profile
Door
More view more view
Recording the apparel supply chain
through shop-floor numbers and data

The Couple Who Put Light Bulbs on Workers’ Fingers: From the Gilbreths’ Motion Study to Sewing SMV

Old-photo-style image of the cyclegraph experiment, with a bulb on a sewing operator's fingertip drawing loops of light at a sewing machine
The cyclegraph experiment, in which the path of a bulb on the fingertip is left as a line of light, moved to a sewing machine.

KEY POINTS

More than 100 years ago, Frank and Lillian Gilbreth, an American couple, fixed small light bulbs to workers’ fingers and photographed them, leaving the path of the hand as a line of light. The couple broke human motion into basic units and named them “therbligs”, their surname spelled backwards. This classification led to MTM in the 1940s and then to GSD, a standard time system for sewing, in 1978, and it became the root of the SMV that sewing factories use today. What the Gilbreths did with bulbs and film is now done by a tablet next to the sewing cell that reads joint positions.

A worker sits in a dark room. A small bulb is fixed to the fingertip, and in front stands a camera with its shutter open. When the worker does the job as usual, the photograph shows no person, only the path of the hand as a line of light. This is an experiment that Frank and Lillian Gilbreth actually ran in the United States more than 100 years ago.7

If you trace the standard time (SMV) used in sewing factories back far enough, you arrive at this dark room. This post follows how the couple turned human motion into numbers and how that method found its way to the sewing floor.

The question a bricklayer’s helper asked

Frank Gilbreth was born in Maine in 1868. Instead of going to college, he started work as a bricklayer’s helper in July 1885, at the age of 17.2

The first thing he noticed on site was that every bricklayer worked differently. Laying the same brick, each person used the hands in a different order and a different number of times, and worked at a different speed. This is where he formed the idea that there is “one best way” to do any job.2

What he changed was not the worker’s hands but the working environment. He built a scaffold whose height could be adjusted easily, so that the bricklayer always stood at the best height for the work. Bricks were placed in advance within reach of the hand. As a result, the motions needed to lay one brick went down from 18 to 4.5.1

Old-photo-style image of a bricklayer on an adjustable scaffold with bricks and mortar at waist height
An adjustable scaffold that keeps bricks and mortar at waist height.

The man with the stopwatch, the couple with the camera

In 1904 Frank married Lillian Moller. Lillian had studied literature at the University of California and later earned a doctorate in applied psychology from Brown University.3 A husband with a feel for the building site and a wife who studied the human mind began to look at the same problem.

In 1907 Frank met Frederick Taylor, famous for “scientific management”.1 Both men talked about efficiency, but they looked at different things. Taylor’s symbol was the stopwatch. He measured how many seconds a task took and focused on cutting that time. The Gilbreths looked at motion rather than time. They believed that if wasted motions were removed, time would fall as a result and workers would be less tired.2

The difference eventually turned into conflict. Frank broke with Taylor in 1914, and after Taylor died he stayed on bad terms with Taylor’s followers for a long time.12

Why he put a clock inside the frame

To study motion, you first have to capture it. Frank took a hand-cranked movie camera into factories. Between 1910 and 1924 he shot 250,000 feet of 35mm film. He filmed bricklaying, typing, soap packing, conveyor work and even surgery.56

There was one problem. A hand-cranked camera ran faster or slower depending on the person turning it. From the film alone, you could not tell how fast the worker really moved. So he filmed a clock in every scene. It was called a “microchronometer”. With the clock hand turning inside the frame, the time taken by each motion could be worked out regardless of the filming speed.5

Old-photo-style image of a worker filmed with a hand-cranked camera and a large clock placed beside the worker
A hand-cranked camera and a large clock filmed inside the same frame.

It is the same idea as showing a timecode in the corner of the screen when a sewing factory analyzes operations on video today.

Motion drawn with light

Film showed motion cut into moments, but it was hard to see at a glance which path the hand had taken. The couple’s answer was the bulb experiment from the opening of this post. Fix a small bulb to the worker’s finger, leave the shutter open for a long time, and the path of the hand stays in the photograph as a line of light. The couple called these photographs “cyclegraphs”.7

A tangled line meant the hand went back and forth over the same spot several times, and a short, smooth line meant a motion with nothing wasted. The couple went one step further. They bent wire into three-dimensional models of the light lines. Workers held these wire models and studied them to learn what a good motion looks like.7

These experiments did not stay in the laboratory. Among the surviving photographs is one of a worker operating a drill press at a factory called New England Butt Co. A clock appears on one side of that photograph too.7

Old-photo-style image of a motion model made by bending wire along the path of the hand
A motion model made by bending wire along the path of light. Workers looked at it to learn the shape of a good motion.

