Blog Archives

One Man’s Lean Journey: Exposure to SMED and Shigeo Shingo

My lean journey has taken a very common path. I started out learning the lean tools and concepts. Then I evolved to learning how the tools and concepts fit together to create a system that relied on people behaving differently than I was used to seeing. Finally, I was enlightened and understood the thinking that was behind it all and why it works.

For me, I was fortunate enough that my lean journey started back in college. I wish Purdue had incorporated more lean into their Industrial Engineering program, but it didn’t. My lean experience came from my four year internship with Thomson Consumer Electronics. It was a great experience. I would work full time for the company for a semester and then go back to school for a semester. It was a program Purdue had designed with several companies.

My manger at TCE had read books by Shigeo Shingo, specifically A Revolution in Manufacturing: The SMED System. During my second semester with the company, he required all of his interns read the book. The manufacturing geek in me found it absolutely fascinating. I couldn’t wait to put some of the SMED techniques into practice.

My first opportunity to use SMED was in our injection molding house. I worked with a group of machine technicians to reduce the changeover of molding tools. This was a good first opportunity. Shingo had discussed very similar machinery in his book, so it was easier to translate to this actual application. We were able to reduce the changeover time from three hours to one and a half hours. A 50% reduction. I was excited at the time, but looking back now we were still a long way off.

My second opportunity was more a learning because it was on a manual paint booth in one of TCE’s facilities. I had to translate how it would work based on my own understanding. I decided to use videotaping this time. I video taped the changeover on second shift. I was on the road and it was a last minute project. This didn’t go so well. I was able to make my manager laugh hysterically when we watched the playback of the changeover. At one point it became lunch time for the crew so they dropped everything and left. I kept the tape rolling so I could time the downtime. I didn’t have a tripod and got bored quickly, so there is a period where I am spinning and dancing around the area with nobody around. I was loopy by 1AM. We did get a 30% reduction in changeover time so it wasn’t all bad, but the last minute planning definitely showed. Plus, we didn’t included anyone from the facility in the redesign.

It was years before I reflected on these two SMED events and changed my approach.

Reflections:
* Proper planning in advance is critical to a successful SMED event. If you are videotaping, get a tripod. More importantly, how are you going to document the changeover? What is the current changeover time and push to reduce by 75%

* Always include people who are involved in the changeover in the SMED event. Their insights and buy-in is critical to sustaining the gains.

Advertisements

The SMED-ing of Football

I was looking for a change of pace for the whole Pit Crew/Racing example used to illustrate the SMED process.   Maybe I just got frustrated with it because, although it does show an example of a fast changeover, I’m not sure how “Lean” the whole process is.   Luckily, with football season around, I have found a new example to talk about.  (For those who may want to stop now, I’m talking about “American Football”, not what most everybody else in the world calls football.)

Judging by ratings, more people watch the NFL and College Football than motorsports.  That’s kind of important if I want to come up with something other than the tried and true pit crew metaphor.  Chances are if you’ve watched a game over the past few years, the talking heads in the booth have spent some of their time talking about “hurry up” or “blur” or some other variance of a no-huddle offense that is the greatest thing since the forward pass.  This is likely to be a huge topic of conversation early in the NFL season as one of the most well known practitioners, Chip Kelly, has left the University of Oregon and is now performing some degree of his voodoo for the Eagles.  What does that have to do with Lean and changeovers?  Hopefully I can show you.

One of the perceived benefits of the no huddle offense is that you can run more plays in the same amount of time because you can run them faster.  How does that happen?  Well, it starts with looking at a huddle as a changeover.  If you can exchange in your head the whistle stopping the previous play for “last good piece” and being in place for the snap of the ball for “first good piece”, the process is actually quite logical.  Here is a typical huddle:

1

In a no huddle system, you identify the steps in the changeover that don’t add value.  In this example, the steps that don’t add value are the “running back to the huddle” and the “communication in the huddle” steps.  From there, the steps of going to line up in your position and coaches communicating the play have to occur in parallel, and add in speeding up the movement from the end of the last play to getting back to the line for the next play.  The diagram starts to look like this:

2

3Okay, it’s not a perfect metaphor, but it’s not a bad start.  Plus, it makes both watching football and talking about SMED a little more interesting.  (For the sake of clarity, yes, I realize that no-huddle offenses aren’t a new development in the past 2 years.  Also, from a football standpoint the speed of the plays is mostly important because it allows you to constantly tweak the pace of plays being executed so you can outflank the defense…but that’s a topic for another time and place.)

