Quality Engineering

Quality engineering is the branch of industrial engineering that designs the means by which a product or process is made to meet a standard—reliably, measurably, and at acceptable cost.

11 frameworks, from 1900 to today · Industrial Engineering · Engineering

still a working view ended · each leads to its entry below

Overview

Quality engineering is the branch of industrial engineering that designs the means by which a product or process is made to meet a standard—reliably, measurably, and at acceptable cost. Its practitioners do not usually inspect quality into a product at the end of the line; they build it in upstream, by choosing tolerances, designing experiments, setting control charts on the process, and organizing the management systems that keep a factory or a service honest about its own performance. What is at stake is the difference between a product that works most of the time and one that works predictably, and between a company that discovers its defects and one that prevents them. The field's history is largely a story of moving quality control earlier and earlier in the production cycle—from inspection, to the process, to the design, to the whole organization.

Inspection and the birth of statistical quality control

Before the twentieth century, quality was a matter of craft and inspection12. Skilled workers made parts to fit, and inspectors sorted good from bad at the end. The limit of this approach appeared with mass production: at the Western Electric Hawthorne plant in the 1920s, telephone equipment was built in enormous volumes, and 100 percent inspection was slow, expensive, and imperfect—inspectors missed defects, and testing every item could destroy it.

The decisive move came from Walter Shewhart, a physicist at Bell Telephone Laboratories, who in 1924 proposed that a manufacturing process could be monitored by plotting measurements over time and watching for patterns that signaled trouble1345. His insight was statistical: every process has natural variation, and as long as that variation stays within limits predicted by the process's own random behavior, the process is "in control."5 When points fall outside those limits, or form non-random runs, something assignable has changed and should be investigated5. Shewhart published these ideas in Economic Control of Quality of Manufactured Product (1931), introducing what became the control chart and the distinction between common-cause variation (inherent to the process) and special-cause variation (from a specific, findable disturbance). This was the founding act of statistical quality control (SQC): quality became something measured continuously during production rather than judged at the end13.

Shewhart's colleague W. Edwards Deming carried the ideas to Japan after the Second World War, and Joseph Juran, another American, taught there as well. Their influence on Japanese industry in the 1950s is one of the great transfers in the field's history1.

Sampling, acceptance, and the wartime push

The Second World War forced quality engineering to scale1. The U.S. military needed to accept or reject huge lots of munitions and equipment without testing everything, and statisticians responded with acceptance sampling: draw a sample, count defects, and decide whether the lot passes6. The work of Harold Dodge and Harry Romig at Bell Labs, and the military standards built on it, gave industry a practical, statistically defensible way to make that decision36. The same wartime urgency produced the techniques that would dominate the next generation.

Design of experiments

The deepest methodological contribution came from agriculture and moved into industry. Ronald Fisher, working at Rothamsted Experimental Station in England in the 1920s, developed the analysis of variance and the principles of experimental design—randomization, replication, and blocking—so that several factors could be varied at once and their effects separated. His The Design of Experiments (1935) is the foundational text7. The practical payoff for industry is large: instead of changing one factor at a time (slow, and blind to interactions between factors), an engineer can run a structured set of trials and learn how several inputs jointly affect an output.

George Box and K. B. Wilson extended this in 1951 with response surface methodology, a way to find the settings that optimize a response by fitting a model and climbing toward the optimum. Box's later work with Norman Draper made designed experiments a standard industrial tool. The concrete case that shows what this meant: an engineer tuning a chemical process can, with a fractional factorial design, learn in a handful of runs which of many variables matter and how they interact, rather than running hundreds of one-at-a-time trials.

Reliability engineering

As products grew more complex—radar, aircraft, electronics—the question shifted from "does this batch meet spec?" to "how long will this system keep working?" Reliability engineering emerged from military and aerospace work in the 1950s. The bathtub curve, describing high early failure (infant mortality), a long low-failure middle life, and rising wear-out failures, became a standard mental model. The exponential and Weibull distributions gave engineers ways to model time-to-failure, and methods such as failure modes and effects analysis (FMEA) and fault tree analysis let designers reason systematically about what could go wrong before it did. Reliability is now treated as a sibling discipline, but its roots in the same wartime and postwar industrial engineering are shared.

Robust design and the Taguchi contribution

In the 1950s and 1960s, Genichi Taguchi in Japan developed an approach that reframed quality around loss rather than conformance. His central idea, the quality loss function, held that any deviation from a target value costs society, and that reducing variation around the target matters more than merely staying inside tolerance limits. His methods—orthogonal arrays for designing experiments, and the use of "noise" factors (uncontrollable variation such as temperature or user behavior) to find settings that make a product insensitive to them—became known as robust design or the Taguchi methods. They spread to the West in the 1980s and provoked both enthusiasm and statistical criticism, but the underlying aim—designing products that perform well despite variation—became a permanent part of quality engineering.

