Where Was Toyota’s First Andon Cord?
For years, I've been asking a very specific question about Lean history:
Where was Toyota's first andon cord?
Jeff Liker didn't know. Neither did Isao Yoshino. I've asked other former Toyota people.
Toyota histories and books about the development of the Toyota Production System usually explain what an andon does. Some describe how the system evolved. Very few identify where Toyota installed the first one.
Takehiko Harada's Management Lessons from Taiichi Ohno may finally provide the answer.
Toyota's First Andon Was at the Honsha Plant
Harada writes:
“The first andon was put on the engine machining line in Toyota's Honsha plant.”
That is the most specific answer I have found.
Toyota's Honsha plant–originally called the Koromo Plant–was the company's first dedicated automobile factory. It opened in 1938 in what later became Toyota City.
Other historical accounts suggest that Toyota introduced its first andon lights around 1957. If that date is correct, the plant was still officially called Koromo at the time; Toyota renamed it Honsha in 1960.
So the most precise answer may be:
Toyota's first known andon was installed on an engine machining line at the Koromo Plant, later renamed the Honsha Plant, probably during the 1950s.
But that answer comes with a complication.
Harada does not describe a cord.
Toyota's First Andon May Have Used a Button
Harada explains that when a worker believed the job would not be completed within the allotted time, the worker pushed a call button.
That button illuminated a yellow andon light.
If the team could not recover before the designated stopping point, the light turned red and the line stopped.
That is clearly an andon system. But it may not have involved the familiar horizontal cord hanging above an assembly line.
The distinction matters.
We often use “andon” and “andon cord” interchangeably, but the cord is only one possible activation mechanism. Toyota has used buttons, switches, wireless devices, automatic machine sensors, and pull cords.
The andon is the signaling and response system.
The cord is one way to activate it.
Toyota's current description of the Toyota Production System makes that distinction clear. On a manual production line, a worker can pull a cord to call attention to an abnormality. On an automated line, the equipment can detect a problem, stop itself, and illuminate the andon without waiting for a person to intervene.
In both cases, the purpose is the same: make the abnormality visible and bring help to the point where it is needed.
Some Toyota plants today use a call button instead of a cord. It's not the hardware that matters as much as the management software. The reaction to the cord is what matters.
Sometimes the Worker Calls for Help. Sometimes the Machine Does.
Harada says Toyota soon extended andon systems to automated machining lines.
Toyota divided the line into sections. If a part failed to move through a section as expected, the corresponding andon light turned yellow.
Why yellow instead of red?
Harada explained that the signal meant:
“The machine is calling us for help.”
That phrase broadens the usual image of andon.
We tend to picture an assembly worker reaching overhead to pull a cord. That image matters because it represents frontline authority: the person doing the work can expose a problem and ask for assistance.
But Toyota also calls it andon when a machine detects the abnormality.
Sometimes the worker calls for help.
Sometimes the machine calls for help.
The andon makes that call visible.
We Have to Weave In the Loom
Pun intended.
Toyota's andon system did not emerge from the automotive assembly line alone. Its deeper lineage runs through Sakichi Toyoda's automatic weaving looms and the development of jidoka.
In 1896, Sakichi developed a power loom with a mechanism that stopped the machine when a thread broke.
That was a major departure from conventional automation. A machine could run faster than a person, but if it continued operating after something went wrong, it could produce defective material just as quickly.
Sakichi's loom did not continue weaving bad fabric until someone happened to notice. It detected the abnormality and stopped itself.
His Type G Automatic Loom, completed in 1924, extended this principle. One person could monitor multiple machines because each loom was capable of identifying certain problems and stopping automatically.
Toyota describes this as the origin of jidoka, sometimes translated as “automation with a human touch” or “automation with human intelligence.”
The key principle is not merely automation.
It is stopping when something goes wrong.
That same logic appears in the andon system. Whether a person pulls a cord or a machine activates a signal, production should not continue blindly while defects accumulate or an abnormal condition worsens.
The sequence is straightforward:
- Define normal conditions.
- Detect a departure from normal.
- Stop or contain the process.
- Make the abnormality visible.
- Respond.
- Prevent recurrence.
The loom supplied the jidoka lineage.
The andon made the call for help visible.
“When It Flashes, Go”
Harada's larger lesson was not about lights, buttons, or cords.
It was about the response.
