## Takt Time: How Slow Can You Go?

A long, long time ago – in the days when computer programs were coded as holes in punch cards – I was in ROTC* in college. Twice a year we had a “PT” (Physical Training, I think) test that consisted of measured performance on five “events.” One of those events was a 2 mile run. To get a maximum score of 100 points, the participant had to complete the 2 mile run in 14 minutes and 9 seconds. Why? I have no idea. But that was the way it was.

My buddy John and I enjoyed running, and would often work out together. Our school was in Potsdam, New York, a place known for rather brutal winters, so we ran on a 1/10 mile indoor track.

We practiced the PT test. John would track our total time for 2 miles (20 laps around the 1/10 mile track). I would measure lap times. Since 14 minutes is 840 seconds, we knew that if we could consistently make 42 second laps, we would complete two miles in 14 minutes, and get the maximum score with 9 seconds to spare.

The track had hash marks at each quarter point, so we knew we had to hit quarter 1 between 10 and 11 seconds, half way at 21, the 3/4 mark between 31 and 32, and the lap at 42. We would check and adjust our pace accordingly, striving to hit exact 42 second laps every time.

To be clear, we could hold that pace many times further than 2 miles. It wasn’t a matter of conditioning. We weren’t going all out. We were going fast enough. And that was the point.

When we took the test, other cadets were going all out, passing us, and turning in much faster times. Others would try to “pace themselves” and then sprint as fast as they could for that last lap. As a general rule, although the instructors were calling out elapsed times as people went by, these cadets weren’t all that aware of the speed they had to hold. They were just going as fast as they could.

Meanwhile, John and I, running together, and tracking our cycle times (lap times) vs the takt time (42 seconds / lap) would come in at pretty much exactly 14 minutes and get the same score as everyone who had finished ahead of us: 100 points.

After our timed run was done, we would keep running, offering encouragement and pacing for those cadets who were struggling a bit. Everyone finishes, nobody left behind.

A couple of the instructors were curious why we didn’t go for a faster time. And many times we did when we were just working out. We were capable of breaking 12 minutes – not competitive times in a track meet, but respectable. The simple fact was that going faster wasn’t necessary to accomplish the goal.

## Takt Time and Cycle Time

This is the whole point of having a takt time. It answers the question, “How fast must we go?” It doesn’t answer “How fast can we go?” nor does it answer “How fast should we go?” I fact, John and I could have run those laps almost (but not quite) half a second slower than we did – which would have eaten into the 9 second buffer we established. Aside from making the math easier, that buffer also gave us a small margin for even something as bad as taking a stumble and standing back up. Also, of course, we could make up 5-10 seconds a lap if we really had to. But we never did. Time: 14:00. We were very consistent.

## Rate vs. Output

I encounter a lot of production managers who are so conditioned to focus on the daily output that they don’t even think about the critical factor: How fast are you running vs. how fast do you need to run? In other words what is your rate of output vs. your takt time?

Instead they tend to, at best, count units of output without really paying attention to the time interval between one and the next. In the worst case many units are started at once, and people swarm from one operation to the next during the day – the equivalent of that mad sprint trying to make up as much time as possible. They don’t really know if they will succeed or not until the end of the day (or month!).

## It Isn’t a Race

The “just” in “just-in-time” is “just enough resources” to “just make the output you need” in any given time interval. As the operation is streamlined, the same effort is able to accomplish more. Where to put that additional capacity (which costs nothing additional since it has been there all along) to create more value should be the challenge the organization is trying to meet.

My experience has been that managers and leaders often struggle to adopt this “rate” mindset and let go of chasing an inventory number. In the words of the late great philosopher Kenny Rogers – “There’ll be plenty of time for counting when the dealings done.”

*ROTC (Reserve Officers Training Corps) is a U.S. program where college students can earn a commission in the U.S. Armed Forces while earning their degree.

## Toyota Kata: What If There Is No Takt Time?

The default Starter Kata for “Grasp the Current Condition” places heavy emphasis on takt time and variation in timing of a regular process. However a lot of processes, both within manufacturing as well as in other domains such as health care, don’t seem to have any kind of regular heartbeat.

As Steve Medland pointed out at KataCon, this can present a struggle for a novice learner, as well as for a lot of coaches.

Before we get into ways to deal with this, I want to level set on what takt time is, and what it does for us.

## Why Takt Time?

From an industrial engineering standpoint, takt time is an expression of how much capacity you need.

