Today I’m opening a discussion about of 2 terms that come from the Lean lexicon: TAKT TIME, and ACTUAL TAKT TIME. Remember, “Lean” is the occidental word defining what it is basically the Toyota Production System, implemented in Japan in the 60’s, and that the rest of the world tried to embrace 20 years too late, and that trying to make it very, very short, it is based in the identification of the production wastes -jp. muda– and the processes to eliminate them. But we will talk about these terms here because they are very related to the production costs.

So, Toyota came with the term TAKT TIME, meaning the rhythm at which you need to produce to meet customer demand. Hence the term “takt”, think about a metronome “…tac, tac, tac…”, at every “tac” your customer wants one part. In other words, it is the theoretical period of time (seconds, minutes, hours…) required to produce what you need to produce. Not more, not less. Takt Time (from now on TT) is very easy to calculate and to understand, but it is often misinterpreted or incorrectly used.

Always better with one example: Our customer needs 100,000 pcs/yr, and your plant runs 48 weeks/year, 1 shift/day, 8 hours/shift. But every Friday the shift is shorter due to cleaning and self-maintenance, running only 6 hours.

Available production time:

  • Per week = (4 x 8) + 6 = 38 hrs/wk in 1 shift
  • Per year = 38 x 48 = 1,824 hours/year, or 6,566,400 seconds in one shift

Therefore, if we plan to run only 1 shift, then  TT = 6,566,400 sec / 100,000 pcs = 65.66 sec/pc

So, to make our customer demands, our line should be making 1 part every 65.66 seconds. If your line is design to make only this part and it’s producing faster that TT, you’ll either create inventory, or you’ll need to stop the line. Both alternatives are a waste in the system that cost money and don’t produce any profit.

What if your company runs 3 shifts?  Then my new TT = 196.99 sec/pc

But life is not perfect, and even less in our manufacturing lines, and we must take into account the inefficiencies and stops, like equipment breakdown, tool changes, bad parts, sample and trial runs, change-over time for shared equipment, etc. and therefore the cycle time planned in your lines must be shorter than the TT. How much shorter? It will depend on how well your company manages those inefficiencies. At the end, your line will be built and programmed to make parts at a certain cycle time (“engineered cycle time” or “planned cycle time”), and this cycle time is what is called ACTUAL TAKT TIME (probably a bad translation from Japanese, not very fortunate, and the reason for confusing with TT). This is, the pure or engineered cycle time of your lines.

In the example before

If you already have the equipment running at 100 seconds/pc (your ATT), how many shifts do you need to run?

100 / 65.66 = 1.52 shifts.

If you decide to run 2 full shifts, it means you are allowed to have 1.52/ 2.00 = 76% efficiency. If you can’t make your production in 2 shifts, it means you have more wastes than you thought of. These cost you money and give no profit, they are non-value added seconds that must be removed. Better to run at 85% OEE and perform part of the tasks that were initially thought for Fridays, and perhaps you can remove one full shift at the end of the week!

    So you see, for the same customer demand, the TT changes depending on the available time, or the available time is adapted to meet your TT.

A correct design of your line will make it more cost effective.  If you already have the equipment, plan your production time to meet your customer demand at your desired efficiency, do not run extra shifts “just in case” (waste of resources and cost). On the other hand, if you need to build a new line, once you have calculated what your ATT should be, design these lines to run at this ATT (another Lean principle is, make the maximum utilization of your equipment, so try to run as much time as possible, and design this ATT accordingly)

The TT can also be used to calculate the appropriate number of operators and machines needed to run a line for a specific product. ATT / TT gives the % of time these machines and their operators need to run to produce this product. If ATT=40 sec and TT=50 sec, it means that this product is consuming 80% of this manufacturing process, the other 20% needs to be used in another product and can’t be charged to the first one.

Of course, this is a simplification, and easier to apply to dedicated equipment ((please see the blog about dedicated or shared equipment), but it all starts here. Your product variation, the share of products in your lines, the change-over time, etc, all this needs to be considered in your TT/ATT, and therefore, if you already have the equipment, you adapt your running time accordingly (to optimize your labor rate), and if you need to build a new line, you design it according to your calculated ATT.

As cost engineers, this might seem like a conflict between BEST PART COST and BEST LEAN PRACTICE, like, “If I run faster, cost per piece is reduced”. Well, here we cost engineers have something to say. First thing, is the line dedicated? If the line or machine is custom-made, DEDICATED to a product, and in-line with the part process, then definitively this line / machine should be designed to run at the desired ATT. If you design it to run faster, it will need to be more complex and expensive (more $/hr), and you’ll have to stop it when production needs are met (you will not be complying with the Lean principle to make the maximum utilization of equipment), or you can fall into the trap of running at lower efficiency, this is, at higher cycle time than what the machine was built for (your limit will be the TT!), thus spending the same time per part as you would have with a simpler equipment, but of course now at a higher cost.

If your TT is 60 seconds and you expect an efficiency of 85%, then your ATT should be 70.6 seconds; a line with ATT=60 sec will run with less labor and less stressed than a line designed to run, for example, at 40 seconds. On the other hand, if the line will be SHARED, then of course you want to make as many different parts as possible, and the ATT for each different part must have been calculated to optimize the line (or lines) usage, and of course to meet their TT so to comply with your customers demand.

For example, imagine a new plastic part required at a volume of 500,000 pcs/yr. We have a plastic injection machine with enough tonnage to make up to 10 pcs/shot (10-cavity mold). To facilitate the example, let’s consider that the plastic injection time is so fast that it takes the same time to make 1 part in a 1-cavity mold than 10 parts in a 10-cavity mold, and let’s say it is 30 seconds/cycle. Running at 3 x 8h shifts during 240 days/yr (we want to maximize the machine), and at 80% efficiency, TT= 33.2 sec. This means that in theory, you could run the machine with a 1-cavity tool and be almost dedicated to this part (30/33.2=90% utilization). But the manufacturing cost of running this part at 1 part/cycle is 10 times more expensive than putting a 10-cavity mold in the same machine!  This machine should run at 10 pcs/cycle (ATT/TT = 9% occupation) and use the 91% of remaining capacity in other parts. This will meet the best cost and best utilization of the machine, so there is no contradiction at all between Lean and Best Cost.

To summarize:

TT is a factor exclusively of your customer demand and your available time

ATT is the engineered cycle time of the line or machine, and it is a factor of TT and other variables

I hope this blog is of your interest, and as always, if you have any comments or questions, please write to lhernandez@societyofcostengineers.com. Feedback is always welcome (continuous improvement, you know ;)).

Leo Hernandez

Leave a Reply

Your email address will not be published. Required fields are marked *