History of Six Sigma Study-ready

Pre-1980s foundations

• SPC / control charts (Shewhart) and Deming–Juran quality management set the analytic culture Six Sigma later packaged.^[1]
• ASQ’s History of Quality timeline situates Six Sigma inside a century of industrial quality practice.^[1]
• 1919 — Charles R. Allen’s four steps: WWI shipyard instructor training (U.S. Shipping Board Emergency Fleet Corporation) yields The Instructor, the Man and the Job (Lippincott, 1919): Preparation → Presentation → Application → Testing (or Inspection) — the method TWI Job Instruction was built on.^[3][4] See WWI · Charles Allen. Internet Archive scan →
• Training lineage: WWII U.S. TWI (born from factory restaffing under the draft) → postwar GHQ introduction in Japan → Toyota JI/JM/JR adoption (1950s) shaped how Lean plants teach and improve standards — see WWII factory restaffing and Toyota & TWI.^[4][14][15]

1940–1945 — WWII restaffing & Training Within Industry

• Selective Service and war expansion pulled skilled factory workers into the Army and Navy while plants had to produce more matériel with a newer, less experienced workforce (women, minorities, younger workers, transfers from other industries).^[5][4]
• The U.S. created Training Within Industry (TWI) (1940–45, later under the War Manpower Commission) so supervisors could train replacements fast on the job — not in long classroom courses.^[5][4]
• By war’s end: certifications for ~1.6 million workers across ~16,500 plants. JI / JM / JR became the portable “how we teach / improve / lead” kit.^[5]
• Full picture below → WWII factory restaffing.

1986 — Motorola invents Six Sigma

• Engineer Bill Smith at Motorola formalizes defect-reduction / variation control labeled Six Sigma (commonly dated 1986).^[16][7]
• Stretch goal: ~3.4 DPMO under the classic 1.5σ shift convention — a quality target, not a slogan.^[2][7][18]
• Motorola builds curricula and practitioner qualifications; wins Malcolm Baldrige (1988).^[7]
• Mikel Harry and others help spread martial-arts-inspired belt vocabulary as a training/deployment system.^[17]

1990s — GE and Fortune 500 scale

• Jack Welch makes Six Sigma a GE corporate initiative (mid-1990s), tying training, projects, and leadership to results.^[16]
• AlliedSignal/Honeywell and others adopt similar programs; many Fortune 500 firms follow.^[16]
• Roles harden: Champions sponsor; Black Belts lead complex work; Green Belts lead scoped projects; Yellow/White support awareness.^[2][19]

2000s — Lean Six Sigma + professional BoKs

• Lean (TPS waste/flow) merges with Six Sigma (variation/defects) as Lean Six Sigma.^[2][18]
• Professional bodies (notably ASQ) publish belt certifications with explicit Bodies of Knowledge.^[20]
• DMAIC remains the workhorse for existing processes; DFSS/DMADV for new design.^[2][18]

2011 — ISO publishes Six Sigma standard

• In 2011, the International Organization for Standardization (ISO) published the first standard ISO 13053:2011 defining a Six Sigma process.^[8]
• ISO 13053 gives an international, consensus definition of Six Sigma methodology (DMAIC-oriented), beyond company curricula and professional BoKs.^[8]

The WWI shipyard squeeze

• Fleet expansion: Germany’s U-boat campaign forced the United States to acquire, build, and operate cargo ships to supply an army overseas. Congress created the U.S. Shipping Board (1916) and, days after the U.S. entered the war, the Emergency Fleet Corporation (16 Apr 1917) to drive emergency construction.^[22]
• Workforce explosion: Edward N. Hurley, wartime Shipping Board chairman, reported that U.S. shipyards had fewer than 50,000 workers when America entered the war and about 350,000 by the end of 1918 — with another ~180,000 in mills and factories supplying engines, boilers, and materials. Within about nine months, yard employment rose from roughly 50,000 to 350,000.^[22]
• Training problem: That surge meant huge numbers of inexperienced people had to become productive on shipyard jobs quickly. Long craft apprenticeships could not keep pace; yards needed structured on-the-job instruction by people who could teach as well as do the work.^[3][4]