The traditional way to calculate takt time is to divide the total time available by the output required in that time. This gives us a “time per unit of output” that we have to achieve if we are going to get everything done. In other words, takt time is the required rate of production.

The whole goal of an ideal “Just-in-Time” system is that we have only the capacity required to meet the demand. If the system is even able to run faster than the takt time, we have excess capacity. Excess capacity = extra cost, and “overproduction” is a symptom of having excess capacity. Note that this is the “ideal” – it isn’t anything we can realistically achieve. The concept gives us a direction to strive toward.

Also Note: Determining how much capacity you need has absolutely nothing to do with how much capacity you have.

OK, that answered the question “What is takt time?” but not the question I posed: “Why takt time?” After all, I could just as easily say I need the capacity to product 96 units per day. But that doesn’t answer the question, “How fast do I need to go?” And that is what takt time gives us.

Think of it this way. If I need to make 96 units during the course of an 8 hour day, then I need to have made 48 units in four hours.

To make 48 units in four hours, I need to make 24 units in 2 hours. Which means 12 units in 1 hour. 6 units in 30 minutes. 3 units in 15 minutes. One unit every 5 minutes. And that is my takt time. (This is an oversimplification since I did not subtract break times to make the math easier to make the point.)

Thinking of it this way gives the people doing the work a quick way to know, very quickly, if they are ahead or behind. They can ask, “How long did this unit take?” and compare that with, “How long did we have?”

## A Point of Comparison

Which brings us to the “Why.”

If I measure the time between units output at the end of the line, I can compare the actual time interval (the cycle time) with the required time interval (the takt time).

• If we need to complete a unit every five minutes (takt time), AND
• We know that our process can do that when there are NO problems, THEN
• We can see very quickly if there IS a problem. We have to attack a source of variation and work on stability.

On the other hand,

• If we need to complete a unit every five minutes (takt time), and
• We know that our process is not capable of doing so even when running smoothly, then
• We know we have to change the entire system to make rate.

Distinguishing between these two conditions is the main benefit of building a cycle time run chart (step 3 in the Starter Kata of Grasp the Current Condition.) That is a topic for another post, or just get a copy of the Toyota Kata Practice Guide ;).

The issue comes when people don’t see a steady rate of demand.

Sometimes they generally know how much time is in the day, but they see demand fluctuating all over the place.

This is true in manufacturing work as well as other cases, such as engineering work, software development, and a lot of cases in health care. The time to complete one “unit of work” varies, so it is hard to see any kind of cadence to the output even if the work is steady.

## Key Question: Are you ahead or behind?

And how can you tell?

Regardless of all of the variation, though, we still want to know the answer to questions such as “Are we on track to get everything done today?” or “Has my load exceeded my capacity?” or “Is the backlog increasing, decreasing, or holding steady?” unless we are simply relying on luck. Asking and answering these questions is the purpose of many of the “lean tools” – including takt time.

## When to Bring it Up

Everything above, though, is information for the coach to keep in mind. What a lot of us (myself included) do all too often is take beginners into advanced application way too fast. We forget what it is like to be overwhelmed with just getting through the day, and the limits that places on anyone for taking in new concepts.

I bring it up for the coach because I think the chances of the learner discovering it on their own are significantly lower (perhaps close to zero) if the coach is starting in the same place.

If you are getting pushback on the concept, it might be time to back off a bit and give your learner some space.

Steve Medland had a mini (5 minute) presentation at KataCon 2020 that briefly addressed some of his experience with this situation.

He pointed out that the default worksheet templates for Grasp the Current Condition and Establishing a Target Condition emphasize process timing and cadence pretty heavily.

And the alternative is a blank template:

Steve makes a good case that there ought to be something that provides a more general structure without removing all structure. I agree – as a coach, structure is one of the things you are bringing to the table. Any learner, at any level of expertise, is more deeply embroiled in the process itself than in the process of improving the process. Having some structure really helps.

The key is to adjust the structure to fit the situation.

With beginners, the concept of takt time can be distracting, even paralyzing. Even more so if we use alien jargon like “takt time.”*

Using a hybrid structure like Steve proposes can get the learner moving into process analysis without getting wrapped up on terminology, or struggling with something she sincerely believes has nothing resembling a cadence.

Then, when the opportunity arises, the coach can still gently, but persistently, ask “How often does this need to be done?” and “How do we know if we are ahead or behind?”

Often the resistance is less about knowing there is some kind of schedule, and more about just being overwhelmed by all of the chaos that prevents any kind of stability.

Steve and I agree that timing is important. And I agree that it is important enough that it might be best to hold off on introducing it as a concept so we don’t create resistance unnecessarily.

But please don’t completely throw away measuring time just because it is hard. In fact, the harder it is, the more important it likely is. When it is easy to determine takt time, we likely already have an idea how long things are taking. The less appropriate takt time seems, the more critical it becomes to dig deep into where the time is going.

*Believe it or not, Godwin’s Law can even be applied to “takt time” if you research your history enough.

## Heavy Equipment Overhaul: Flow at Takt in 1938!

This is a great contemporary film from 1938 describing the complete overhaul of a mainline 4-6-0 steam locomotive in the U.K.

What is interesting (to me) is:

• The overhaul involves stripping the locomotive down to individual parts. Each of the parts, in turn, flows through a process of inspection / repair or replacement, with a strict timing to ensure it is delivered back to re-assembly when required.
• There are 6 positions with a takt time of 10 hours 44 minutes. Everything is timed to this cadence.
• I can only speculate, but with that degree of rigor in the timing, they are going to be able to see a delay or problem very quickly, and get out in front of it before it causes a delay in the main-line work.
• The parts that come off are not necessarily the exact once that are put back on. Everything is flowing – there are multiple locomotives in overhaul.

More thoughts below the video.

(Here is a direct YouTube link for those who don’t get the embed in the email subscription: https://www.youtube.com/watch?v=ktHw1wR9XOU)

## Flow in Overhaul and Repair

This is a great working example of a process flow that proves difficult for some organizations: Overhaul and repair. “We don’t know what we will find, so there is no way we can sequence and index it on a timetable.”

I’ve seen a similar operation overhauling helicopters. The intended flow was exactly the same.

• Like the locomotive flow, they stripped everything down to the airframe. The various components had different flow paths for sheet metal, hydraulic components, power-train (engine / transmission), rotor components, electrical, avionics, and composite parts.
• The objective was to deliver “good as new” items on time back to the re-assembly process.

Here is where they ran into problems:

• If an item needed repair, then the repairs were done, and the item flowed back.
• But if an item could not be repaired (needed to be scrapped and replaced) it was tagged, and returned to the “customer” – the parts bin in main assembly. It arrived just like any other part except this one was tagged as unusable. It was up to the assembly supervisor to notice, and initiate ordering a new one.

## Who is your customer? What do they need?

The breakdown was that the repair line(s) saw themselves as providing a repair service. If it couldn’t be repaired, sorry.

What their customer needed was a good part to install on the helicopter. If they can create a good part by repairing the old one, great. But if it isn’t repairable, their customer still needs a good one and they need it on time.

## The Importance of Timing and Sequencing

In the locomotive video, they emphasize the precise timing and sequencing to make sure each part arrives in the proper sequence, when it is needed, where it is needed.

Even if it actually worked like they describe, I can be sure it didn’t work like that when they first started.

The timing and sequencing is a hypothesis. Each time they overhaul a locomotive, in fact each individual part flow, is an experiment to test that hypothesis. Over time, it is possible to dial things in very precisely.

Why? So you can quickly identify those truly anomalous conditions that demand your intervention.

## Normal vs Abnormal

Just because there are frequent issues does not negate the fact that most of the time things can probably flow pretty well. What we tend to do, however, is focus on the problem cases and give up on all of them. “What about this? What about that?” bringing up the legitimate issues and problems, causes us to lose sight of the fact that underneath it all there is a baseline pattern.

What is important is to define the point at which we need to intervene, and set up the process to detect that point. When we can clearly distinguish between routine work and true exceptions, and not try to treat everything as a special case.

## A Machine Productivity Search Question

Here is another test or quiz question that showed up in my search logs:

a machine tool is producing 90 pcs per day .using improved cutting tools,the output is raised to 120 pcs per day. what is the increase in productivity of the machine?

I guess the answer seems pretty obvious…

90 x 1.33, rounded a bit, is 120, which gives us a 33% productivity improvement, right?

# Not so fast…

(pun intended)

How fast does the machine need to run?

Is there demand for the additional 30 pieces per day, or are they just being put into inventory with the hope they will sell at some point in the future?

This is actually pretty common where cost accounting systems allocate overhead against production output rather than actual sales.

But what if you are only selling 90 pieces per day?

After three days you will have a day’s worth in inventory. You are running the machine more than you have to, adding wear and tear. You are consuming material to make parts you aren’t selling. At some point you are going to have to shut down the machine – idle it. What is your productivity then?

# What Problem Are You Trying to Solve?

It always comes down to this question. Is there a real-world, customer-impacting reason you need the additional output? If so, then yes, this is a valid countermeasure, similar to one I have overseen myself. If the machine is too slow, what do we have to do to run it faster (while maintaining quality and not breaking anything)?

But if the machine is fast enough, then why are you trying to make it run faster?

And what will happen if we do? Use real numbers. You don’t have revenue until a customer with real money (not transfer pricing) actually pays for your product. Pretending otherwise looks great on the balance sheet for a while, but the paper profits aren’t tangible, you can’t use that “money” to buy anything else, or distribute to share holders. In fact, it is just money you have spent not money you have earned.

# Machine Utilization at Home

At least here in the USA, a typical home washing machine will run a cycle in about 25 minutes. The dryer takes about 40 minutes to complete a cycle. If you wanted “maximum efficiency” from the washing machine, all you will get is a big pile of wet laundry. There is no point in running the washer any more often than once every 40 minutes. The dryer is pacing the system.

If I could modify the washing machine to run in 15 minutes instead of 25, how much more productivity do I have? The question is nonsense.

This example makes perfect sense to people. Then I often get arguments about how the factory floor is somehow different?

# It’s The System, not the Machine

Key Point: You can’t look at one machine in isolation and calculate how “efficient” or “productive” it is unless it is pacing your system. In this case, we don’t have enough information.

Now, I know this example was just a made up case. But I have seen well-meaning production people fall into this trap all of the time. You have to look at the system, not individual machines.

## More Cycle Time Questions from Search Results

A couple of more exam questions showed up in search results.

4. what can happen if the cycle time is much shorter than the takt time

The searcher didn’t even bother typing this one – just cut and paste the entire thing, including the question number.

Fortunately the answer is really easy:

If you keep this up, you’ll need another building.

OK, here’s one that is a little more subtle:

185 units in 14 hours is what cycle time

Hmmm. My first thought is to go back to Who is Grading the Questions?

think whoever wrote this question is looking for something like 14 hours  (840 minutes) / 185 units = 272 seconds / unit (more or less). But that’s an average, it isn’t a cycle time.

## The Average Rate of Output is not “Cycle Time.”

The average doesn’t give you any more information than the original question. This is simply a rate of output: 185 units in 14 hours. Cycle time is the measured time to complete one cycle.

There might be a single cycle that produces a batch of 185 units. Let’s say, for example, in a cure oven process that takes 14 hours to load, run and unload. Then the cycle time is 14 hours.

If the pieces are moving in a sequence of operations, but in a lot (which is different from a batch), where Operations 1 is completed on all 185, then Operation 2 is completed on all 185, etc. and maybe that all takes 14 hours? I have no idea what the cycle time(s) for the operations are. Most of the time the units are waiting in queue for the other 184 items to be done at each operation.

If we have true one-by-one flow then I have at least 185 cycle times. Hopefully they are all about the same, but likely that isn’t the case.

Are we measuring exit cycles (the cadence of output at the end of the process), or operator or machine cycle times?

By taking the average we are obscuring all of this information. Calling the average the “cycle time” just makes it worse because it gives people the impression that they are the same.

Cycle time is not calculated. It is measured. You cannot determine cycle time by applying mathematical operations on numbers. You have to go observe with a stopwatch.

And finally:

within each four hours worked, workers can take a total of 48 minutes in allowance (so allowance factor is based on workday). if the observed time is 6 minutes, and performance rating is 95%, what is the standard time (in minutes; give three significant digits after the decimal point)?

Even if I could understand the question, down to thousandths of a minute? Seriously?

</rant> <!– Until next time –!>

## Using Takt Time to Compute Labor Cost

How can I use takt time in computing labor cost?

Sometimes the searches that lead here give us interesting questions.

While simple on the surface, this question takes us in all kinds of interesting directions.

Actually the simplest answer is this: You can’t. Not from takt time alone.

## Takt Time

Takt time is an expression of your customer’s requirement, leveled over the time you are producing the product or service. It says nothing about your ability to meet that requirement, nor does it say anything about the people, space or equipment required to do it.

## Cycle Time

Cycle time comes in many flavors, but ultimately it tells you how much time – people time, equipment time, transportation time – is required for one unit of production.