His chapter is titled “Standard Work for the Andon Is, ‘Go There When It Flashes.'”
Someone at Toyota had asked Taiichi Ohno how to define standard work for the people supporting a production line.
For an operator performing a repeating cycle, standard work could describe the sequence and timing of each step. But support personnel could not know in advance when or where a problem would occur.
How do you define standardized work for unpredictable events?
Ohno's answer was simple:
“Once the andon flashes, start moving; go fix the problem.”
Harada summarized it even more directly:
“When it flashes, go.”
That may sound obvious. In practice, many organizations install the signal without designing the response.
They give workers a cord or button but no readily available person to answer the call.
They tell employees to stop when something is wrong while keeping supervisors fully occupied producing parts, completing reports, or attending meetings.
They ask people to expose problems but treat every interruption as a failure to hit the production target.
The result is predictable.
The cord remains in place.
The light still works.
People stop using it.
The Team Leader Cannot Respond While Doing Someone Else's Job
Harada warned specifically against assigning routine production work to the team leader merely because no andon light was flashing at that moment.
The logic sounds efficient: when there are no problems, why should the team leader appear to be waiting?
But if the team leader is already performing a production assignment when a worker calls, the response is delayed. The worker waits. The process falls farther behind. One unresolved problem creates another.
What looks efficient on a staffing spreadsheet makes the system less responsive.
The team leader's standard work is not to remain busy producing parts every second. It is to be available when the process needs support.
And when the andon is quiet, Harada wrote, the leader should investigate the causes of previous calls and work to prevent them from happening again.
That creates the complete management loop:
- The worker or machine detects an abnormality.
- The andon makes it visible.
- The team leader responds immediately.
- The immediate problem is contained.
- The underlying cause is investigated.
- The process is improved.
The cord is not the system.
The response is the system.
Read more: The Most Misunderstood Role in the Toyota Production System
So Where Was Toyota's First Andon Cord?
Based on Harada's account, Toyota's first andon was installed on an engine machining line at the plant now known as Honsha.
It was probably introduced during the 1950s.
It may have used a button rather than a cord.
That means the answer depends slightly on the question.
Where was Toyota's first andon?
At the Honsha–or then Koromo–engine machining line.
Where was Toyota's first literal pull cord?
I have not yet found a source that answers that question definitively.
But Harada gives us something more valuable than the location of a piece of hardware. He explains the management thinking Toyota built around it.
An andon was not simply a way to stop production.
It was a way to expose the gap between normal and abnormal conditions, bring assistance immediately, and create an opportunity for improvement.
That appears to settle the Toyota part of the story.
But then another historical thread appeared.
Before Toyota, Ford Had a Stop Cord
The earliest documented automotive andon cord I have found was not at Toyota.
It was at Ford's River Rouge complex in Dearborn, Michigan.
Edwin P. Norwood described the system in his 1931 book, Ford: Men and Methods.
Inside Ford's Motors Building, a control room overlooked the production operation. An instrument board contained numbered signal lights corresponding to locations along the conveyor.
When a worker encountered a problem, he could stop the conveyor by throwing a switch or pulling an overhead cord. Norwood compared it to the cord passengers used to signal the driver of a city bus.
Pulling the cord stopped the conveyor and illuminated the corresponding light in the control room.
If the stoppage continued for more than two minutes, a mechanic was dispatched. The control-room operator recorded where the interruption occurred and how long it lasted.
Most strikingly, Norwood wrote that any worker was free to stop the line–and was expected to do so.
By 1931, Ford had the recognizable elements of an andon system:
- A worker could detect a problem.
- The worker could pull an overhead cord.
- The conveyor stopped.
- A visual display identified the location.
- A support person responded.
- The location and duration of the interruption were recorded.
Ford did not call it an andon.
But it is difficult to call it anything else today.
Did Toyota Learn the Andon Cord From Ford?
The timeline makes the question unavoidable.
Ford's stop-cord system was documented in 1931.
Eiji Toyoda visited Ford operations in the United States in 1950, including the enormous River Rouge complex. Toyota executives studied Ford's factories, methods, and management practices closely.
Later that year, another Toyota executive, Shoichi Saito, trained at River Rouge. After returning to Japan, he helped introduce an employee suggestion program based explicitly on Ford's system.
Toyota was clearly willing to study Ford, adapt what it found useful, and incorporate those practices into its own developing production system.