Giving motions a name: therbligs

As photographs and film piled up, the couple found that what the human hand does could be divided into a few basic motions, such as search, grasp, transport, position, use and release. They named these basic motions “therbligs”, from their surname Gilbreth spelled backwards.4

In a 1915 paper Frank first listed 16 basic motions, and today the list stands at 18.4 Applied to sewing work, they read like this.

TherbligExample in sewing work
SearchLooking for the next ply in a stack of cut pieces
SelectChoosing the right piece between front and back
GraspPicking up a cut piece
Transport loadedBringing the cut piece under the presser foot
PositionLining up the seam allowance and notches
UseStitching
InspectChecking the stitched line
Release loadStacking the finished piece
Transport emptyReaching for the next ply
HoldHolding the fabric with one hand and waiting
Pre-positionPlacing the next ply within reach
Unavoidable delayRethreading after a thread break
Avoidable delayStopping to wait for materials

13 of the 18 therbligs that are easy to apply to sewing work. The other five are Find, Assemble, Disassemble, Plan and Rest.4 The sewing examples are added to make the list easier to follow.

In this table, only “stitching” is time that actually adds value to the fabric. Everything else is time spent preparing for stitching or tidying up after it. These preparing and tidying motions are exactly what the Gilbreths set out to reduce.

Motion study outside the factory

The couple’s method did not stay in factories. Frank was the first to propose that in the operating room, a nurse hands the surgeon each instrument the surgeon calls for.2 During the First World War he served as an army major and created a training method that let soldiers take apart and reassemble a gun blindfolded.2

On 14 June 1924, Frank died of a heart attack at the railway station in Montclair, New Jersey, while talking to Lillian on the telephone. He was 55.2

Lillian carried on. While raising the children, she kept running the company she had founded with her husband. She took on work for the department store Macy’s, and in 1926 she did market research on sanitary pads for Johnson & Johnson.3 When engineering work dried up because she was a woman, she turned to motion in the home.7 In 1929 she presented the “Kitchen Practical”, a kitchen designed to cut walking. The foot-pedal trash can and the shelves inside the refrigerator door are also credited to her.3

In 1935 Lillian became a professor of management at Purdue University. She is called the first female engineering professor in the United States. In 1965 she became the first woman elected to the National Academy of Engineering, and the following year she was also the first woman to receive the Hoover Medal. She died in 1972, at the age of 93.3

Life in the Gilbreth household was recorded in Cheaper by the Dozen, a book written by two of their children in 1948. The book was made into a film three times, in 1950, 2003 and 2022.23

From therbligs to SMV

The Gilbreths divided motion, but other people attached time to each motion. In the 1940s, commissioned by Westinghouse in the United States, H. B. Maynard, J. G. Stegemerten and J. L. Schwab developed MTM (Methods-Time Measurement) and published it in 1948.810 It gives a predetermined time value to each basic motion, such as reach, move, grasp, position and release, so that a standard time can be calculated from the work method alone, without a stopwatch. The therblig “transport empty” is now called “reach”.4

MTM crossed over into sewing. In 1978 Peter Evans in the United Kingdom built GSD (General Sewing Data), a system made for sewing, on the basis of MTM. GSD uses the same time unit as MTM, the TMU.8 1 TMU is 36 milliseconds, and one hour is 100,000 TMU.10

The SMV (Standard Minute Value) that an IE in a sewing factory writes into the operation breakdown comes from standards built this way. It is the sum of the times attached to each motion, such as grasping the cut piece, positioning it, stitching and putting it down. A question that started on a building site has become the number that sets quotations and line targets in today’s sewing factories.

The cyclegraph 100 years later

For the Gilbreths, analyzing a single motion took a dark room, bulbs, a camera, film and the time to go through the film frame by frame. So for a long time motion study was something done on a special day with a few selected people. The same is true in sewing factories. An IE measures a few cycles with a stopwatch or on video to set the standard time, and that value is used as it is until the next order.

SIJE’s Monolog turns the same question into something asked every day. The tablet camera next to the sewing cell does not save the work as video. It extracts only the positions of the person’s joints as points. It reads about 75 points per frame: 33 body joints such as the shoulders, elbows and wrists, plus 42 finger joints across both hands.9 Instead of fixing a bulb to a fingertip, it places a point on every finger joint.

Old-photo-style image of a sewing operator's hands with a point on every finger joint
What today’s hand joint recognition might have looked like through a camera from 100 years ago.

Following how the points move over time shows when each piece of work started and ended, and within it, separates the time the needle spends stitching the fabric from the time spent picking up, positioning or waiting for the fabric.9 In terms of the table above, it separates “stitching” from the other therbligs in every cycle.

There is another difference. The Gilbreths’ films captured the workers’ faces and bodies as they were. Monolog deletes the original footage within 500ms and keeps only values extracted from the joint points.9 Just as the couple thought first about worker fatigue in their motion studies, we believe that putting a camera on the floor has to be designed together with rules that protect the workers.

Wrapping up

The question Frank Gilbreth asked as a bricklayer’s helper was “why does each person do the same job differently?” That question passed through bulbs, film and therbligs, became MTM and GSD, and led to the SMV of the sewing factory. SIJE’s work of building operation know-how into measured data starts from the same question. The difference is that instead of measuring a few selected people on a special day, it can now be measured in every cell, every day.

Glossary

  • Motion study: a method of dividing work into basic motions to find and remove unnecessary ones.
  • Cyclegraph: a long-exposure photograph that records the path of a bulb fixed to a worker’s finger.
  • Therblig: a basic unit of human hand motion defined by the Gilbreths. The name is Gilbreth spelled backwards.
  • MTM (Methods-Time Measurement): a method that calculates standard time by giving a set time value to each basic motion.
  • GSD (General Sewing Data): a standard time system for sewing built on MTM.
  • TMU (Time Measurement Unit): the time unit used by MTM and GSD. 1 TMU is 36 milliseconds.
  • SMV (Standard Minute Value): the standard time it takes to finish one operation, expressed in minutes.

👉 One Tablet per Sewing Cell: Data Collection Designed for a Small Line
👉 How to Read an Operation Breakdown: Operations, SMV, Takt, Man Power
👉 Why Production Data Measured on the Floor Gets Lost

References

  1. ASME, “Frank Bunker Gilbreth” Motions to lay one brick reduced from 18 to 4.5, the adjustable scaffold, meeting Taylor in 1907, the break in 1914. ↩
  2. Wikipedia, “Frank Bunker Gilbreth Sr.”. Born 1868, started as a bricklayer’s helper in July 1885, “one best way”, differences with Taylor, proposal of the role of handing instruments to the surgeon, First World War service and training method, died 14 June 1924, Cheaper by the Dozen and its films. ↩
  3. Wikipedia, “Lillian Moller Gilbreth” Doctorate in applied psychology from Brown University in 1915, work for Macy’s and Johnson & Johnson, the Kitchen Practical in 1929, the foot-pedal trash can and refrigerator door shelves, professor at Purdue University in 1935, National Academy of Engineering in 1965, Hoover Medal in 1966, died 1972, 12 children. ↩
  4. Wikipedia, “Therblig” The list of 18 therbligs, the origin of the name, the 16 in the 1915 paper, transport empty now called reach. ↩
  5. Jill Lepore, “Screen Time” 250,000 feet of 35mm film from 1910~1924, the hand-cranked camera, the clock (microchronometer) placed in every frame. ↩
  6. Internet Archive, “Original Films of Frank B. Gilbreth” Films of bricklaying, typing, soap packing, conveyor work, surgery and cyclegraph demonstrations. ↩
  7. Issues in Science and Technology, “Gilbreth Roots of Motion Capture” (2025) Small bulbs fixed to the fingers and the paths of light, wire models, the photograph of a drill press operator at New England Butt Co., Lillian’s turn to efficiency in the home. ↩
  8. Apparel Resources, “MTM: an institution by itself” MTM developed at Westinghouse in the late 1940s (Maynard, Stegemerten, Schwab), GSD by Peter Evans in 1978, GSD’s use of the TMU. ↩
  9. SIJElog, “One Tablet per Sewing Cell: Data Collection Designed for a Small Line” 33 body joints and 42 finger joints, about 75 points per frame, original footage deleted within 500ms, value-adding and non-value-adding work. ↩
  10. Wikipedia, “Methods-time measurement” 1 TMU = 36 milliseconds, 1 hour = 100,000 TMU, published in 1948 together with the book Methods-Time Measurement released the same year. ↩
back icon Back