Takt time and cycle time together can help you determine the required capacity to meet the customer’s demand, however they don’t give you the entire story.

In the simplest scenario, we have a leveled production line with nothing but manual operations (or the machine operations are trivially short compared to the takt time).

If I were to measure the time required for each person on the line to perform their work on one unit of the product or service and add them up, then I have the total work required. This should be close to the time it would take one person to do the job from beginning to end.

Let’s say it takes 360 minutes of work to assemble the product.

If the takt time says I need a unit of output every 36 minutes, then I can do some simple math.

How long do I have to complete the next unit?  36 minutes. (the takt time)

How long does it take to complete one full unit?  360 minutes (the total manual cycle time)

(How long does it take) / (How long do I have) = how many people you need

360 minutes of total cycle time / 36 minutes takt time = 10 people.

But this isn’t your labor cost because that assumes the work can be perfectly balanced, and everything goes perfectly smoothly. Show me a factory like that… anywhere. They don’t exist.

So you need a bit more.

## Planned Cycle Time (a.k.a. Operational Takt Time and “Actual Takt”)

How much more? That requires really understanding the sources of variation in your process. The more variation there is, the more extra people (and other stuff) you will need to absorb it.

If we don’t know, we can start (for experimental purposes) by planning to run the line about 15% faster than the takt time. Now we get a new calculation.

85% of the takt time = 0.85 x 36 minutes = ~31 minutes.  (I am rounding)

Now we re-calculate the people required with the new number:

360 minutes required / 31 minutes available = 11.6 people which rounds to 12 people.

Those two extra people are the cost of uncontrolled variation. You need them to ensure you actually complete the required number of units every day.

“But that cost is too high.”

## Getting to Cost

12 people is the result of math, simple division that any 3rd grader can do. If you don’t like the answer, there are two possible solutions.

1. Decide that 360 / 30 = something other than 11.6 (12). (or don’t do the math at all and just “decide” how many people are “appropriate” – perhaps based on some kind of load factor. This, in fact, is a pretty common approach. Unfortunately, it doesn’t work very well for some reason.
2. Work to improve your process and reduce the cycle time or the variation.

Some people suggest slowing down the process, but this doesn’t change your labor cost per unit. It only alters your output. It still requires 360 minutes of work to do one unit of assembly (plus the variation). Actually, unless you slow down by an increment of the cycle time, it will increase your labor cost per unit because you have to round up to get the people you actually need, and/or work overtime to make up the production shortfall that the variation is causing.

So, realistically, we have to look at option #2 above.

This becomes a challenge – a reason to work on improving the process.

## Really Getting to Cost

Challenge: We need to get this output with 10 people.

Now we have something we can work with. We can do some more simple math and determine a couple of levers we can pull.

We can reverse the equation and solve for the target cycle time:

10 people x 30 minute planned cycle time-per-unit = 300 minutes total cycle time.

Thus, if we can get the total cycle time down to 300 minutes from 360, then the math suggests we can do this with 10 people:

300 minutes required / 30 minutes planned cycle time = 10 people.

But maybe we can work on the variation as well. Remember, we added a 15% pad by reducing the customer takt time of 36 minutes to a planned cycle time (or operational takt time, same thing, different words) of 30 minutes. Question: What sources of instability can we reduce so we can use a planned cycle time of 33 minutes rather than 30?

Then (after we reduce the variation) we can slow down the process a bit, and we could get by with a smaller reduction in the total cycle time:

330 minutes required / 33 minutes planned cycle time = 10 people.

(See how this is different than just slowing it down? If you don’t do anything about the variation first, all you are doing is kicking in overtime or shorting production.)

So which way to go?

We don’t know.

First we need to really study the current process and understand why it takes 360 minutes, and where the variation is coming from. Likely some other alternatives will show themselves when we do that.

Then we can take that information, and establish an initial target condition, and get to work.

## Summarizing:

• You can’t use takt time alone to determine your labor cost. Your labor cost per unit is driven by the total manual cycle time and the process variation.
• With that information, you can determine the total labor you need on the line with the takt time.
• None of this should be considered an unalterable given. Rather, it should be a starting point for meeting the challenge.

And finally, if you just use this to reduce your total headcount in your operation, you will, at best, only see a fraction of the “savings” show up on your bottom line. You need to take a holistic approach and use these tools to grow your business rather than cut your costs. That is, in reality, the only way they actually reach anywhere near their potential.

