Learnings from a Kanban Implementation

A few years back I had the pleasure of setting up my first kanban system.  At least one that I was consciously setting up, unlike the one I blogged about in the pastJoe Wilson, my recent guest blogger, and I were tasked with developing a kanban system, train 550 people across 3 shifts using a simulation we develop, and implement all within 8 weeks.

(click on image to see larger version)

This was no small task as you can imagine.  The facility had 4 main process: injection molding, painting, electroplating, and assembly.  The processes were spread out over 450,000 square feet.  We also thought the only way anyone would have a chance to retain how to use the kanban system was to have all 550 people touch/participate in the simulation.

We designed the simulation using the actual kanban cards that would be used out on the floor so people would be used to seeing them.  The simulation also only used 6 people at time.  We weren’t dummies.  We knew how many sessions that meant and we were going to do it in one week.  We decided to train 6 people from HR on the process and the simulation.  They knew as much as we did at the time so why not.  We went over it with them several times until they understood it.

We were the only facility in the company that hit the mandated deadline for this task.  Yes.  It is one to this day that I am proud of.  Now that doesn’t mean the system worked all that smoothly.  It did allow us to jump into the learning cycle much faster than everyone else and start making improvements.  It was an AMAZING learning experience.

I did all this set up so I could share some learnings……OK and maybe toot Joe and I’s horn a bit for meeting the deadline 🙂

  • Small kanban cards (3×3) on big portable racks didn’t work too well – Our solution was to permanently mount the cards to the racks that were specifically designed for the parts.  We scanned the card when full and then when empty.  Another possibility is to make the kanban cards big (8×4) so a card can’t be stuck in a pocket easily or is easy to see if missing.
  • Transporting the cards large distances to put in the “Return To Supplier” bin did not work – Taking the cards across the department allowed people to stick them in pockets until they walked over there and also gave them more opportunity to drop cards on their way to the bin.  The permanently mounted cards helped with this because we went visually off the empty bins.  This forced us to create a visual management system to see them easily.  Another solution is have a “Return to Supplier” bin no further than 12 inches from where the card is removed.
  • Start with too much inventory instead of too little – When parts ran out because we sized the kanban too small people wanted to blame the new process and not bad math on our part.  In most cases, we sized properly or too large.  When someone argued the process was to blame we showed how it was working for the other parts and we just needed to add more kanban cards to the system.
  • The final one was timing of launch – We were an automotive supplier and we went go live with the process in the middle of June.  In automotive, almost all manufacturers shutdown for a week around July 4th to retool for the new model lines.  The suppliers do to.  So we were live for one week and then told everyone to violate the kanban because we had to build a bank of parts for the few customers that didn’t shutdown.  Whoops!  That was a hard pill to swallow but we did and we put a process in place of using non-replenishment kanbans (my next post will talk more about this) for building a bank of parts.

The list could go on forever on what we learned from this experience.  These were the highlights that spurred other learnings.  In the end, the system worked very well but it took us some time to get there.  I hope others can take from out learning and not have to make some of the same mistakes we did.

Posted on February 4, 2011, in Flow, Learning, Manufacturing, Problem Solving, Tools and tagged , , , , , , , , . Bookmark the permalink. 3 Comments.

  1. Matt – you should add that you asked HR people to help. As Cheryl Jekiel points out in her new book, Lean Human Resources, the HR department is too often left out of the lean transformation. As you showed, they can really supplement your need to train people. They can also help document job content, help with a skills matrix, and help you root out policies that conflict with lean objectives and help better align them. Her book is not just for HR. It will help ops managers figure out what to demand from both company and HR leadership in breaking down that silo.

    • Karen, thanks for the information on Cheryl’s book. That sounds very interesting. There aren’t many books, to my knowledge, out about lean and HR. You are right. HR should be helping in many other ways. This was just one way they helped us in the plant I was in at the time.

      Chris, thanks for the compliments. I appreciate it.

  2. Matt,

    Great job! This is one of your best posts and you clearly delivered for your company. The illustration is very helpful and the lessons learned are valuable. Thanks for sharing.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: