matt hrivnak . com

One of the single most effective projects that I worked on in the past five years was also the simplest; in fact, I proposed this change in my first week of work.

During my interview, the engineering director took me for a walk through the plant.  It was obvious that a lot of changes needed to be made.  He complained, “Every time we have a defect in this cell, we have to scrap the whole lot, 50 pieces.”  I was shocked that a man with a title of “Engineering Director, North America,” could not see the solutions that were right in front of him.  

My initial reaction was to ask a question, “Have you tried reducing your lot size?  Maybe cutting your lots in half?”  His response was even more dumbfounding, “What do you think that would do?” 

So, I went on to describe all of the benefits of Single Piece Flow, and added the obvious fact that if the company requires the entire lot to be scrapped because of one defect, then you would only be scrapping 25 instead of 50 every time that one occurred.  He really got the point once I asked him to think about it on a larger scale, “How many would you have to scrap if you had a lot size of 1,000?”  Of course, he said, “1,000.”

When I got there, the process looked something like this:

There was an obvious disconnect between each process, even though the stations were literally five feet apart.  Each station operated as its own entity, not caring whether or not it was receiving or shipping defective products.  The operators were there simply to put in their time and collect their money.  They cared nothing for the products they were making and took no pride in quality workmanship.  If something failed at one of the inspection stations, then the entire lot was scrapped – big deal – they did whatever someone told them to do for eight hours and that was it – product or no product.

As you can see by looking at the diagram of the process, they had incorporated Inspection stations.  Inspection is a large portion of waste in many manufacturing processes. Sure, it may be necessary in some instances, but the inspection should still be dealt with in-process instead of having it as its own workstation.  

By having Inspection as three entirely separate workstations, defects accumulated lots of 50 waiting in queue and valuable resources were tied up in labor, fixed overhead and much desired floor space.

As a result, I suggested eliminating Inspection from the process.  Well, I should say, I suggested eliminating the wasteful aspects of Inspection.  Because we are dealing with electrical devices that are tested to a standard, the cables must be checked during manufacturing to certify the product as passing the standard, so the Inspection needs to be in there somewhere.

I suggested that we put the inspection testing equipment within the previous station’s area (i.e. the pre-inspection station, e.g. wire insertion, pre-mold, mold), to be checked one at a time as they are made.  The results showed instantly!

• In each of the cases, the inspection operator was eliminated and added to other cells for more value added work.  
• Each of the pre-inspection stations would make one unit and test it instantly.  If there was a problem, it was solved immediately and no other cables would be tainted by the same issue.
• Additionally, in all cases, the inspection portion of the manufacturing could be done (by the testing apparatus) while the operator was preparing the next sample.  

At this point, we also rearranged the cell so that it was in the classic ‘U’ shape which cut the travel distance by 100’.  It had been segmented into two lines, with two operations being 110’ apart from another, this was shortened to 10’.

The bottleneck of the entire line was the Pre-Mold operation which was considerably slower than the other processes (the Wire Stripping and Crimping operations were very slow also, but each had 3 workstations, as these were inexpensive compared to a molding machine).  This was also the highest offender when it came to quality issues.  At this step, the individual wires were consistently getting snipped by the mold, causing complete electrical failure of the connector.

Before the improvements, it was easy for this process to make upwards of 200 bad cables before the Inspection station got around to discovering their was a problem!Because it was the bottleneck, it was absolutely imperative to always keep the Pre-Mold operation filled with work.  To achieve this, we setup a sequenced pull system that started with a supermarket between the Wire Insertion w/ Inspection operation and the Pre-Mold operation.  To keep the supermarket fed, we introduced FIFO lanes upstream.  The supermarket handled multiple varieties of cables because the Insertion operation was the first step in the system where product variety appeared.

In the operations upstream from there we used FIFO lanes because the product mix was entirely the same.  Each was held to a maximum storage amount; the supermarket with 5 and the FIFO lanes with 2.With the addition of the FIFO lanes and the supermarket, we were able to work at the pace of the bottleneck.

Granted we were able to speed up the processing of the bottleneck through a SMED event, which required some machine design from the maintenance department, but it was still the bottleneck regardless. 

Efforts to justify the purchase of another molding machine for that area were just not cost effective. (As part of the SMED, we added a second bottom half to the mold which could be loaded while the current cable was being molded.  After the current one finished, the top half lifted up and the other bottom half slid into place, and the molding continued while the other bottom half was unloaded and reloaded for the next cable.)