## Another Homework Question

Another interesting homework question has shown up in the search terms. Let’s break it down:

23. if the slowest effective machine cycle time in a cell is 55 seconds and the total work content is 180 seconds, how many operator(s) should operate the cell so that labor utilization is at 100%?

I find this interesting on a couple of levels.

At a social level, the idea of cutting and pasting a homework question into Google hoping to find the answer is… interesting. Where is the thinking?

What are we teaching?

The question is asking “How many people do we need to run as fast as we can?” (as fast as the slowest machine). But how fast do they need to run? Maybe they only need a part every 95 seconds. If that is true, then I need fewer people, but I am going to run the slowest machine even slower.

In other words, “What is the takt time?” What does the customer need? How often must we provide it?

Then there is the “labor utilization” metric, with a target of 100%. Assuming the planned cycle time is actually 55 seconds (which it shouldn’t be!), we need 3.3 people in this work cell. (180 seconds of labor cycle time / 55 seconds planned cycle time: “How long does it take?” / “How long do you have?” = Minimum Required Capacity)

How about improvement? What do we need to do to get from 3.3 people to 3 people? We can solve for the labor cycle time.  55 seconds of planned cycle time * 3(people) = 165 seconds of total labor. So we need to get that 180 seconds down to a little less than 165 seconds.

Now we have a challenge. We need to save a bit over 15 seconds of cycle time. That might seem daunting. But we don’t have enough information (the current condition) to know where to begin. Then we can establish the next target condition and get started making things better.

These types of questions bother me because they imply all of these things are fixed, and they imply we run “as fast as we can” rather than “as fast as we must.”

Edit: Today I saw two more searches for:

total work content divided by slowest machine cycle time

so it looks like at least two others are working on the same assignment.  🙂

Thoughts?

## Boeing Moving Line

Boeing’s “PTQ” (Put Together Quickly) videos show a time lapse of an airliner in production. They have been producing the for years – certainly since I was working there.

This one, though, shows something a little special.

When I first started working there, the idea of a line stop was unthinkable. The plane moved on time, period. Any unfinished work “traveled” with the plane, along with the associated out-of-sequence tasks and rework involved.

The fact that the 737 is now built on a continuously moving assembly line in Renton is fairly well known.

But what struck me in this PTQ video is that one of the things highlighted in it is a line stop. It happens pretty quickly at about 1:57.

The video is also full of rich visual controls to allow the team to compare the actual flow vs. the intended flow. See many many you can spot.

## Taktzeit

Now and again someone wonders out loud why, in this lexicon of Japanese terms, we have the word “takt.”

I had always passed along what I had heard – that the word was German, short for taktzeit and used in their factories to represent the pace of production. During WWII, the Germans had helped the Japanese set up more efficient production lines, and the word migrated into Japanese usage.

All of this had been anecdotal.

But today I ran across Alan Hamby’s phenomenally in-depth reference site on the German WWII Tiger Tank. Alan has extensive detail on the Henschel Tiger Tank Factory, and in some of the photos are signs indicating the number of the “takt” or production position.

But don’t stop here. Take a look at Alan’s site, and look at how this factory is set up and operated. This plant was set up to produce a Tiger I tank every six hours, and built a total of 1375 of them between 1942 and early 1945. Yes, there is a lot of waste, but bluntly, I have seen 21st century factories making products of similar size and complexity that are far worse than this.

The idea of pacing and balancing production is not new. By the time these photos were taken around 1943, the concept had been proven for over 20 years. Yet when I visit factories today this is a seemingly novel concept. I always wonder why today’s operations managers are not insisting on at least the efficiencies that were achieved by 1935.

Thanks to Alan for his kind permission to bootstrap from his research and use these rare photos here.

Just to be clear, though, having a pace for production does not make a line “lean.” Far from it. But it is a foundational element. It may not be sufficient, but it is (in nearly all cases) necessary. What makes it foundational element for improvement, however, is not so much the pacing and balancing aspect. Rather, the concept of takt time can be used as a way to structure improvement goals and targets in a way that is meaningful to the people doing the work.

We talk a lot (all to much, in my view) about metrics, but tend to think of the things management is interested in – like labor productivity. But the way you get labor productivity is to focus on the takt time, the total cycle time, and the stability of that cycle time. Those are the things that determine how much gets done by how many people. You can measure “labor productivity” all you want, but you can’t change it unless you get down another couple of levels. Fortunately (for us) Reichsminister Speer didn’t figure that out.