Are there any other good parallels that are in use to talk about SMED or another Lean concept?  In the interest of space, I’ve condensed some of the “how” out of this post.  If you’re interested, post a comment or drop me a line (joewilsonlean at gmail dot com) and we can discuss this concept or your other examples further.

Bill Waddell Highlights the Importance of Quick Changeovers

Recently, Bill Waddell published a great blog post highlighting the benefits of reducing changeover time.  The post was about reducing the manufacturing cycle which is the time it takes to produce every product.  Bill used an actual story from a client of his.

To be sure there were other inputs to the improvement – a simple demand pull method and more statistically valid methods of determining the inventory needed to cover the cycle, but set-up reduction was at the heart of it, and the improvements there translated into significantly less inventory, better on time delivery and lower costs.

Hearing stories like the one Bill wrote about just reaffirms the importance of reducing changeover time.  It is something that companies take for granted.  Most companies don’t see it as critical to achieving the business needs and goals.

Bill gives two great examples of where changeovers have been deemed to critical or their business would die.

I recently saw a cruise ship go through the change-over process and it is really quite similar.  Dock and disembark some 3,000 passengers and their luggage and take on 3,000 new ones, restock tons of food and supplies, perform necessary maintenance to the ship, then sail again all in the course of a few hours.  They have all sorts of specially designed devices and a very well trained crew of folks to do it … but they have to.  That turnaround is the key to their success.  In that regard they are a lot like the NASCAR or Indy cars – change-over fast or die.

Manufacturing companies don’t take this view.  My question is “Why not?”

If results like the company Bill talks about receives such incredible benefit that help them stay viable and profitable, why aren’t more companies doing it?  What other evidence is needed?

Does your company consider quick changeover critical to it’s success?

Lean Concepts are Intertwined

When working with an area, department or organization to implement lean people like to focus on implementing a specific tool or concept, but it’s not that easy.  The concepts and tools are so intertwined that focusing on one is really difficult to do.

An example would be implementing SMED (or quick changeover) across a facility.  A vast majority of the time a large opportunity for improvement is through organization, having the tools you need where you need them and knowing when and where to be for the changeover.  Immediately, other concepts that come to mind are visual management to understand when and where to be without having to ask.  Also, 5S which can help with organization and having the right tools at the right spot.  5S is also a component of visual management.

A second example is implementing strategy deployment.  There is standard work to how to cascade catchball down through the organization and it should be documented to be repeatable.  Then the strategy is usually documented on an A3 to help communicate the message and most companies use visual management to show progress to the entire organization as time progresses.

As a person working to help others implement lean, it is OK to let them believe they are only focusing on one concept to start.  Sometimes thinking about the intertwined concepts can become overwhelming.  Let the customer focus on the one concept and introduce the other concepts through the backdoor.  There is no need to call out the lean concept.  Just discuss what a way to help them solve their problem in further implementing the concept the are focusing on.

At a later date, you can show them how they have actually implemented other lean concepts successfully.  This helps build their confidence, shows further progress then what they believed and builds momentum to continue moving forward and taking more on.

Don’t get hung up on explaining all the intertwined concepts.  Delivery on the needs of the customer and it will all work out.

Real Life Lean

I live in a part of the United States where houses are made of board siding which requires the siding to be painted every 5-7 years.  This is new to me because the parts I have lived in prior the houses were made of brick or aluminum siding.  Both do not require any regular maintenance.  So a few weeks ago, we had our house painted.

I had watched paint crews take a couple of days or so to paint neighbors’ houses.  The crew we hired painted the house in 1 day.  The process was amazing.

Most crews were 2 people.  This was a 4 man crew.  When they started, two men started taping off the windows and fixtures on the front of the house.  The main painter started mixing the paint and hooking up his spray gun.  The main painter started painting the house the base color and following the tapers around the house.  When the tapers finished, the main painter was on the back of the house and the front of the house was dry enough start painted the trim a different color.  The tapers started painting the trim while the sprayer was still working around the rest of the house.  When the sprayer finished the two working on the trim color got help from him.  Then one member broke off and started taking off the tape and then started doing the touch ups.

The fourth crew member was a runner.  He mixed paint, brought paper and tape to the tapers, relieved painters during breaks and anything else that was non-value added work.  He was the support system that kept everything going.

1 day.  10 hrs.  4 men. 1 custom painted house.

It was incredible to watch.  They get paid by the job so everyday can be a payday but not if you take more than one day to do the job.

What real life examples have you seen?