Since we were so constrained by the Pre-Mold operation, we were able to use the same operators that ran the operations upstream from Pre-Mold to run the operations on the back end of the system.The results speak for themselves:

• Output of quality products increased from 1.20 units/hour to 5.56 units/hour.
• Quality problems and rework was down by 90%.
• The number of associates went from 14 to 9, allowing those additional 5 people to be moved to areas where they could perform more value added work.
• Associates saw the real effects of their work, taking on more responsibility and having more respect for themselves, the jobs that they perform and the products that they produce.

One of the toughest aspects of the Toyota Production System (TPS) for people to understand is why single piece flow is so important, and how it works.  I can’t count the number of times I’ve answered questions and statements like these: 

“What is single piece flow?”

“If we make one, move one, then our efficiencies will fall and we can’t have that…”

“Why use smaller batch sizes?”

“How will smaller batch sizes help?  We’ll just have more changeovers and our costs will go up.”

“We can’t achieve single piece flow and if we are able to, it will be too costly for us to operate that way.”

“Our customers will never get their orders if we do that because all we’ll ever be doing is changing over and setting up machines…” 

When it comes down to it, single piece flow is the best way that a manufacturing system can be set up.  Now, in most industries, systems are setup that will never allow for single piece flow in the traditional sense because of machining capacities and capabilities.  Examples of this are piece work items that are manufactured automatically by machines that have multiple machining heads that are performing the same task concurrently – something like the minting, pressing and stamping of coins – multiple dies punching hundreds of coin blanks out at one time, etc.  With time and revolutionary machine designs, single piece flow would ultimately be possible. 

In order to highlight the benefits of single piece flow, I’m going to use a list that is characterized by Liker in The Toyota Way (if you don’t own it, buy it – it is well worth the money), pages 95-97.  I’ve kept the list the same, but have added my own reasons as to why these benefits occur: 

1. Builds in quality – this is the aspect that is most overlooked by opponents to single piece flow.  Since you are not dealing with batches, in particular, large batches, any defects can be correctly instantly or removed from the system at that time by the operator.  Defects are fixed or removed instead of being passed on.  Defects also become more noticeable and do not become hidden amongst a batch.  The most significant benefit is that any quality issues are more apt to begin and end with that one particular unit.  This happens because the defect is located, a cause is determined and a solution is remedied (PDCA in action).
2. Creates real flexibility – because you are dealing with a lot size of 1 you can end production for that product at any point throughout the day within the number of minutes it takes for that 1 unit’s cycle time to elapse.  This is improved with the advent of SMED as changeovers are reduced and a larger mix of products can be produced within a shift, servicing more customers than a system processing larger batches of products.  If processing a lot size of 20 takes 5 hours of machine cycling time to complete, then processing a lot of 1 will take .25 hours.  In this way, you would be able to switch products and start producing something else after 15 minutes, instead of waiting 5 hours for the previous lot to complete its cycle.
3. Creates higher productivity – operators focused on single piece flow are working on mostly value added activities leaving less time for non-value added time to interfere.  In addition to this, as each piece is processed, it can be moved onto the next workstation and processing can begin there – eventually you will get to the point where you are producing units at an output rate nearly equivalent to your slowest individual process (i.e. theory of constraints).  Also, any quality issues can be quickly removed or remedied within minutes on a single piece as opposed to reworking an entire lot, this saves larges amounts of rework time that would normally bog down a production line and utilize operators in a completely non-value added manner.
4. Frees up floor space – because single piece flow naturally works within a cell there is less space for the accumulation of inventory between processes.  A cell is setup to maximize production floor space and improve communication between processes to improve quality and increase throughput.  There is no waste associated with defects, scrap, unneeded stacks of raw material, stacks of finished goods waiting for the next process – none of that because as soon as inventory is created (in the form of 1 unit) it is absorbed and processed by the next station and so on, down through the line.
5. Improves safety – single piece flow means that there is no need for large batches to be shuttled back and forth, over thousands of feet within a production facility.  All of the processes are arranged in a cell with minimal space between them.  Again, inventory does not build up and will not require movement, batch sizes are 1 so bins and containers used to move products will be very small, allowing for operators to lift small, light packages instead of large, heavy packages that may contain multiple units.
6. Improves morale – this is a natural phenomenom that occurs because each operator gets to see the outcome of their hard work instantly.  This instant gratification builds a passion for creating well-made, quality items.  Overtime, operators will pride themselves on high levels of quality and products that they produce because they can actually see the benefits that they add to the product – and each one they produce is, in and of itself, a unique special well crafted item.
7. Reduces cost of inventory – simply due to the fact that you have less inventory of raw materials, inventory of WIP, and inventory of finished goods means that your company will be able to dedicate their capital and resources in other areas instead of overhead.  Additionally, any inventory that becomes obsolete because it is sitting around waiting to be processed will no longer occur.  This could mean expanding by purchasing new capital or technologies, improving existing work centers, giving more benefits, providing higher salaries to attract a better workforce.

