Operations Deliver

Total Quality Management

Total quality management is an organization-wide approach to quality, aimed at long-term customer satisfaction rather than inspection at the end.

The questionWho owns quality here, and is it built into the work or checked at the end?

TQM. Grew out of total quality control (TQC) and the Japanese company-wide quality movement.

Model · Deming, Juran, Feigenbaum, Ishikawa, and others, 1950s to 1980s

Simplified view of the commonly cited TQM principles around quality. Sources list them differently.

When to use it

Defects, rework, or complaints trace back to handoffs between departments, not to the work of one department.
A client relies on final inspection, and scrap, returns, or warranty claims are a major cost.
Improvement projects keep starting and fading, and nobody at management level owns the results.
A hospital, bank, or public agency wants quality measured in terms its customers or users would recognize.

When to avoid it

One process has one measurable defect. A single improvement cycle or a focused method such as Six Sigma's DMAIC will do it with less overhead.
Management wants a training program for staff and does not expect to change how it makes decisions.
The client has no stable processes or reliable data, so there is nothing for fact-based decisions to rest on yet.
The client already runs ISO 9001 or a Baldrige-based system. A branded TQM program on top duplicates it, and the label carries baggage from the 1990s.

Overview

TQM is a management approach that makes quality the responsibility of every function and every level, with the customer’s requirements as the definition of quality. It supports a choice about where to spend effort: on preventing defects at their source across departments, or on inspecting and reworking them at the end.

There is no standard version, certification, or agreed list of practices, so ask a client what they mean by the term. Many programs run under the name in the 1990s were reported as disappointing, and the label faded as ISO 9001, Six Sigma, and lean took over its tools. The underlying questions about ownership, process, and data are still the right ones.

The commonly cited principles

Quality defined by the customer

Quality is what the customer requires and notices, not what the plant's specification says. Return reasons, complaints, and service-level misses are the starting data. Where the plant's definition and the customer's differ, the customer's wins.

Management responsibility

Executives own quality through funding, training, goals, and their own attention. The Navy's early version stated this directly. Without it the rest tends to be left to a quality department.

Process approach

Defects are traced to the process that produced them, which often sits upstream and in another department. Improvement work then goes to the process and not to blaming the last person who handled the item.

Everyone involved

People who do the work are expected to find and fix problems, usually through cross-functional teams and standing improvement groups. This only works if they have time, training, and authority to change something.

Fact-based decisions

Decisions rest on data, using basic tools such as check sheets, Pareto charts, and cause-and-effect diagrams. The principle is easy to endorse and hard to carry out when the data are late or disputed.

Continual improvement

Improvement is a permanent routine, not a project with an end date. Teams normally run it through a cycle of plan, do, check, and act. Sources list these principles differently, and some add supplier relationships, systems thinking, or communication.

What it is not

A specific method with defined steps. There is no standard TQM, and programs under the name varied widely. Ask what the client actually means.
The same as Six Sigma or ISO 9001. ISO 9001 is a certifiable management system standard, and Six Sigma is a project method with a strong statistical core. TQM was broader and looser, and those two absorbed much of its toolkit.
A quality department. The idea was that every function owns quality, and moving the job to one team brings back inspection at the end.

The steps

Define quality in the customer's terms

Collect return reasons, complaints, and late or incomplete deliveries, and take the customer's categories as given. What comes out is a short list of measures that the customer would recognize.
Trace the cost of poor quality to its source

Sort scrap, rework, returns, and claims by where the defect began, not where it was caught. The usual finding is that a few upstream causes sit in departments that never see the cost.
Give each cross-functional cause an executive owner

A cause that crosses departments needs someone with authority over all of them. Name one person per cause and put the item on their regular review.
Put teams on the process, with data

Charter a small number of teams, each on one cause, with the people who touch the process and a measure from step one. Have them run improvement in short cycles and report results, not activity.
Build the routines that keep it

Update standards where a fix worked, add the measures to the management review, and decide how new people learn the changes. Check after a few months whether the results held.

In practice

Returns from the plant's largest customer are rising, and the plant's answer is more final inspection. The consultant asks for six months of return records and sorts them by where each defect started, not where it was found. Few began at final assembly. They came from drawings released with unchecked tolerances, a resin supplier switched on price, and a changeover procedure that each shift performs differently. Those causes sit in engineering, purchasing, and production, and none of those groups sees the return data.

What follows is small. One return dashboard is owned by the plant manager. Purchasing and engineering must sign off on supplier and drawing changes, and two cross-functional teams take the largest causes, using the customer's own return categories as the plant's definitions. Inspection stays in place, but it stops being the main control.

What you leave with

Quality measures expressed in the customer's categories.
Cost of poor quality sorted by where each defect began.
A named executive owner for each cross-functional cause.
A short list of chartered improvement teams, each with a measure.
A review routine that puts the measures in front of management.

Failure points

A program in place of management change

Training, posters, and teams appear, but executives keep making decisions the old way. Teams find cross-department causes, nobody with authority acts on them, and participation falls off. ASQ lists lack of management commitment first among common difficulties.

Quality owned by one department

A quality group is made responsible, and engineering, purchasing, and operations carry on as before. The result is inspection and audit under a new name.

Measurement without action

Teams collect many measures and chart them, but no one is assigned to act when a number moves. Reports grow, and defect causes stay where they were.

Certificates as the goal

Effort goes into winning an award or passing a certification audit, and the documentation is made to fit the audit. The customer sees no change in what they receive.

Payback expected in a quarter

Fixing upstream causes takes months, and short-term profit targets push the work aside. The program is cut before the first cause is closed and is then reported as a failure of the approach.

The failures above are usually Leverage & IP: it was used once, and nothing was left that makes the next engagement easier.

Origin

TQM is a body of practice and not one author's model. W. Edwards Deming began teaching statistical quality control to Japanese engineers and executives in 1950. Joseph Juran taught managers how to control and improve quality, Armand Feigenbaum published Total Quality Control in 1951, and Kaoru Ishikawa's writing and quality-circle work shaped the Japanese approach. The origin of the term Total Quality Management is disputed. It is often credited to the United States Navy, which used it in 1985, and a British government quality campaign in 1983 may have used it earlier.

The Malcolm Baldrige National Quality Award was created in the United States in 1987, and the first ISO 9000 standards were published the same year. Neither the term nor the approach has a single owner, a standard, or a certification, and there is no trademark. The Navy later renamed its own effort Total Quality Leadership.