Total Quality Management and the organizational turn

Through the 1970s and 1980s, Japanese manufacturers gained market share with products that were both better and cheaper, and Western firms looked for the cause. Part of the answer was that quality had become everyone's job, not the inspection department's8. Deming's "fourteen points" and Juran's quality trilogy (planning, control, improvement) framed quality as a management responsibility12. Armand Feigenbaum had coined "total quality control" in 1961, arguing that quality is affected by every function of a firm. These strands converged into Total Quality Management (TQM): a management philosophy holding that quality is defined by the customer, achieved through continuous improvement and the involvement of all employees, and led from the top8. Its tools included quality circles, benchmarking, and statistical thinking across the organization. TQM's influence was enormous and its record mixed—many programs became slogan-driven and faded—but it permanently moved quality engineering from the shop floor into strategy.

Quality management systems and standards

As quality became organizational, it needed to be auditable and comparable across firms and borders. The result was the quality management system (QMS): a documented set of processes, responsibilities, and records by which an organization ensures and demonstrates quality2. The ISO 9000 family, first published in 1987 by the International Organization for Standardization, standardized the requirements for such a system, and certification against it became a commercial necessity for many suppliers. The QMS idea descends directly from TQM's insistence that quality be systematic, but it added the discipline of external audit and the risk that documentation could become an end in itself.

Six Sigma and lean quality

In the 1980s, Motorola developed Six Sigma, a disciplined, project-based improvement methodology aimed at reducing defects to a few per million opportunities. Its name refers to a statistical target: a process so capable that its specification limits sit six standard deviations from the mean9. Six Sigma packaged familiar statistical tools—control charts, designed experiments, measurement system analysis—into a structured sequence (define, measure, analyze, improve, control) and a hierarchy of trained practitioners (Green Belts, Black Belts, Master Black Belts). It spread widely after General Electric adopted it in the mid-1990s9. In parallel, lean quality drew on the Toyota Production System: the idea that quality is built in at the source, that defects should stop the line (jidoka), and that waste and error share root causes. Lean and Six Sigma were later combined as Lean Six Sigma, and both remain among the most widely taught improvement frameworks.

Multivariate monitoring and the modern toolkit

Classical control charts watch one variable at a time, but real processes have many correlated measurements—temperatures, pressures, dimensions—that move together. Multivariate statistical process monitoring, developed from the 1940s onward and maturing in the later twentieth century, uses methods such as Hotelling's T² statistic and principal component analysis to detect when the joint behavior of many variables departs from normal, and to diagnose which variable is responsible. This is the direct descendant of Shewhart's chart applied to the complexity of modern manufacturing and chemical processes.

The present

Quality engineering today is a layered discipline. Statistical process control and acceptance sampling remain the daily tools of the factory floor; designed experiments and robust design are standard in product and process development; reliability methods are embedded in design for aerospace, automotive, and electronics; and quality management systems, Six Sigma, and lean improvement structure how organizations govern and change themselves. The frontier has moved toward data: automated sensing, machine learning for fault detection, and multivariate monitoring of increasingly instrumented processes, alongside a persistent concern with measurement itself—whether the numbers a system reports can be trusted. The questions still open are old ones in new dress: how to design quality into products whose complexity outstrips any single engineer's understanding, and how to keep improvement from becoming paperwork.

References

  1. Quality management (History), Wikipedia
  2. Quality management system (History, introduction), Wikipedia
  3. Statistical process control (History, introduction), Wikipedia
  4. Walter A. Shewhart, Wikipedia
  5. Control chart (introduction, History), Wikipedia
  6. Acceptance sampling (introduction, History), Wikipedia
  7. The Design of Experiments, Wikipedia
  8. Total quality management, Wikipedia
  9. Six Sigma (introduction, History), Wikipedia

The frameworks of Quality Engineering

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1900-1935
  2. 1924-Present
  3. 1935-Present

    Design of Experiments

    Which inputs actually change the output?

    Fly to its tower in Nousfarer

  4. 1940-Present

    Acceptance Sampling

    How many items must we check to judge a lot?

    Fly to its tower in Nousfarer

  5. 1947-Present
  6. 1950-Present

    Reliability Engineering

    How long will a product keep working?

    Fly to its tower in Nousfarer

  7. 1960-Present

    Taguchi methods

    How do we design products that resist variation?

    Fly to its tower in Nousfarer

  8. 1970-Present

    Lean Quality

    How do we stop defects at their source?

    Fly to its tower in Nousfarer

  9. 1970-2000

    Total Quality Management

    Whose job is quality in an organization?

    Fly to its tower in Nousfarer

  10. 1986-Present

    Six Sigma

    How do we cut defects to near zero?

    Fly to its tower in Nousfarer

  11. 1987-Present

    Quality Management Systems

    How do we prove quality to outsiders?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Inspection-based quality controlStatistical process controlDesign of ExperimentsAcceptance SamplingMultivariate Statistical Process Moni…Reliability EngineeringTaguchi methodsLean QualityTotal Quality ManagementSix SigmaQuality Management Systems

Beyond the field

Where Quality Engineering's ideas came from in other fields, and what they led to there: the strongest 12 of 90 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Quality Engineering is a planet in Nousfarer

Its city has a tower for each of these 11 frameworks. Land, walk in, and climb one concept at a time.

Land on Quality Engineering

Related fields

Fields its ideas are linked to

More in Industrial Engineering

See where Quality Engineering sits on the Map of Human Knowledge