The Importance of TPM and SMED

There are two main types of manufacturing environments that I have worked in.  One is a heavy manual operation environment usually assembly lines.  The other is a heavy machining environment with painting and plating lines or injection molding machines, etc…

In a heavy machining manufacturing environment, there are two lean concepts that I believe need to be done very well to have sustained success…Total Productive Maintenance (TPM) and Quick Changeover (SMED).

Some may say continuous flow is the most important concept.  The issue is flow can not happen if the machines are down unexpectedly or if the changeovers are taking a long time.

If a machine goes down, then flow stops all together or WIP is built up between the processes so when the machine goes down unexpectedly the following process does not go down also.

The longer a changeover takes the more WIP needs to be built up in front of the next process.

In either case, this causes the flow to be interrupted. Any WIP that is introduced between the process interrupts the flow of that particular item even when things are running smoothly.  The item has to stop and wait will the items in queue in front of it go through the process first.  This can make it difficult to link processes together to create the smooth flow.

Yet, these are two tools that I rarely see as a high priority to master in a manufacturing environment.  5S seems to be the tool to try and master first and then standard work.  I do believe these are extremely valuable and necessary concepts, but if you don’t implement them trying to solve a particular business need then sustaining the gains is hard.

5S and standardized work can both be implemented through a TPM and SMED program.  5S can be used so tools are in their place and ready to go when a changeover is taking place or maintenance is being done so the machine is down for less time and production can start back up.  There should be standardized work to make sure the TPM is happening.  Checklists is a common way to have standardized work for TPM.  There should also be standardized work for changing over a machine in the quickest, safest, most efficient manner that is known.

All the concepts are important.  I see TPM and SMED as two concepts that are vehicles to other concepts introduction and success.

Are there other areas TPM and SMED are as important?  Maybe with IT systems?

Production Sacrificed in the Name of Changeovers

Is it ever OK to value the number of changeovers you do in a day over your production numbers?  I say no.

I was with a customer recently that did just this.  The customer has done a great job of setting production goals for a press per shift.  On the production board, they write the production numbers in green if the meet or exceed the goal and in red if they do not.

Normally, this is great.  The customer is making the problem visible and easy to see.  Then I noticed that a number below the goal was written in green.  So, I asked about it.  The customer replied the operator did a lot of changeovers that day so we give them green if they do so many changeovers but don’t hit the production goal because the changeovers eat up a lot of their time.

The managers were giving a built in excuse for the operators to not meet the production goal.  If the goal was set with capability and meeting customer demand, then why is it alright to produce anything less than the goal?  This tells me they are not putting a big enough emphasis on changeover reduction.

The question should be changed to understand what is the changeover time needed.  If the largest number of changeovers I need to do in a shift is X and I am accounting for time T to do the changeovers, then my changeover time target should equal T/X.  Example: I allow 1 hour for changeovers and I need to be able to handle 10 changeovers in a day, then my changeover time target should be (60 min) / (10 changeovers) or 6 min/changeover.

If my current changeover time is more than 6 minutes, then I should be doing some sort of SMED (Single Minute Exchange of Die) activity to get the time to 6 minutes or less.

The number of changeovers can never be an excuse for why it is ok not to hit a production goal.  The mindset should be to continue to reduce the changeover time and ideally eliminate the changeover time so the production goals can be met.

SMED Part 3 – Reducing Trials

A commonly used lean tool/concept in manufacturing is Single Minute Exchange of Dies (SMED) or quick changeover.  By definition changeovers from one job to the next is waste.  It does not add any value to the product/service, nor is the customer willing to pay for it.  Since it is waste but necessary in many operations, the goal should be to be as quick and as efficient when changing over as possible.

Shigeo Shingo showed how getting changeovers done in just a few minutes can reduce the batch size that can be produced, which creates less inventory and increases the cash flow.  When achieved, a changeover that is done in less than 10 minutes will save a lot of money.  The ideal state is to get the changeover to instantaneous so no capacity is lost.

During my time I have seen what I call three levels of the SMED concept that can help depending on where you are with implementing quick changeovers or lean.  This is the third of three parts explaining the different levels I have seen.  I hope this will help others with their SMED efforts.

The last topic in this three part series is about reducing the number of trials at start up.  The definition of a changeover is from the last good piece of the current run to the first good piece of the next run.  So any trials to align or purge or anything else in order to get a good piece is considered part of the changeover.

