Ask a factory why it bought forty sewing machines for a new line instead of thirty-two, and the answer is sometimes just “that’s what fit the floor space.” That’s backward. Floor space should get planned around the machine count the product actually requires, not the other way around, and the machine count itself comes from a fairly straightforward calculation once you have three inputs: the garment’s SMV, the target daily output, and a realistic efficiency figure for the line.
What SMV Actually Measures
SMV, standard minute value, is the time a trained operator working at a standard, sustainable pace takes to complete one operation. It’s not a theoretical minimum and it’s not what a fast operator can hit on a good day. It’s a benchmark pace that can be sustained across a full shift without excessive fatigue or quality drop-off, and it’s typically established through time study or from a predetermined motion time system.
Every operation in a garment’s construction, from attaching a pocket to closing a side seam, has its own SMV. This SMV and efficiency variance reporting guide covers how this figure gets tracked against actual line performance once production starts. Add up every operation’s SMV and you get the garment’s total SMV, the total standard time needed to construct one complete unit.
Where SMV Figures Actually Come From
The entire calculation method in this article depends on having a reasonably accurate SMV for each operation, and it’s worth understanding where that figure should come from rather than treating it as a given input. Formal time study, observing a trained operator perform the operation across multiple cycles and averaging the results with an appropriate allowance for rest and personal time, is the traditional method and remains the most reliable for operations unique to a specific factory’s process. Predetermined motion time systems, standardized tables that assign time values to basic hand and body motions, allow an SMV to be built up analytically without needing an operator to physically perform the operation first, useful during product development before a line is actually producing the style. Some factories rely on SMV figures inherited from a similar previous style, adjusted by estimation, a faster but less reliable approach that introduces real risk of the machine requirement calculation being built on a figure that doesn’t actually reflect the specific operation as designed for the current style.
Whichever method a factory uses, the accuracy of every downstream calculation in this article, machine count, capacity planning, and cost estimation, depends entirely on the SMV figure being genuinely representative of the operation as it will actually run in production, not an optimistic or outdated estimate.
The Core Formula
Machine requirement for a single operation comes down to:
Machines needed = (Operation SMV × Target pieces per day) ÷ (Available minutes per day × Line efficiency)
Available minutes per day is simply shift length in minutes. Line efficiency accounts for the gap between theoretical output and what a real line actually achieves once you factor in operator skill variation, minor stoppages, and the natural learning curve on new styles. Efficiency in a well-run line commonly falls somewhere in the 50 to 65 percent range, though it varies by product complexity, operator experience, and how mature the line is on a given style.
A Worked Example
Take an operation with an SMV of 0.8 minutes, on a line targeting 600 pieces per day, running an 8-hour shift (480 minutes), at an assumed line efficiency of 55 percent.
Machines needed = (0.8 × 600) ÷ (480 × 0.55)
= 480 ÷ 264
= 1.8, rounded up to 2 machines
That rounding matters. A calculation that comes out to 1.8 machines doesn’t mean you can run at 90 percent utilization on a single machine and call it solved. It means two machines are needed to reliably hit target output, with some slack built in for the operation’s real-world variability. Rounding down to save on machine count is one of the most common ways factories create a bottleneck they didn’t intend.
Doing This Across the Whole Line
The example above covers one operation. A full garment has dozens, and each one needs this same calculation run against its own SMV. Once every operation’s machine requirement is calculated, the results reveal something beyond a shopping list: they show where the line’s natural bottleneck sits, since the operation with the highest machines-needed-to-SMV ratio relative to what’s actually planned is where output will get capped, regardless of how well every other operation is staffed and equipped.
This is also where product mix comes back into play, a point covered in more depth in the full machinery guide. An operation unique to one style, say a decorative topstitch that only appears on 20 percent of a factory’s output, doesn’t need a dedicated machine sized for full-line output. It needs capacity sized to the actual volume of styles that use it, which is exactly the kind of nuance a rough headcount-based estimate misses entirely.
Scaling the Calculation Across a Full Garment
The single-operation example above is the building block, but the real value of this method shows up once it’s applied across every operation in a garment. Take a simplified four-operation sequence for a basic woven shirt front: attach pocket (SMV 0.6), join shoulder seam (SMV
0.5), attach collar (SMV 0.9), and topstitch front placket (SMV 0.7),
all running on a line targeting 500 pieces a day, an 8-hour (480 minute) shift, at 55 percent line efficiency.