Could Toyota have learned about the stop cord the same way?
Possibly.
John Shook reportedly saw a copy of Norwood's Ford: Men and Methods in a Toyota library. The detail makes a possible connection more intriguing, but it does not tell us when Toyota obtained the book, who read it, or whether it influenced the development of Toyota's andon system.
But it does not prove the chain of influence.
A book on a shelf does not tell us:
- When Toyota obtained it.
- Who read it.
- Whether Eiji Toyoda or Taiichi Ohno knew about the passage.
- Whether the Ford system was still operating during Toyota's visits.
- Whether Ford's stop cord directly influenced Toyota's andon.
Toyota may have learned the mechanism from Ford.
It may have developed a similar solution independently.
Or it may have woven together something observed at Ford with the jidoka logic inherited from Sakichi Toyoda's loom.
The evidence makes Ford influence plausible.
It does not make it proven.
The Ford Irony
The larger irony is that Ford appears to have had the cord first–and then, at least in many plants, lost the culture required to use it.
A 2007 BBC report compared andon use at Toyota's Georgetown, Kentucky, plant with Ford's Dearborn truck plant.
Toyota workers reportedly pulled the andon cord about 2,000 times per week.
Ford workers pulled theirs about twice per week.
That difference cannot reasonably mean that Toyota had 1,000 times more problems.
It means Toyota workers were far more willing–or more strongly expected–to make problems visible.
Ford had introduced a team-leader structure and an andon system at Dearborn. But those mechanisms were entering an organization shaped by decades of adversarial labor relations, mistrust, production pressure, and different management habits.
The cord existed.
The credible expectation of a helpful response apparently did not exist to the same degree.
That is the historical reversal.
Ford appears to have developed an early physical mechanism for frontline voice.
Toyota built and sustained the stronger social and managerial mechanism around it.
American manufacturers then spent decades trying to copy Toyota's visible tools, sometimes without recognizing that Ford had used remarkably similar hardware generations earlier.
Ford did not need to discover the cord.
It needed to remember how to make pulling it worthwhile.
The First Cord Is Less Important Than What Happened Next
The origin story matters, but not because one company deserves permanent credit for inventing a rope connected to a switch.
The more important question is what happened after someone pulled it.
Did a leader come?
Did the leader thank the person for exposing the problem–or treat the interruption as an inconvenience?
Did the organization contain the immediate risk?
Did anyone investigate why the problem occurred?
Did the process improve, or did the same person have to pull the same cord tomorrow?
Ford may have had the physical cord first.
Toyota may have learned something from Ford–or arrived at a similar mechanism through its own experimentation.
What Toyota clearly developed was a management system in which the call for help was expected, the response was designed, and the abnormality became an opportunity to improve the process.
That is why the andon cord became associated with Toyota rather than Ford.
The power was never in the cord.
It was in what workers believed would happen after they pulled it.
Where was the first andon cord used?
The best available account says Toyota's first andon was installed on an engine machining line at the Koromo Plant, later renamed the Honsha Plant. The source describes a call button, however, so it does not prove that the first andon used a physical cord.
When did Toyota introduce the first andon?
A TPS historical timeline prepared by Art Smalley, citing Michael Cusumano, dates Toyota's first andon lights to 1957. Harada's account identifies the location but does not provide a year.
What is an andon?
An andon is a visual signaling system that shows where an abnormality has occurred and where assistance is needed. It may be activated manually by a cord or button, or automatically by production equipment.
Does pulling an andon cord immediately stop the line?
Not necessarily. In a typical Toyota assembly-line system, the initial signal calls the team leader. The line stops at a designated point only if the problem cannot be resolved within the remaining work cycle.
How is andon related to jidoka?
Jidoka is the broader principle of detecting abnormalities and stopping rather than continuing to make defects. Andon is one method for making the abnormal condition visible and calling for a response.
Did Ford have an andon cord before Toyota?
Edwin P. Norwood's 1931 book Ford: Men and Methods describes workers at Ford's River Rouge complex using overhead cords to stop conveyors and activate numbered signal lights. It is the earliest documented automotive andon-cord system I have found.
Did Toyota copy Ford's andon system?
Toyota executives studied Ford operations extensively in 1950, including River Rouge, so Ford influence is plausible. No source found so far proves that Toyota directly copied Ford's stop-cord system.