Who Charles R. Allen was

• Vocational educator: Allen came from industrial/vocational education. His 1919 title page identifies him as sometime Agent for Industrial Training of Boys and Men for the Massachusetts Board of Education, and as Superintendent of Instructor Training for the U.S. Shipping Board Emergency Fleet Corporation.^[3]
• EFC role: Under the Emergency Fleet Corporation he directed instructor-training work for shipyards — training experienced mechanics/trade people to become instructors so they could teach new workers on the job, rather than relying on informal “watch the guy next to you” pickup.^[3][4]
• Book as field notes: The Instructor, the Man and the Job (Lippincott, 1919) was written as both a plant instructor handbook and “instruction notes” for instructor-training courses. Allen says the material grew out of earlier instructor courses and was then modified for EFC shipyard instructor training under his direction.^[3]

What he taught — the four-step lesson

• Four steps: Preparation → Presentation → Application → Testing (or Inspection). Purpose, in Allen’s words: get the learner ready; instruct him; check up errors; give a final inspection of the instruction job. Steps always in that order.^[3][4]
• Job analysis: Much of the book is trade analysis — listing what must be taught, breaking work into teachable units/steps, and arranging an effective instructional order (easier learning difficulties before harder ones) so experienced workers can teach systematically.^[3]
• Instructing ≠ showing: Allen insists that merely showing a job is not instructing. If a learner is “shown” and still cannot do the work, “the man has failed to learn, not because he is stupid, but because he has not been really taught, and the blame for the failure rests not on the man but on the instructor.”^[3]
• Scale (verified): In the 1919 preface Allen reports that, in the EFC instructor-training work under his direction, over one thousand instructors had been trained. (He does not publish a matching total of shipyard workers trained by those instructors in the preface.)^[3]

How this leads to TWI — and Lean plants

• TWI Job Instruction: WWII Training Within Industry built Job Instruction on Allen’s four-step method. TWI leaders (including Mike Kane, who had worked with Allen-era WWI training) adapted the same prepare–present–try out–follow up logic into the JI pocket-card program.^[4][5]
• The famous motto: TWI’s JI card slogan — “If the worker hasn’t learned, the instructor hasn’t taught” — is the later wartime phrasing. Allen’s 1919 book already carries the same accountability idea (failure after mere “showing” is the instructor’s fault), but that exact motto wording is TWI’s, not a verbatim Allen quote.^[3][4][5]
• Lineage forward: TWI’s J programs later moved into occupied Japan and into Toyota supervisor training — so Allen’s WWI instructor method is an ancestor of how Lean plants still teach standards. Full WWII restaffing story → WWII factory restaffing. Method detail → TWI / Job Instruction. Toyota handoff → Toyota & TWI.^[4][14][15]

What it is

• Watch and copy: A new hire shadows or sits beside an experienced worker, observes, asks occasional questions, and gradually imitates. British industry long called this colloquially sitting by Nellie (or “watching Joe” / “standing by Fred”) — learning by observing and copying an experienced worker rather than by a planned course.^[23][24]
• Unstructured OJT: In modern training literature this is unstructured (unplanned) on-the-job training: the work itself sets the sequence; there is no written standard to teach from; the “trainer” is primarily a producer who also happens to have a new person nearby.^[27]
• Allen’s “absorption”: In 1919 Allen described training by absorption as shops with no definite training arrangements — new people “pick up” the work as they can, get help from whoever is friendly at the next machine, and “use their eyes and their mouths.” The plant assumes no real responsibility for seeing that they get trained.^[3]

Why companies use it

• Looks free and fast: No classroom, no course fees, no trainer curriculum — just put the new person on the floor. That is why it feels cost-effective, even when the experienced worker’s lost production is a hidden cost.^[24]
• Keeps production running: Training happens inside real work, so managers feel they are not taking people offline. The expert is already there; tradition says skill will transfer naturally if you watch long enough.^[24][27]
• No written standard to teach: If the “one best way” lives only in someone’s head, shadowing seems like the only option. Unstructured OJT is shaped by whatever work happens to arrive that day, not by a planned list of what the new person must learn.^[27]
• Social fit: Pairing with a colleague also looks like a way to integrate socially — another reason the method persists even when critics call it outdated or haphazard.^[23][24]

Why it doesn’t work (or works poorly)