Recently, I facilitated a kaizen event which was targeted at dramatically reducing the setup time of a Former-Winder twisting machine in a textile plant.  Naturally, we performed a SMED event; dedicating 3 days to the event and 2 days of prior videotaping (performed by me, under ‘normal’ conditions). A little background on the Former-Winder’s operation.  It is a machine that takes in raw material Yarns or Ends, twists them together into a strand, and winds them onto a 3 foot long bobbin.  The Yarns/Ends used can number anywhere from 2 and go up to 150 in some cases.  The bigger the yarn needed, the more raw material packages that will be required.  These raw material packages are put up onto fixtures, called Creels, that hold them while they feed out yarn.  The pictures show the creels in this case.

After loading the creels, the operator needs to install gearing that produces different levels of twist in the finished strand.  Creating a small sample allows the operator to perform a quick in-process check after changing the gears.  If it is okay, the operator puts in an empty bobbin and starts the machine.  If not, then he changes gears with engineering’s help (that’s what the ISO documentation says at any rate). Watching the video we put together this timeline, identifying the CURRENT STATE (i.e. the way in which it was currently being performed)External Setups in GREEN and the Internal Setups in RED:

• Get & read traveler (1 min.)
• Check prioritization (1 min.)
• Get gears (15 min.)
• Install gears (5 min.)
• Return gears to proper place (1 min.)
• Clean/grease machine (5 min.)
• Program computer (if necessary) (1 min.)
• Obtain/order material (45 min.)
• Creel placement (20 min.)
• Create material space (10 min.)
• Get/prepare cart (if necessary) (1 min.)
• Teardown (16 min.)
• Return material (several trips) (9 min.)
• Prepare & inspect creel (foam donuts, check eyelets, inc.) (2 min.)
• Check for/install small packages (10 min.)
• Get/install new packages (several trips) (22 min.)
• Untie packages & feed thru eyelets (38 min.)
• Setup distribution plate(s) (13 min.)
• Setup wax tank (if necessary)
• Locate & get proper die (5 min.)
• Install die (1 min.)
• Return old die to proper place (1 min.)
• Feed machine (6 min.)
• Run (.5 min.)
• Chalk for twist (TPF) (.5 min.)
• Stop & check TPF (.5 min.)
• Sign off by other person (.5 min.)
• Make adjustments (6.5 min.)
• Run (couple minutes)
• Stop & calculate weight/counts (repeat if necessary)
• Check traveler for QTY (.5 min.)
• Run Machine

—————————————————

76% Internal:  179 min = 2.98 hours

24% External:  58 min = .97 hours

Total Time:   237 min. = 3.95 hours

We then went through this list and created a ‘theoretical’ best case scenario that showcased the true Internal and External Setups.  Some of these are not even possible due to current technologies owned by the company:

• Get & read traveler (1 min.)
• Check prioritization (1 min.)
• Get gears (15 min.)
• Install gears (5 min.)
• Return gears to proper place (1 min.)
• Clean/grease machine (5 min.)
• Program computer (if necessary) (1 min.)
• Obtain/order material (45 min.)
• Creel placement (20 min.)
• Create material space (10 min.)
• Get/prepare cart (if necessary) (1 min.)
• Teardown (16 min.)
• Return material (several trips) (9 min.)
• Prepare & inspect creel (foam donuts, check eyelets, inc.) (2 min.)
• Check for/install small packages (10 min.)
• Get/install new packages (several trips) (22 min.)
• Untie packages & feed thru eyelets (38 min.)
• Setup distribution plate(s) (13 min.)
• Setup wax tank (if necessary)
• Locate & get proper die (5 min.)
• Install die (1 min.)
• Return old die to proper place (1 min.)
• Feed machine (6 min.)
• Run (.5 min.)
• Chalk for twist (TPF) (.5 min.)
• Stop & check TPF (.5 min.)
• Sign off by other person (.5 min.)
• Make adjustments (6.5 min.)
• Run (couple minutes)
• Stop & calculate weight/counts (repeat if necessary)
• Check traveler for QTY (.5 min.)
• Run Machine