Running each operation through the same formula:
- Attach pocket: (0.6 × 500) ÷ (480 × 0.55) = 300 ÷ 264 = 1.14, rounded up to 2 machines
- Shoulder seam: (0.5 × 500) ÷ 264 = 250 ÷ 264 = 0.95, rounded up to 1 machine
- Attach collar: (0.9 × 500) ÷ 264 = 450 ÷ 264 = 1.70, rounded up to 2 machines
- Topstitch placket: (0.7 × 500) ÷ 264 = 350 ÷ 264 = 1.33, rounded up to 2 machines
Total machines needed across these four operations: 7. Notice that the collar operation, with the highest SMV, needs the same machine count as the pocket and placket operations despite its higher time demand, simply because of how the rounding works at this particular target output. This is exactly the kind of result a rough headcount estimate would miss, and it’s also where the line balancing consideration below becomes relevant, since the collar operation, running closest to its calculated capacity limit at 1.70 rounded to 2, is the operation most likely to become a bottleneck if daily target output increases even modestly without adding a third machine to that specific operation.
Line Balancing: Matching Machines to Actual Flow
Calculating machine requirement per operation is only half the job. The other half is line balancing, arranging operations so that no single station becomes a bottleneck that starves every station downstream of it. A perfectly calculated machine count for each individual operation can still produce an unbalanced line if the sequencing doesn’t account for how work actually flows from one operation to the next.
In practice, this means grouping operations with similar cycle times together where possible, and paying particular attention to any operation whose SMV runs well above the operations immediately before and after it in the sequence. That operation, even with the “correct” number of machines calculated against target output, often needs an extra machine or a skilled operator specifically to prevent it from becoming the line’s limiting factor once real-world variability enters the picture.
Theoretical Versus Actual Machine Requirement
The formula above gives a theoretical machine requirement, built on assumed efficiency and a clean SMV figure. Actual requirement on a running line almost always differs from this, usually upward, for a few consistent reasons: new styles run below assumed efficiency during their learning curve, some operators genuinely work below the SMV standard pace regardless of training, and unplanned absenteeism means a line rarely runs with every calculated machine fully staffed on any given day.
A practical approach is to use the theoretical calculation as the baseline machine count to purchase, then plan staffing with a small buffer, rather than trying to staff to the exact theoretical number and hoping real-world variability doesn’t create a shortfall. Machines are a one-time capital cost. Understaffing a bottleneck operation is a recurring output loss that compounds every single day the line runs short.
Revisiting the Calculation When Product Mix Changes
Machine requirement isn’t a one-time calculation done at factory setup and then forgotten. Every time product mix shifts meaningfully, whether that’s a new style entering rotation or an existing style’s volume changing, the machine requirement for affected operations shifts too. Factories that treat their original machine ratio as fixed, even as the product running through the line changes, are the same factories that end up with idle machines in one category and a chronic bottleneck in another, months after the original calculation stopped reflecting reality.
Revisiting this calculation whenever a major new style or volume change comes in, rather than only at initial setup, keeps the machine list aligned with what the line is actually being asked to produce.
Frequently Asked Questions
What happens if my machine requirement calculation gives a fraction like 1.8 machines?
Always round up, not down. A calculated result of 1.8 machines means the operation genuinely needs 2 machines to reliably hit target output, since the extra 0.2 reflects real capacity the operation needs even though it looks like a rounding formality.
Where do I get an accurate SMV if my factory has never done a formal time study?
Formal time study on a trained operator is the most reliable method, but predetermined motion time systems can build an SMV analytically before a style is even in production. Avoid relying purely on estimates carried over from a “similar” past style without verification.
Does line efficiency really need to be as low as 50 to 65 percent in the formula?
That range reflects real-world line performance once operator variation, minor stoppages, and learning curve are accounted for, not a pessimistic assumption. Using a theoretical 90 to 100 percent efficiency figure will understate real machine requirement and create a bottleneck.
How often should I recalculate machine requirement?
Any time product mix shifts meaningfully, a new style enters rotation, or an existing style’s volume changes significantly, since the original calculation stops reflecting reality the moment the underlying inputs change.