• Experts aren’t teachers: Knowing a job and knowing how to teach it are different skills. Allen insisted that merely showing is not instructing: if a learner is shown and still cannot do the work, “the blame for the failure rests not on the man but on the instructor.”^[3]
• Key points and reasons stay tacit: In unstructured shadowing, neither trainer nor trainee may know precisely what to emphasize. Subtle “make or break” details and the why behind steps often go unspoken, so the learner copies motions without understanding.^[25][26]
• Bad habits travel with good ones: If “Nellie” uses outdated methods, ignores safety rules, or has a personal shortcut, the new person inherits that drift. Without a shared standard, methods diverge shift-to-shift and trainer-to-trainer — a “telephone game” of informal handoffs.^[24][25]
• Slow, inconsistent, unchecked: Shadowing can take days or weeks with uneven results and no clear check that the learner can do the job correctly, safely, and conscientiously. Trainers also use different methods each time; content is often incomplete or inaccurate.^[25][26][27]
• Production cost is real: The experienced worker’s time coaching (or being interrupted) reduces their own output — so the “free” method still costs money, just off the training budget line.^[24][27]
• TWI’s answer: Job Instruction exists specifically because unstructured “show them / have them shadow” is not instruction. JI’s credo — “If the worker hasn’t learned, the instructor hasn’t taught” — puts ownership on the instructor and replaces absorption with prepare–present–try out–follow up, using a job breakdown of important steps, key points, and reasons.^[5][4][26]

What works instead

• Structured on-the-job instruction: Teach from a job breakdown; demonstrate with key points and reasons; have the learner try under coaching until you know they know; then follow up. That is Allen’s four-step lineage and TWI Job Instruction — the opposite of hoping skill transfers by proximity.^[3][5][26]
• On this library: How Allen built instructor training → WWI · Charles Allen. Why WWII plants needed structured JI → WWII factory restaffing. How to run JI → TWI / Job Instruction. Whole-trade, multi-year path → Apprenticeship compared.

What a Registered Apprenticeship is

• Earn-and-learn career pathway: A U.S. Registered Apprenticeship (approved by the Department of Labor or a State Apprenticeship Agency) combines paid work with a mentor, progressive wage increases, related classroom (or equivalent) instruction, and a portable, nationally recognized credential on completion.^[28]
• Federal standards (29 CFR 29.5): Program standards must include a written plan, a work-process schedule of supervised on-the-job learning, organized related instruction, a progressively increasing wage schedule, performance review, and a completion certificate from the Registration Agency. For a time-based program, the industry standard for on-the-job learning is at least 2,000 hours for the apprenticeship term. Related instruction carries a recommended minimum of 144 hours for each year of apprenticeship (classroom, industry courses, electronic media, or other approved forms).^[29]
• U.S. history in brief: Craft apprenticeship long predates modern registration — early American trades (printers, silversmiths, shipwrights, and others) trained people under masters.^[9] Wisconsin enacted the first state apprenticeship law placing authority under an industrial commission in 1911 (and earlier required classroom hours for apprentices); Congress enacted the National Apprenticeship Act (Fitzgerald Act) in 1937, authorizing the Secretary of Labor to promote labor standards for apprentices and cooperate with the states.^[9][32] A WWII-era example already on this page: St. Louis-area machinist standards of 8,000 hours / about four years.^[21]
• Who teaches: On-the-job learning is supervised under journeyworkers / mentors; related-instruction instructors must meet state vocational-technical requirements or be subject-matter experts (e.g. journeyworkers) and have training in teaching techniques and adult learning styles (before or after they start teaching related instruction).^[29]

Side-by-side comparison

Notes: “≥2,000 hours” is the federal minimum on-the-job learning for a time-based apprenticeship term (not a separate yearly quota). “144 hours” is a recommended minimum of related instruction per year of apprenticeship. Competency-based and hybrid programs also exist under 29 CFR 29.5.^[29]

