprise Institute.
Search Posts:
The Work of Management
By Jim Lancaster
Was this post... Click all that apply
HELPFUL INTERESTING INSPIRING ACCURATE
4 people say YES
4 people say YES
1 person says YES
1 person says YES
Related Posts
8 Comments | Post a Comment
Anonymous September 21, 2017
1 Person AGREES with this comment

"When that happens, then there's an expectation that we go find out why we performed less well than average yesterday. There is no tolerance. Once you get a red, then you go look." 

At the risk of stating the obvious, aren't about half of your data points going to be worse than average? It seems like you'd be chasing a lot of noise... There's not usually any simple root cause or reason why performance is lower some days in a stable system.



Reply »

Claire Everett September 21, 2017

It depends on what your goal is.  If you're happy with your current results then you would only investigate results that are outside of tolerance, but if you want to drive continuous improvement and lift your average, then this is not enough. 

By investigating everything that falls below average you're saying that 'we want to get better'.  You get better by finding out what causes your lowest levels of performance and fixing that, in this case, what falls below average.

If you do this constantly then you create an upward trend in performance, by moving your lowest performance upward.

Isn't this a better goal than creating a stable system?



Reply »

Anonymous September 21, 2017

The goal is not a stable system. But, if you have one, then reacting and asking why to every below average point is not the way to improve the performance of the system. Asking "what happened yesterday?" is only appropriate if there's evidence of "special cause" variation. Below average is noise unless is 3 sigma lower than the average, basically.

Investigating every below average point says, "I am interested in the impression of improvement." It's a distraction from real systemic improvement work. We should ask, "How do we improve the system?" instead of "what went wrong yesterday?"



Reply »

Claire Everett September 22, 2017

Except that if you ask Lantech "how are you improving the system?" they'll answer "by investigating what went wrong yesterday."

This is not about control charts and signals that a system is in or out of control, this is about creating an envirnment where you're always striving to be better than you were yesterday.  It's not about tweating the system because of every small variation, it's about improving by saying we want to be above this level of performance and when we aren't we're going to find out why.  Whether that target is average or something else is really just semantics.

You say that reacting to every below average point is not the way to improve, but Lantech has created large improvements using this method.  Whether you agree with this method or not, the results speak for themselves.



Anonymous September 22, 2017
1 Person AGREES with this reply

But we can't live out the thought experiment of "maybe they would be more successful if they didn't waste time chasing noise?"

Jim writes about their need to limit the number of improvement investigations... if they're missing or ignoring the chances to investigate signal because their improvement dance card is already full... that sounds like lost opportunity to me.

They've been successful... that doesn't mean that's as successful as they can be. Just because something "works" doesn't mean you'd recommend it to others and it doesn't mean that things can't be better.



Anonymous September 21, 2017
1 Person AGREES with this comment

Q: Do you teach structured root-cause problem solving to everybody in your plant?

Jim: No. We just teach it to the team leaders and the folks that are doing the improvements. Operators generally are not doing problem solving because they’re too busy assembling or welding or whatever.

So "the folks doing the improvements" doesn't include operators?

When you visit Toyota, they say they teach the 7 basic QI tools to every employees. This includes fishbone diagrams, part of the root-cause problem solving process.

Yes, people are busy building cars and trucks. But on a tour, somebody asked "When do team members have time for Kaizen?" The tour guide (a paint shop team member) said if they can't find time during the work day (when the line is down, which happens), then management will approve overtime to stay late and work on Kaizen.

Being "too busy" doesn't seem like a great reason to not engage everybody in continuous improvement.

Or am I missing something in the way this was written?

 



Reply »

Claire Everett September 21, 2017

I think it depends on what is meant by 'teach structured root-cause problem solving'.  To me this means teaching someone how to do the full problem solving process so that they could lead a team.  If this is the case then teaching this to everyone is potentially wasteful, unnecessary, and a bad use of valuable time.

If you mean teaching basic problem solving tools used as part of problem solving, then over time I think everyone should be taught this.  This also happens as a side effect of involving people in various projects and improvement initiatives without the need for formal training.

So are you missing something?  Maybe.

Are we missing something?  Yes... your name.



Reply »

Anonymous September 21, 2017

Yes, it depends on what Jim meant... it's unclear. So I'm asking.

There's no requirement to use my name.



Reply »

Search Posts:
The Work of Management
By Jim Lancaster
Job 1?
Rethinking the Model Line
Working on the Management
A Lean Leap of Faith
A Sensei in One's Own Land