—————————————————

11% Internal:  26.5 min = .44 hours

89% External:  210.5 min = 3.51 hours

This gave us a starting point for base lining and making improvements.  “How could we improve our current setup process to be as close to this as possible?”  First, we took all of the ‘current’ internal setups that were really external setups, and made them into external setups, changes shown in BLUE.  This saved about 25%:

• Get & read traveler (1 min.)
• Check prioritization (1 min.)
• Get gears (15 min.)
• Install gears (5 min.)
• Return gears to proper place (1 min.)
• Clean/grease machine (5 min.)
• Program computer (if necessary) (1 min.)
• Obtain/order material (45 min.)
• Creel placement (20 min.)
  • Preparation (10 min.)
  • Physical move (10 min.)
• Create material space (10 min.)
• Get/prepare cart (if necessary) (1 min.)
• Teardown (16 min.)
• Return material (several trips) (9 min.)
• Prepare & inspect creel (foam donuts, check eyelets, inc.) (2 min.)
• Check for/install small packages (10 min.)
• Get/install new packages (several trips) (22 min.)
• Untie packages & feed thru eyelets (38 min.)
• Setup distribution plate(s) (13 min.)
• Setup wax tank (if necessary)
• Locate & get proper die (5 min.)
• Install die (1 min.)
• Return old die to proper place (1 min.)
• Feed machine (6 min.)
• Run (.5 min.)
• Chalk for twist (TPF) (.5 min.)
• Stop & check TPF (.5 min.)
• Sign off by other person (.5 min.)
• Make adjustments (6.5 min.)
• Run (couple minutes)
• Stop & calculate weight/counts (repeat if necessary)
• Check traveler for QTY (.5 min.)
• Run Machine

—————————————————

57% Internal:  134.5 min = 2.24 hours

43% External:  102.5 min = 1.71 hours

So, after a few hours of suggestions, some time working the setup process on the gemba, we were able to make a second pass, saving around 72%!  We designed a new creel that could have one side loaded with the current job and a second side that could be prepared for the next job.  This required a 30 day list be made up and completed because of the nature of the maintenance work needed and a few other action items that were external to the SMED event itself:

• Get & read traveler (1 min.)
• Check prioritization (1 min.)
• Get gears (15 min.)
• Install gears (5 min.)
• Return gears to proper place (1 min.)
• Clean/grease machine (5 min.)
• Program computer (if necessary) (1 min.)
• Obtain/order material (45 min.)
• Creel placement (20 min.)
  • Preparation (10 min.)
  • Physical move (10 min.)
• Create material space (10 min.)
• Get/prepare cart (if necessary) (1 min.)
• Teardown (16 min.)
• Return material (several trips) (9 min.)
• Prepare & inspect creel (foam donuts, check eyelets, inc.) (2 min.)
• Check for/install small packages (10 min.)
• Get/install new packages (several trips) (22 min.)
• Untie packages & feed thru eyelets (38 min.)
• Setup distribution plate(s) (13 min.)
• Setup wax tank (if necessary)
• Locate & get proper die (5 min.)
• Install die (1 min.)
• Return old die to proper place (1 min.)
• Feed machine (6 min.)
• Run (.5 min.)
• Chalk for twist (TPF) (.5 min.)
• Stop & check TPF (.5 min.)
• Sign off by other person (.5 min.)
• Make adjustments (6.5 min.)
• Run (couple minutes)
• Stop & calculate weight/counts (repeat if necessary)
• Check traveler for QTY (.5 min.)
• Run Machine

—————————————————

21% Internal:  49.5 min = .825 hours

79% External:  187.5 min = 1.71 hours

We looked heavily into streamlining the leftover Internal Setups and determined that most of the streamlining had already been done. These machines (literally, these same machines) had been working in this plant since the 1940s and people had come up with quicker and better ways of doing things in that time.  Of course, we are still working towards streamlining and looking for those opportunities daily.

So, some team work, some more aware operators and better methods will allow us to service the internal customers of this process a lot faster, in the exact quantities they need and that means we can service a much greater variety of product lines within the same timeframe as before!