Shortcomings of apprenticeship

Apprenticeship remains one of the strongest earn-and-learn paths available — but it is not a universal fix. Multi-year terms are too slow when demand surges overnight (the same reason WWII plants could not lean on craft apprenticeships alone — see WWII factory restaffing). Sponsors pay progressive wages, journeyworker mentoring time, and related instruction while apprentices are still ramping up, and they face a classic free-rider risk: competitors may hire away graduates (“poaching”). In a nationally representative Urban Institute / DOL survey of sponsors, about one in four called poaching a significant problem and about one in four said the same about dropouts — even though most current sponsors still strongly recommend the model. Completion is real attrition, not a rounding error: RAPIDS figures published with DOL’s 2016 apprenticeship EEO rule showed all-industry completion around the low-to-mid 40s percent for men in FY2015 (higher for women overall, but lower for women and Black apprentices in construction). Access has long been uneven. Construction has historically trained the majority of U.S. apprentices, yet in FY2015 women were only about 2.8% of new construction enrollments; as recently as FY2024 women were still only about 14% of active registered apprentices nationwide. Courts and DOL’s own EEO history document decades of exclusion and informal networks that shut women and Black workers out of many trade programs — the same pools war plants later drew on when plants restaffed. Quality on the job is not automatic either: federal standards require structured OJL and related instruction, but journeyworker mentors often have no required training in how to teach, so floor practice can slide back into training by absorption unless the sponsor adds Job Instruction–style methods. Finally, U.S. registered apprenticeship is still small and narrow relative to dual systems abroad — roughly 0.3% of the labor force, with about half of civilian registered apprentices in construction — whereas large shares of German and Swiss youth cohorts enter apprenticeship across many occupations. Traditional time-based terms (at least 2,000 hours OJL under 29 CFR 29.5) can also hold back fast learners; DOL now allows competency-based and hybrid approaches for that reason.^{[4][9][21][33][35][34][32][36][29]}

• Speed vs. surge demand: Years to journey level cannot match a draft-driven restaffing crisis; that is why TWI JI existed alongside — not instead of — apprenticeship.^[4][5][21]
• Employer cost and poaching: Wages, mentor time, and classroom costs are real investments; about 25% of surveyed sponsors called competitor poaching a significant problem (46% saw no poaching problem).^[33]
• Attrition: Sponsors often worry about dropouts; public RAPIDS completion rates in the mid-2010s sat well below “most finish,” especially for some groups in construction.^[33][35]
• Equity gaps: Women’s share of active apprentices ~14% (FY2024) vs nearly half the labor force; construction enrollments historically far lower for women; Black apprentices underrepresented and with lower completion in construction in FY2015 RAPIDS data.^[34][35]
• Teaching quality on the job: Journeyworkers are not required to be trained instructors; without structured methods, OJL can become absorption — which is why JI fits inside apprenticeships.^[32][26][3]
• Narrow U.S. footprint: ~0.3% of the U.S. labor force; ~50% of civilian RAs in construction — vs. dual systems where large youth cohorts enter apprenticeship across many fields.^[36]
• Time clocks vs. competence: Time-based minima can over-constrain fast learners; competency-based and hybrid terms are allowed under current federal standards.^[29][33]

How they fit together

• Different jobs, different tools: Apprenticeship builds broad trade mastery over years. JI-style structured OJT gets someone productive on a specific job in days or weeks. Absorption is the unstructured default both are improvements over.^[28][27][3]
• JI inside apprenticeship: Nothing stops a sponsor from using Job Instruction (job breakdowns, four steps, instructor accountability) to teach each work-process element inside an apprenticeship. Structured OJT is a teaching method; apprenticeship is a multi-year occupational pathway that still needs good instruction on the floor.^[5][26][29]
• Why WWII needed JI, not more apprenticeships: When plants restaffed for war production, multi-year craft apprenticeships could not keep pace. Less-skilled production jobs had mostly been learned by watching. TWI Job Instruction was the national answer for fast, structured on-the-job teaching — see WWII factory restaffing.^{[4][9][21][5]}
• Modern relevance: Registered Apprenticeship remains DOL’s earn-and-learn model across trades and newer sectors. DOL’s Apprenticeship.gov notes that about 90% of apprentices who complete continue employment with the employer afterward. An Urban Institute / Abt evaluation summary of DOL’s American Apprenticeship Initiative reported average earnings up about 49% from the year before the program to the year after exit (with caveats — outcomes, not pure causal impacts; longer construction programs largely excluded).^[28][30]
• Manufacturing example: Toyota launched the Federation for Advanced Manufacturing Education (FAME) in 2010 as an apprenticeship-style Advanced Manufacturing Technician pathway (community-college academics plus substantial paid on-the-job hours). Stewardship moved to The Manufacturing Institute in 2019 for national scaling; FAME USA reports 1,800+ hours of hands-on experience plus an associate degree, and that nearly 90% of graduates secure full-time roles with their sponsoring employer.^[31]

How many factory workers?