Here are a few suggestions to get the thinking started on how to reduce the number of trials during a start up:

  • Pre-set adjustments – This is where you can adjust your machine quickly to a pre-determined setting that should be very close if not exact for the job you will be running
  • Offline fixturing – Create a fixture offline that simulates the functionality of the machine.  Then set up the press to receive the pre-positioned die just like the fixture does.  This should help on the accuracy of the process.
  • Set pins – Build pins that would allow a fixture (such as a screen) to be set in the same place every time.  This will help with pre-set adjustments and sliding in a fixture that was aligned offline to be accurate each time.

The three things I have discussed about SMED are not the only parts of SMED.  There is a lot more to SMED and quick changeovers that can help.  Learning to become very quick with changeovers can really help drive the business in reducing inventory and increasing cash flow.

SMED Part 1 here

SMED Part 2 here

SMED Part 2 – Quick Releases

A commonly used lean tool/concept in manufacturing is Single Minute Exchange of Dies (SMED) or quick changeover.  By definition changeovers from one job to the next is waste.  It does not add any value to the product/service, nor is the customer willing to pay for it.  Since it is waste but necessary in many operations, the goal should be to be as quick and as efficient when changing over as possible.

Shigeo Shingo showed how getting changeovers done in just a few minutes can reduce the batch size that can be produced, which creates less inventory and increases the cash flow.  When achieved, a changeover that is done in less than 10 minutes will save a lot of money.  The ideal state is to get the changeover to instantaneous so no capacity is lost.

During my time I have seen what I call three levels of the SMED concept that can help depending on where you are with implementing quick changeovers or lean.  This is the second of three parts explaining the different levels I have seen.  I hope this will help others with their SMED efforts.

Another big concept in SMED that I have seen help many times, is the quick release and tool modification concept.  Too many times, I have seen examples of turning screws or bolts that are 2 inches long in order to secure something.  Quick release clamps give the functionality of holding something in place without the need to screw something in.

(click on image for larger view)

The example above shows a screw with a knob that was used on 4 corners of a screen to hold it in place.  A team changed this to 4 lock down clamps that take less time to secure than one of the screw knobs.

Another concept is trying to find ways to modify tools that are used in order to prevent wasted movement during the changeover.  I worked with a team one time that needed 2 different size wrenches to do the changeover inside a piece of equipment.  Everything they needed the other size they would have to get out of the equipment, get the wrench and then get back in to work.  The team decided to cut the two wrenches in half and weld the sizes they needed together in order to make things quicker.  Here is another example:

(click on images to see a larger view)

On the left the operator has to scoop the powder and then strain it into a container and then pour it into the machine.  The modification on the left had the strainer built right into the pour slot for the powder on the machine.

Getting quick releases and modifying tools may be something that can help your SMED efforts.

SMED Part 1 here

SMED Part 1 – 5S Will Improve Your Changeover Times

A commonly used lean tool/concept in manufacturing is Single Minute Exchange of Dies (SMED) or quick changeover.  By definition changeovers from one job to the next is waste.  It does not add any value to the product/service, nor is the customer willing to pay for it.  Since it is waste but necessary in many operations, the goal should be to be as quick and as efficient when changing over as possible.

Shigeo Shingo showed how getting changeovers done in just a few minutes can reduce the batch size that can be produced, which creates less inventory and increases the cash flow.  When achieved, a changeover that is done in less than 10 minutes will save a lot of money.  The ideal state is to get the changeover to instantaneous so no capacity is lost.

During my time I have seen what I call three levels of the SMED concept that can help depending on where you are with implementing quick changeovers or lean.  This is the first of three parts explaining the different levels I have seen.  I hope this will help others with their SMED efforts.

The most basic level of SMED I have seen is organization.  The lean tool/concept that comes to mind is 5S.  Knowing when to use 5S is the key.  If the area where the changeover occurs is not organized or operators and technicians spend a lot of time looking for the proper tools/parts for the changeover then 5S is a good tool/concept to pull out of the lean toolbox.  Operators shouldn’t leave the area to go look for the tools/parts that should have been in the area of the changeover.

Here are just a few ideas to get you started in thinking about how to use 5S to help reduce changeover time:

  • Have everything that is needed brought to the area and ready before the last good piece of the run comes off the machine (some of this may be brought to an area on an organized cart)
  • Have the tools/parts needed within reach of the physical location from where it will be used (no walking, even in the area, to get what is needed)
  • Think about organizing the tools in the order they will be used (this can also help make visual where someone is in the changeover process)
  • Be sure to have plenty of tools (you have a wrench but if it is being passed between two people why not buy a wrench for each person so they can work at the same time)

I have seen changeover times drop by 50+% just from implementing good 5S practices around changeover tools/parts.