It is always important to recognize people’s efforts and to strengthen team building.  One way to achieve this is through the inclusion of memorabilia that states the purpose of a certain project or cross functional team, or even a company wide goal that may span years.  Here are a couple of examples of a button and a mug that were handed out during my days at General Cable.  The button, which lists the plant’s Lean Sigma goals of reductions in Scrap, Rejections and Inventory, were handed out to all associates working in the plant. 

The mug, which simply states “General Cable Lean Associate”, was given to anyone who participated in any Kaizen event.  Several other items were used for motivational purposes and to show the company’s gratitude.  I still have several golf shirts and a jacket that say similar things to these on them.

The main point is to allow every associate and every team member to feel that they are part of something bigger - something worth working towards.  Along the way, each one of these buttons, mugs, etc. remind you of the task at hand.  General Cable’s Lean Sigma program has worked well for the company since its launch back in 2002 (BGC).

Although SMED was clearly developed during the 1900s, there are several cases from throughout history that can be referenced when trying to explain SMED.  The very basic idea of SMED, i.e. preparing and completing External setups while the machine/process is working on the current job, while streamlining Internal setups, etc. (if you are new to SMED, check out my article SMED…What is SMED?), can be seen in common processes that are around us.

The most common of these, which is more or less the cliche reference now because it seems every Lean trainer uses it, is the quick changeover of tires witnessed by Nascar and Indy racing fans at every race.  When you think about changing a tire, you think of several elements, like:

1. Getting your tools
2. Loosening 4/5/6 nuts
3. Jacking up your car
4. Pulling off the old wheel
5. Putting on the new wheel
6. Individually picking up the nuts and resetting them
7. Tightening the nuts
8. Putting the old tire in the trunk
9. Putting your tools away

I assume it is something like that for most people.  Now, the quick connection they make here is that in an auto race, the pit crew is completely ready with their power tools, 1 guy jacks up the car, another takes the a) in cart racing, 1 (and only) nut off while maintaining it in the tooling head, b) in Nascar, the nuts are fixed to the wheel so only tightening is required, another guy removes and replaces the tire, and the nut(s) is tightened, etc.  You can see how that works.

Well, when thinking about SMED, there is a better example that I like to use.  So far, I’ve been a little on my own referencing this and some of my professors had a hard time grasping the concept of this being related to SMED.  Eventually, they saw my point and some of them use it in their classes to this day.  Unfortunately, this is not the most pleasant example of a SMED because of the nature of the resulting action, but it really gives a solid example of SMED.

Since the beginning of firearms, reloading has been a critical, essential, and frankly, a crucial aspect of warfare.  For any of you who are not that knowledgeable when it comes to the evolution of the gun, I’ll fill you in.  After the first couple hundred years of firearm development, the gun was stuck in a period of lackluster progression.  At the time of the American Revolution, muskets, a smooth bored muzzle loading rifle was the most widely used rifle in the world.  Due to the loading nature of the musket, a well trained infantryman could fire about 3 or 4 shots per minute.  Looking at this from a process standpoint, this gives us (assuming ~3 shots/minute):

• Setup time:  20 seconds (includes aiming)
• Cycle time:  ~1/10th of a second

The process for loading a musket, looks something like this:

1. Stand the musket on its butt end, with the open barrel facing the sky
2. Get out the gun powder (from a powder horn)
3. Pour the required amount of powder down the barrel
4. Put the powder horn away
5. Get out the ball (i.e. bullet) (from a snap enclosed pouch)
6. Put the bullet in the barrel
7. Pull out the rifle’s ram rod from its casing on the side of the barrel
8. Stick the ram rod in the barrel and push the bullet fully down
9. Put the ram rod back in the side casing
10. Pick up the musket
11. Cock the hammer (which contains a piece of flint for creating a spark)
12. Get out the powder
13. Pour some more powder on the flash pan
14. Put the powder horn away
15. Aim the rifle
16. Pull the trigger and hope the flint sparks and ignites the powder, firing the gun

That’s a lot to accomplish in 20 seconds. A small pictorial of an American Revolutionary War soldier is below (source:  http://www.americanrevolution.com/images/ContinentalArmy.jpg)

Like many processes, loading a rifle was held back because quicker changeovers were not possible due to technological issues and lack of progress in that area.  Slowly, the technology did improve and the flint-lock hammer was replaced with a preloaded percussion cap (like the kind you can buy in a toy store that makes a loud pop) that could easily be placed on a pin, before being struck by the hammer. 