• Before the surge: BLS’s Current Employment Statistics series (all employees, manufacturing) puts U.S. manufacturing employment at roughly 9.5 million as an annual average in 1939 and about 10.1 million in 1940 — on the order of ten million factory payroll jobs as the defense program started.^[37]
• Wartime peak: The same series rises through the conversion years and peaks near 16.6 million in late 1943 (November CES reading), roughly a 70%+ climb from the 1939 level — even while the draft was pulling men out of civilian life.^[37][40]
• Skills gap: Multi-year craft apprenticeships and informal absorption could not staff that expansion. Shortages of skilled metal-trades workers showed up during the 1940–41 defense buildup; that is the skills backdrop for TWI and for why plants later needed Job Instruction, not more four-year apprenticeships alone — see Apprenticeship compared and WWII factory restaffing.^[40][4][5]

How many went into the armed forces?

• National total: About 16 million Americans served in uniform during World War II (out of a 1940 Census population of about 132 million).^[39]
• No clean “factory workers drafted” national total: Published BLS/WMC/Selective Service summaries do not give a single, authoritative count of how many of those 16 million came from manufacturing payrolls specifically. What they do show is the scale of the drain on the civilian labor market and the simultaneous boom in factory jobs.
• Best sourced proxies: From July 1940 to July 1943 the civilian labor force fell by about 1.4 million while nonfarm employment still rose — the draft and enlistment removed men faster than new entrants fully replaced them in the civilian pool, even as manufacturing employment jumped nearly 70%. Contemporary mobilization planning (summer 1942) estimated that about 3.4 million men would be withdrawn from the labor market for military service in calendar 1942 alone. Plant-level Selective Service “replacement lists” (already cited on this page) tracked which experienced factory workers were leaving each firm — evidence of a real industrial drain, even without a neat national manufacturing-only headcount.^{[40][42][12][5]}

The squeeze

• Draft + enlistment: Selective Service and voluntary enlistment pulled large numbers of experienced machinists, assemblers, welders, and supervisors into the Army, Navy, and other services.^[5]
• War orders up: At the same time, the War Department poured contracts for aircraft, ships, munitions, vehicles, and instruments — plants needed more output, not less.^[5][4]
• Replacement lists: Firms notified of drafted workers compiled workforce composition (including women and African American workers) and estimated how long replacements would take to become productive — a paperwork trail of the restaffing crisis.^[12]

Who filled the jobs

• Women in war plants: Millions of additional women entered paid work; manufacturing’s share of women’s employment rose sharply. “Rosie the Riveter” became the public face of women in aircraft, shipyards, ordnance, and machine shops (many with little prior factory experience).^[10][11]
• Broader labor pool: Plants also hired and upgraded African American workers, younger workers, rural migrants, and people shifting from non-defense industries — often into jobs that had been closed to them before the war.^[10][12][11]
• Demand as much as draft: Historians note war-contract demand and active recruiting pulled women into plants at least as much as draft-driven male shortages alone — both forces reshaped the shop floor.^[13][11]

How training had to change

• Long apprenticeship-style training could not keep pace. Before the war, skilled trades like machinist, toolmaker, and electrician were usually learned through multi-year apprenticeships (often about four years), formalized nationally by the Fitzgerald Act (National Apprenticeship Act) of 1937. Less-skilled production jobs were mostly picked up informally by watching an experienced worker, with little structured instruction.^[9][21][4]
• New people needed to be productive in days or weeks, on real jobs, under real supervisors.^[5][4]
• Training Within Industry (TWI) (Aug 1940 → Sep 1945) gave plants a national, free consulting service: teach supervisors how to instruct, improve methods, and lead people — then multiply trainers plant-to-plant.^[5][4]
• Job Instruction (JI): break the job into steps, demonstrate key points and reasons, let the learner try under coaching, follow up — motto: “If the worker hasn’t learned, the instructor hasn’t taught.”^[4][5][6]
• Job Methods (JM): question every step; eliminate, combine, rearrange, or simplify; sell and standardize the better way.^[4][5][6]
• Job Relations (JR): treat people as individuals; facts-based handling of people problems under high turnover stress.^[4][5][6]
• Scale: by war’s end, over 1.6 million certifications in over 16,500 plants. After 1945 the U.S. government wound down domestic TWI funding, then helped export the methods (including into occupied Japan — the bridge to Toyota).^[5][4][14]