Hertz to Cycles per Hour for Line Throughput Rates

Turning a Station Cycle Time Into Parts per Hour

Every manufacturing conversation eventually reduces to two numbers: how long one station takes to release a part, and how many parts an hour that adds up to. The first is what a stopwatch on the floor gives you, in seconds. The second is what production planning, costing and the customer's delivery schedule are all written in. Cycles per hour is simply the second number, arrived at from the first.

Treating a station as a frequency makes the arithmetic honest. A cycle time is the reciprocal of a rate, so shaving two seconds off a fast station buys far more throughput than shaving two seconds off a slow one, and a rate stated per hour makes that difference visible instead of leaving it buried in a seconds column.

Conversion factor: multiply by 3 600 — 1 Hz = 3 600 cph. A press station releasing a part every 20 seconds runs at 1 ÷ 20 = 0.05 Hz, which is 0.05 × 3 600 = 180 parts per hour if it never stops.

Three numbers that are easy to confuse

Cycle time is observed, takt time is imposed

Cycle time is a property of the process: load, clamp, run, unload, measured with a watch. Takt time comes from outside the factory altogether — available production time divided by customer demand. One is what the equipment does, the other is what the order book requires, and the whole point of comparing them is to see whether the first fits inside the second.

The slowest station is the line

Parts leave a sequential line at the pace of its worst station, no matter how quick the others are. Upstream machines finish early and block, downstream machines wait and starve, and the aggregate rate never exceeds the bottleneck's own. Converting each station's cycle time to an hourly figure puts the constraint in plain sight, along with how much idle capacity sits either side of it.

Theoretical output is not shipped output

Between the ideal rate and the count on the pallet sit changeovers, breakdowns, waiting for material, minor stops, reduced running speed and scrap. Overall equipment effectiveness rolls those into one multiplier, and even a well-run cell rarely clears the mid-eighties, so a station calculated at 180 an hour realistically delivers about 153.

Sizing a Line Rate From the Order Book Backwards

Enter the measured cycle as a frequency

Divide one by the observed cycle time in seconds and type the result into the left field. A 45-second station is 0.02222222 Hz; timings recorded with a decimal comma go in exactly as they were written on the time-study sheet.

Run every station and keep the smallest answer

Work along the sequence one station at a time. The lowest hourly figure is the line rate, and the difference between it and the next slowest tells you how much has to be moved off the constraint before anything upstream matters.

Enter the effective cycle instead of the ideal one

Divide the measured cycle by the effectiveness you actually achieve before converting, and the answer stops being optimistic. Twenty seconds at 85 percent becomes an effective 23.53 seconds, which converts to 153 an hour — the number worth putting in a delivery commitment.

Reverse it when the quantity comes first

Planning usually arrives as a quantity, not a cycle. Press the swap arrows, or type into the right-hand field, and a demand of 80 parts an hour reads back as 0.02222222 Hz — one part every 45 seconds, which is the target each station has to be designed against.

Station Cycle Times and the Throughput They Support

Cycle times spanning fast automated stations through to long manual assembly operations, with the ideal hourly count and a realistic figure once an overall effectiveness of 85 percent is applied.

Station cycle time	Rate (Hz)	Theoretical parts per hour	At 85% effectiveness
4 s — high-speed pick and place	0.25	900	765
6 s — automated inspection gate	0.16666667	600	510
10 s — robot load and unload	0.1	360	306
20 s — press or injection station	0.05	180	153
30 s — manual assembly operation	0.03333333	120	102
45 s — multi-part subassembly	0.02222222	80	68
60 s — test and pack cell	0.01666667	60	51
90 s — long machining cycle	0.01111111	40	34

Time-study figures go in unedited

Long decimals from a stopwatch app and readings written with a comma both convert as typed, so a station timed at 22,4 seconds needs no tidying up first.

Per-minute rates for fast cells

The unit lists reach beyond the hourly count, so a four-second station can be read as fifteen parts a minute, which is the language conveyor and buffer sizing tends to use.

Clean digits for a balance sheet

Copying returns 180 rather than a string carrying a unit, which is what a line-balancing spreadsheet or a capacity model wants dropped into a cell.

Throughput Questions From the Production Floor

Why must takt time never be treated as a cycle time?

Because they come from opposite directions and mixing them up produces a line that cannot ship. Takt is demand-side: available production time divided by the quantity customers want. Two shifts of 450 usable minutes against 1 200 units per day gives 54 000 seconds over 1 200 pieces, a takt of 45 seconds, or 80 an hour. Cycle time is supply-side, and it must land comfortably below takt, because a station running exactly at takt has no margin at all for a jam, a tool change or a slow part. Design the process at some fraction of takt — often around 85 percent of it — and the buffer against everyday disruption is built in rather than hoped for.

How do I find which station is actually holding the line back?

Convert every station's cycle time to an hourly rate and the lowest number is your constraint, but confirm it by walking the line rather than trusting the sheet. The bottleneck is where work-in-progress piles up in front and the machine after it stands idle, and that visual signature is far more reliable than a time study taken on a good day. Once identified, everything you do to faster stations is wasted: speeding up a 10-second station on a line whose worst station takes 45 seconds changes the output by nothing at all. Move work off the constraint onto neighbours with spare time, shorten its changeover, or protect it with a small buffer so it never starves. The constraint often relocates afterwards, which means the exercise repeats.

How much of the theoretical rate does effectiveness take away?

More than most plans allow for, because the losses multiply rather than add. Availability accounts for the time the equipment was scheduled but not running — breakdowns, changeovers, waiting for material or an operator. Performance covers running slower than the ideal cycle plus the micro-stops nobody logs, the ten-second clearances that happen twenty times a shift. Quality removes whatever has to be scrapped or reworked. Multiply three plausible-looking figures such as 0.90, 0.95 and 0.99 and you are already down to 0.846, so a station calculated at 120 an hour delivers 102. Anything near 85 percent overall is a genuinely well-run operation; plenty of real lines sit closer to 60, which halves the theoretical figure once you include the shifts that never started on time.

How do I plan a line rate backwards from a daily quantity?

Start with time that genuinely exists. Take the shift length, subtract breaks, planned meetings, startup checks and cleandown, and you are left with the minutes the line can really produce in. Divide the required quantity by those hours to get the parts per hour you must average, then inflate it by your expected effectiveness before turning it into a cycle time: needing 80 good parts an hour at 78 percent means designing for about 103 an hour, a cycle of roughly 35 seconds rather than 45. Only then decide how many stations the work has to be split across, dividing total task content by the target cycle and rounding up. Skipping the effectiveness step is the classic way to commission a line that hits its numbers on the acceptance run and misses them every week afterwards.

Does a 20-second cycle mean a part is finished every 20 seconds?

One part leaves the line every 20 seconds, yes, but no individual part passes through in 20 seconds — those are different quantities and confusing them wrecks delivery promises. Throughput is how often something comes out of the end; lead time is how long one piece spends inside from start to finish. They are tied together by the amount of work-in-progress: a line producing 180 parts an hour with a dozen pieces somewhere in the sequence is holding each of them for roughly four minutes. Add buffers between stations to protect the bottleneck and throughput may improve slightly while lead time grows considerably, which is exactly the trade-off behind keeping work-in-progress deliberately small.