This removed the clumsly action of pouring small amounts of gun powder into the flash pan while attempting to hold your composure.  Around the same time, ammunition companies began packaging the gun powder and the bullet in a paper encased packet that required the soldier to reach into only 1 compartment instead of 2.  The soldier would simply pull out the packet, bite off the end, and stick it in the front of the barrel.  So, after removing these small, but time consuming external elements, the process looked something like this:

1. Stand the musket on its butt end, with the open barrel facing the sky
2. Get out the paper packet
3. Bite off the top and place the rest of it in the barrel
4. Pull out the rifle’s ram rod from its casing on the side of the barrel
5. Stick the ram rod in the barrel and push the bullet fully down
6. Put the ram rod back in the side casing
7. Pick up the musket
8. Cock the hammer
9. Get out the cap
10. Place the cap on the pin
11. Aim the rifle
12. Pull the trigger

This type of rifle was used extensively around the time of the American Civil War.  At the same time, bullets were being improved through new packaging techniques that encased the bullets and powder in a metal shell that could be loaded into a moveable turret as you would expect to see on a revolver.  Much like the one pictured below:

It took another 25 years or so for this technology to be moved from pistols to rifles, but eventually, the rifle went from muzzle loaded to a pump action carbine to a bolt action rifle to the extremely fast firing machines guns used today.

With the development of the encased bullet, there was no longer a need for a percussion cap and the loading process looked something like this:

1. Preload a small batch of bullets into a ‘clip’ (does not need to be done during battle)
2. Attach the clip to the gun (again, the first clip can be done during down time)
3. Aim the rifle
4. Pull the trigger
5. Repeat steps 3 & 4, as necessary

So, as they do, things progressed and eventually you end up with a mechanism like we see today.  Almost all of the external setup elements have been removed from the gun loading process, however, there is always room for continuous improvement. 

An Uzi (shown), can fire up to 1,200 rounds per minute (rpm)!  That’s 20 rpm or .05 seconds per bullet!!  Some machine guns are capable of firing at rates up to 3,000 rpm.

By advancements in technology that allowed external setup times to be reduced and internal setup times to be eliminated (and the streamlining of all other internal setup times), firearms have seen a throughput increase of 40,000% (using the musket and uzi as examples - (1,200 rpm / 3 rpm) x 100%)!  That kind of increase is unheard of in most manufacturing operations, but can be seen from time to time when dealing with automation.

 How about that for an example of how much improvement you can get from a SMED?!

Over the years I have collected several quotes from various Lean and Six Sigma professionals, historians and trainers.  I have listed most of them below and will continue to expand the list as time goes on.  Enjoy!

“A bad system will defeat a good person every time.” – Deming 

Eliyahu M. Goldratt & Jeff Cox gave us several good quotes related to manufacturing in the book,  The Goal:  Excellence in Manufacturing (later called The Goal:  A Process of Ongoing Improvement).  The book, a work of fiction, leads readers into the concept of the Theory of Constraints through an easy to read novel setting.  These are some famous quotes from The Goal:

“Make the bottlenecks work only on what will contribute to throughput today … not nine months from now. That’s one way to increase capacity at the bottlenecks. The other way you increase bottleneck capacity is to take some of the load off the bottlenecks and give it to non-bottlenecks.” - quote from The Goal

“If we reduce batch sizes by half, we also reduce by half the time it will take to process a batch. That means we reduce queue and wait by half as well. Reduce those by half, and we reduce by about half the total time parts spend in the plant. Reduce the time parts spend in the plant and our total lead time condenses. And with faster turn-around on orders, customers get their orders faster.” - quote from The Goal

“An hour saved at the non-bottleneck is a mirage.” - quote from The Goal

“I say an hour lost at a bottleneck is an hour out of the entire system. I say an hour saved at a non-bottleneck is worthless. Bottlenecks govern both throughput and inventory.”  - quote from The Goal

Again, with any of the lean quotes I present, I try to be as accurate as possible.  If you see any discrepancies, please email me.

Taiichi Ohno (1912 - 1990) was a Toyota executive and one of the chief architects of the Toyota Production System.  He wrote several books about Toyota, most notably Toyota Production System:  Beyond Large-Scale Production and Workplace Management.

