Takt time vs. cycle time

In one line

Takt time is the pace at which a part has to come out to meet what the customer asks for; cycle time is the pace at which your process actually produces it — and the gap between those two numbers, not the value of each one on its own, is the first thing to look at before touching anything else on a line.

What it is

Takt time is calculated with a single formula:

Takt time = Available production time / Customer demand in that period

"Takt" comes from German and means beat or musical pulse — the original idea is that of a conductor setting the pace at which the score has to move forward. Applied to a plant, takt time does not measure anything: it is a number that is defined from two data points external to the process (how much time is available and how much the customer asks for), and it tells the process the pace it has to sustain so it neither falls behind nor overshoots. If demand goes up or down, takt changes right away, even though nobody has touched a single machine or a single work method.

Cycle time is the opposite in origin: it is a figure that is measured, not calculated from demand. It is the real time the process (a machine, an operator, a whole cell) takes to produce one unit, taken with a stopwatch on the process that is actually running today. It knows nothing about how much the customer asks for — it only reflects what the process, with its current method, layout and people, is able to sustain.

The comparison between the two is what matters: takt says what is needed, cycle says what you have. A process is healthy when those two numbers are aligned.

What it is for

It answers a very specific question before you invest time in any other improvement: as it stands today, can my process sustain the pace the customer asks for? Based on that answer, it is used to:

• Decide how many operators or machines a line or cell needs, distributing the total work among stations whose individual cycle does not exceed takt (line balancing).
• Detect the real bottleneck of a line with several stations: it is not the slowest one "in general", it is the only one whose cycle exceeds takt.
• Prioritize where it is worth investing method improvement effort: it only makes sense to shorten the cycle of a station that is above takt today; shortening one that is already below takt adds nothing to the line's output.
• Review staffing every time demand changes (high season, a new customer, a drop in orders), because takt moves with it even if the process stays the same.

How to apply it

1. Calculate the available time for the period. This is the shift time minus the real planned stoppages: lunch, breaks, scheduled maintenance, changeovers already planned. A poorly calculated available time shifts takt without anyone noticing.
2. Take the customer demand for that same period, in the same time unit as the available time (if the available time is daily, the demand also has to be daily).
3. Calculate takt time by dividing the available time by the demand.
4. Measure the real cycle time of each station or of the whole operation, with a stopwatch, averaging several cycles so that an atypical cycle (an interruption, a one-off adjustment) does not distort the number. This pillar will later have an article dedicated specifically to the stopwatch time-study technique.
5. Compare cycle against takt, station by station (not just on the average of the whole line):
   • If the cycle is greater than takt: that station is slower than it needs to be — demand will not be met, and you have to reduce that cycle (improve the method, redistribute tasks, add a shift) or add capacity (another machine, another person).
   • If the cycle is less than takt: there is slack, but be careful with the "we're doing fine" reading. That spare time is capacity that can be reassigned — and if instead the process keeps making parts at its natural pace without pausing, that slack turns into overproduction: parts made before the customer asks for them, which pile up as work-in-process inventory.
   • If the cycle and takt are practically equal, the station is balanced with demand.
6. In a line with several stations, the cycle of the whole line is set by the slowest station, not by the average of all of them — a chain is only as fast as its slowest link.

Overproduction is not a minor detail: it is the first of the seven mudas (waste) of the Toyota Production System documented by this source, and it is considered the most serious of the seven because it generates, in a chain, another waste — excess inventory. Producing faster than the customer asks, without pausing the process, is not "being ahead": it is making something nobody has asked for yet, with the capital, the space and the risk of obsolescence that this implies.

Real example

The source book has two consecutive examples, within the same chapter on layout and balancing of work cells, that together cover exactly the comparison in this article (I cite them with their original numbers).

Part 1 — calculating takt: a factory schedules a production of 600 parts per day, in 8-hour shifts (480 minutes).

• Available time = 8 h × 60 min = 480 minutes.
• Demand = 600 units/day.
• Takt time = 480 / 600 = 0.8 minutes = 48 seconds per part.

That is the target: if the line does not deliver one part every 48 seconds, it does not reach the 600 for the day.

Part 2 — real comparison against the measured cycle: a few pages later, the same text uses a cell of four operators to show what happens when the cycle moves away from takt. In that example the line's takt is 50 seconds, and the measured cycle times of each operator are 28, 52, 32 and 38 seconds:

Operator	Measured cycle	Line takt	What happens?
Operator 1	28 sec	50 sec	Shorter cycle: 22 sec of slack against takt
Operator 2	52 sec	50 sec	Longer cycle: it is the bottleneck of the cell
Operator 3	32 sec	50 sec	Shorter cycle: 18 sec of slack against takt
Operator 4	38 sec	50 sec	Shorter cycle: 12 sec of slack against takt

Even though three of the four operators are below takt, the whole cell cannot deliver one part every 50 seconds: operator 2, with a 52-second cycle, is the one who ends up setting the real output pace of the entire cell. The slack of the other three compensates for nothing — each one works at their own station, and the part does not move faster than the slowest station.

The conclusion the book itself draws from this example is the practical application of balancing: by reassigning the tasks among the four operators (so that none is above the 50-second takt), the same cell can be run with just three people, without losing the pace needed to meet demand.

Template to use

A simple table, station by station, is enough to diagnose any line or cell:

Data	How to get it
Available time for the period	Shift length minus real planned stoppages (lunch, breaks, scheduled maintenance)
Customer demand in that period	Sales or planning data, in the same time unit as the available time
Takt time	Available time ÷ demand
Cycle time per station	Stopwatch, average of several cycles per station or operator
Diagnosis per station	Cycle vs. takt: greater (bottleneck), less (slack/risk of overproducing) or equal (balanced)
Line bottleneck	The station with the highest cycle of all — that is the real cycle of the whole line

So you don't have to build it by hand, there is a takt time and capacity per workstation spreadsheet in Excel with the formulas already in place: it calculates takt, compares each station's cycle against it and flags the bottleneck. The values it comes with are examples and must be deleted.

Benefits

• It gives an objective number (takt) to measure against whether the process can meet demand, instead of the feeling that "the line is doing well" or "doing badly".
• It is a concrete basis for deciding staffing and line balancing: how many stations and people are needed so that none of them exceeds takt.
• It prioritizes improvement effort: it is only worth attacking the cycle of the stations that exceed takt today, not the ones that already have slack.
• It is a living number: if demand changes, takt changes with it, which forces you to review staffing every time the customer's order changes.

Limitations to keep in mind

• Takt assumes a relatively stable demand within the chosen period; with very irregular day-to-day demand, a takt calculated on an average can hide specific peaks that still cause delays.
• Takt does not tell you how to reduce a cycle that exceeds it — for that you need other tools from this same pillar, such as method improvement, line balancing or layout redesign.
• Comparing only the average of the whole line against takt is misleading: as the real case shows, three stations below takt do not make up for a single one above it — you have to look station by station.
• An available time calculated without subtracting the real planned stoppages inflates takt artificially, and hides a pace problem that does exist in practice.

Try it with your own data

Enter your available time, your demand and the cycle time you measured with a stopwatch. The tool calculates takt time and compares it against your measured cycle:

In summary

Takt time and cycle time do not compete with each other and do not measure the same thing: takt is a target calculated from demand and available time (how long a part has to take), and cycle is a figure measured with a stopwatch on the real process (how long it takes today). Comparing them, station by station, is the first diagnosis of any line: it tells you whether you need to reduce the cycle or add capacity (cycle greater than takt), whether there is slack you must take care not to turn into overproduction (cycle less than takt), or whether the process is already balanced with what the customer asks for.