By the way, just to put things into perspective:

In 1943, Boeing Plant 2 was producing one B-17 bomber an hour, sixteen planes a day, six days a week. They did it by using a paced assembly line and continuously working to simply and improve the flow.

## Takt Time – Cycle Time

There has been an interesting discussion thread on “Kaizen (Continuous Improvement) Experts” group on LinkedIn over the last few weeks on the differences between takt time and cycle time.

This is one of the fundamentals I’d have thought was well understood out there, along with some nuances, but I was quite surprised by the number (and “quality”) of misconceptions posted by people with “lean” and “Sigma” in their job titles.

I see two fundamental sources of confusion, and I would like to clarify each here.

#### “Cycle Time” has multiple definitions.

“Cycle time” can mean the total elapsed time between when a customer places an order and when he receives it. This definition can be used externally, or with internal customers. This definition actually pre-dates most of the English publications about the Toyota Production System.

It can also express the dock-to-dock flow time of the entire process, or some other linear segment of the flow. The value stream mapping in Learning to See calls this “production lead time” but some people call the same thing “cycle time.”

In early publications about the TPS, such as Suzaki’s New Manufacturing Challange and Hirano’s JIT Implementation Manual, the term “cycle time” is used to represent what, today, we call “takt time.” Just to confuse things more, “cycle time” is also used to represent the actual work cycle which may, or may not, be balanced to the takt time.

We also have machine cycle time, which is the start-to-start time of a machine and is used to balance to a manual work cycle and, in conjunction with the batch size,  is a measure of its theoretical capacity.

“Cycle time” is used to express the total manual work involved in a process, or part of a process.

And, of course, “cycle time” is used to express the work cycle of a single person, not including end-of-cycle wait time.

None of these definitions is wrong. The source of confusion is when the users have not first been clear on their context. Therefore, it is critically important to establish context when you are talking. Adjectives like “operator cycle time” help. But the main thing is to be conscious that this can be a major source of confusion until you are certain you and the other person are on the same wavelength.

#### Takt time is often over simplified.

The classic calculation for takt time is:

Available Minutes for Production / Required Units of Production = Takt Time

This is exactly right. But people tend to get wrapped up around what constitutes “available time.” The “pure” definition is usually to take the total shift time(s) and subtract breaks, meetings, and other administrative non-working time. Nobody ever has a problem with this. (Maybe because that is the way Shingijutsu teaches it, and people tend to accept what Shingijutsu says at face value.)

So let’s review and example of what we have really done here. For the sake of a simple discussion, let’s assume a single 8 hour shift on a 5 day work week. There is a 1/2 hour unpaid lunch break in the middle of the day, so the workers are actually in the plant “at work” for 8 1/2 hours. (this is typical in the USA, if you are in another country, it might be different for you)

8 hours x 60 minutes = 480 total minutes

But there is a 10 minute start-up process in the morning, two 10 minute breaks during the day, and 15 minutes shut-down and clean up at the end of the shift for a total of 45 minutes. This time is not production time, so it is subtracted from “available minutes”:

480 – 45 = 435

A very common mistake at this point would be to subtract the 30 minute lunch break. But notice that we did not include that time to start with. Subtracting it again would count it twice.

So when determining takt time, we would use 435 minutes as the baseline. If  leveled customer demand was 50 units / day, then the takt time would be:

435 available minutes / 50 required units of production = 8.7 minutes (or 522 seconds)

Note that you can just as easily do this for a week, rather than a day.

435 minutes x 5 days = 2175 total available minutes

2175 available minutes / 250 required units of production still equals 8.7 minutes (or 522 seconds)

All of this is very basic stuff, and I would get few arguments up to this point, so why did I go through it?

Because if you were to run this factory at a 522 second takt time, you will come up short of your production targets. You will have to work overtime to make up the difference, or simply choose not to make it up.

Why? Because there are always problems, and problems disrupt production. Those disruptions come at the expense of the 435 minutes, and you end up with less production time than you calculated.

Then there is the fact that the plant manager called an all-hands safety meeting on Thursday. That pulled 30 minutes out of your production time. Almost four units of production lost there.

I could go on with a myriad of examples gathered from real production floors, but you get the idea.

Here is what is even worse, though.

When are you going to work on improvements?

If you expect operators to do their daily machine checks, when do you expect that to happen?