These are some famous Taiichi Ohno quotes:

“All we are doing is looking at the time line, from the moment the customer gives us an order to the point when we collect the cash. And we are reducing the time line by reducing the non-value adding wastes.” - Taiichi Ohno

“The only place that work and motion are the same thing is the zoo where people pay to see the animals move around” (not exact phrase) - Taiichi Ohno

“Where there is no Standard there can be no Kaizen” - Taiichi Ohno

“Why not make the work easier and more interesting so that people do not have to sweat?  The Toyota style is not to create results by working hard. It is a system that says there is no limit to people’s creativity.  People don’t go to Toyota to ‘work’ they go there to ‘think’” - Taiichi Ohno

“Costs do not exist to be calculated. Costs exist to be reduced.” - Taiichi Ohno

“The key to the Toyota Way and what makes Toyota stand out is not any of the individual elements…But what is important is having all the elements together as a system. It must be practiced every day in a very consistent manner, not in spurts.” - Taiichi Ohno

“The more inventory a company has, the less likely they will have what they need.” - Taiichi Ohno

“Data is of course important in manufacturing, but I place the greatest emphasis on facts.” - Taiichi Ohno

Again, with any of the lean quotes I present, I try to be as accurate as possible.  If you see any discrepancies, please email me.

Shigeo Shingo (1909 - 1990) was an Industrial Engineer who worked as a consultant with a number of companies before finally being brought in to further develop the Toyota Production System.  His most notable contributions and accomplishments include SMED (quick changeovers), Standard[ized] Work and methods of error-proofing.

These are some of his most famous quotes:

“The most dangerous kind of waste is the waste we do not recognize.” - Shigeo Shingo

“When you buy bananas all you want is the fruit not the skin, but you have to pay for the skin also. It is a waste. And you the customer should not have to pay for the waste.” - Shigeo Shingo

“A relentless barrage of ‘why’s’ is the best way to prepare your mind to pierce the clouded veil of thinking caused by the status quo.  Use it often.” - Shigeo Shingo

“Improvement usually means doing something that we have never done before.” - Shigeo Shingo

“The best approach is to dig out and eliminate problems where they are assumed not to exist.” - Shigeo Shingo

“Are you too busy for improvement? Frequently, I am rebuffed by people who say they are too busy and have no time for such activities. I make it a point to respond by telling people, look, you’ll stop being busy either when you die or when the company goes bankrupt.” - Shigeo Shingo

Again, with any of the lean quotes I present, I try to be as accurate as possible.  If you see any discrepancies, please email me.

How to calculate OEE: 

OEE stands for Overall Equipment Effectiveness and is the most comprehensive calculation used to determine how effectively you are utilizing your equipment.  It is an important part of total productive maintenance, can help better calculate efficiency losses, and is useful when incorporated into the cycle time calculations.

In order to calculate OEE, you will need to collect some data:

1. PERFORMANCE RATE - The Performance Rate is calculated by looking at the actual operating speeds of your machines in comparison to the operating speeds that they were originally designed for those machines or the products being run on them.
   • Product A was designed to run on Machine 1 at speed setting 10, the highest speed possible.  Due to the machine’s old age and fragile state, it can only run safely and produce good versions of Product A at a speed setting of 6.  The Performance Rate would then be 60% (i.e. 6 / 10
2. AVAILABILITY RATE - The Availability Rate is calculated by measuring any production losses due to downtime from equipment failing, breaking down, etc. as a portion of scheduled manufacturing time.
   • Machine 1 runs 36 hours for every 40 available manufacturing hours due to breakdowns.  The Availability Rate is then 90% (i.e. 36 / 40).
3. QUALITY RATE - The Quality Rate is calculated by determining the amount of losses due to quality issues like scrap and rework as compared to the total parts processed.
   • Machine 1 ran 100 pieces of Product A, but only 98 met the quality specifications.  The Quality Rate would then be 98% (i.e. 98 / 100).

 Now, once you have these 3 important measures, the calculation of OEE is very simple:

          PERFORMANCE RATE  x  AVAILABILITY RATE  x  QUALITY RATE

Using the examples from above (Performance Rate = 60%, Availability Rate = 90%, Quality Rate = 98%):

          60%  x  90%  x  98% = 52.9%

 Essentially, the OVERALL EQUIPMENT EFFECTIVENESS (OEE) is the % of effective use that you are getting out of your piece of equipment.  It is a compounding, thorough look at your true equipment uptime as a percentage of your total available manufacturing time.  This metric is important for loading workcenters and determining capacities because you are completely aware of a particular machine’s (or machine type) total ‘real’ uptime.