Do you truly expect your team members to “stop the line” when there is a problem?

All of these things take time away from production.

The consequence is that the shop floor leadership – the ones who have to deal with the consequences of disrupted production – will look at takt time as a nice theory, or a way to express a quota, but on a minute-by-minute level, it is pretty useless for actually pacing production.

All because it was oversimplified.

If you expect people to do something other than produce all day, you have to give them time to do it.

Let’s get back to the fundamental purpose of takt time and then see what makes sense.

#### The Purpose of Takt Time

Here is some heresy: Running to takt time is wholly unnecessary. Many factories operate just fine without even knowing what it is.

What those factories lose, however, is a fine-grained sense of how things are going minute by minute. Truthfully, if they have another way to immediately see disruptions, act to clear them, followed by solving the underlying problem then they are as “lean” as anyone. So here is the second heresy: You don’t NEED takt time to “be lean.”

What you need is some way to determine the minimum resource necessary to get the job done (JIT), and a way to continuously compare what is actually happening vs. what should be happening, and then a process to immediately act on any difference (jidoka). This is what makes “lean” happen.

Takt time is just a tool for doing this. It is, however, a very effective tool. It is so effective, in fact, that it is largely considered a necessary fundamental. Honestly, in day to day conversation, that is how I look at it. I made the above statements to get you to think outside the mantras for a minute.

What is takt time, really?

Takt time is an expression of your customer demand normalized and leveled over the time you choose to produce. It is not, and never has been, a pure customer demand signal. Customers do not order the same quantity every day. They do not stop ordering during your breaks, or when your shift is over. What takt time does, however, is make customer demand appear level across your working day.

This has several benefits.

First, is it makes capacity calculations really easy through a complex flow. You can easily determine what each and every process must be capable of. You can determine the necessary speeds of machines and other capital equipment. You determine minimum batch sizes when there are changeovers involved. You can look at any process and quickly determine the optimum number of people required to make it work, plus see opportunities where a little bit of kaizen will make a big difference in productivity.

More importantly, though, takt time gives your team members a way to know exactly what “success” looks like for each and every unit of production. (assuming you give them a way to compare every work cycle against the takt time – you do that, don’t you?)

This gives your team members the ability to let you know immediately if something is threatening required output. Put another way, it gives your entire team the ability to see quickly spot problems and respond to them before little issues accumulate into working on Saturday.

The key point here is that to get the benefit, you have to have a takt time that actually paces production. It has to be real, tangible, and practically applied on the shop floor. Otherwise it is just an abstract, theoretical number.

This means holding back “available time” for various planned (and unplanned) events where production would be stopped.

Further, in a complex flow, there may be local takt times – for example, a process that feeds more than one main line is going to be running to the aggregated demand, and so its takt will be faster than either of them. Likewise, a feeder line that builds up a part or option that is not used on every unit is going to be running slower.

And finally if disruptions do cause shortfalls to the required output, you have to make it up sometime. If you are constrained from running overtime (and many operations are for various reasons), then your only alternative is to build a slight over speed into your takt time calculation. The nuances of this are the topic of a much longer essay, but the basics are this:

– If everything goes well, you will finish early. Stop and use the time for organized improvement of either process or developing people. Continuing to produce is overproduction, and just means you run out of work sooner if you have a good day tomorrow.

– If there are issues, the use the buffer time for its intended purpose.

– If there are more issues than buffer time, there is an operational decision to make. Have a policy in place for this. The simplest is “hope for a better day tomorrow” and use tomorrow’s buffer time to close the gap. If this isn’t enough, then a management decision about overtime or some other remedy is required.

What about just allowing production to fall short? Well.. if this is OK, then you were running faster than customer demand already. So pull that “extra” out of your schedule, stop overproducing (which injects its own disruptions into things), and deal with what just actually have to accomplish. Stop inflating the numbers because they hide the problems, the problems accumulate, and you end up having to inflate even more.

Gee, all of this seems complicated.

Yeah, it can be. But that complexity is usually the result of having an ad-hoc culture that makes up the reactions as you go along rather than a comprehensive thought-out systems-level approach. The key is to work through the “what if…” for what you are doing and thinking about doing, how the pieces actually interconnect and interact, and have a plan.

That plan is the first part of Plan-Do-Check-Act.

Then, as the real world intrudes, you can test your thinking against reality and get better and better rather than just being glad you survived another day.

And that, is the whole point of knowing your cycle times and takt times.