Total Quality Control
QC Background Development in Japan Tools & Techniques TQC Development Six Sigma
QC Background
The notion of Quality Control grew with mass production. Rather than each product being a unique, custom-made item, perhaps hand-built from the raw materials through to completion by one person, the objective became to make each item alike. Consistency ensured that consumers got what they were supposed to get - no more, no less, while the producer's commercial equations balanced.
Inevitably, there would be variation: if different people, different teams, different factories were all aiming to replicate the same thing, individual influences would all creep in. Variation is to be expected, so the idea of 'quality control' was to maintain variation of product within acceptable limits. This took the form of dedicated QC personnel who acted as inspectors, checking on finished product, and filtering out any outside tolerance. 
 Some paradigms developed around this:
 - makers make ...their job is producing, and quantity is the prime goal
 - QC people check. If they can't find any problems their existence may be questioned
 - the later in the process the inspection, the lesser the risk of defects being found by customers
- there is a trade off: the better the quality, the more time and effort (therefore cost) is required to achieve it
- you can't have everything - there has to be a choice!

In 1925 a new approach was pioneered in the Bell telephone Laboratories, USA, and the term 'Quality Assurance' was coined. It introduced SQC (Statistical Quality Control) and started to recognise many of those humanistic and organisational features subsequently associated with TQC. The Depression stifled these initiatives, and they were not seriously re-visited until driven by wartime needs.

 

QC Development in Japan
In the post-war period quality control was eagerly adopted in Japan. The prime focus remained on the shop floor, but a less polarised approach developed: quality was held to be everyone's responsibility, rather than the preserve of the 'quality controllers'. A team approach to solving problems was developed which became known as 'Quality Circles' - people getting their heads together to solve quality problems. 
A different way of thinking started to form on the shop floor: not only was it important to satisfy the consumer, but to satisfy each person's downstream customer. Give the customer what he wants and he can concentrate on doing his job properly. Apply this all the way down the supply chain and the ultimate consumer will be well-served.
Also, the notion that 'quality' is not simply the total of all the attributes on the finished product (achieved by hook or by crook!) but the consistency with which these are achieved. Reliability, dependability, predictability, etc. of people, processes, and plant were recognised as being as important the product itself. 'Right First Time' became a goal.
Thus the remit of Quality Circles was expanded to cover anything that had an influence on the process or product. The only constraint was that teams should only address items within their sphere of control. Methodologies and a structured approach were developed and promoted on a national basis by JUSE (Japan Union of Scientists and Engineers) following initial guidance from American mentors, notably Deming and Juran.

 

Tools & Techniques
Teams need a game plan, so the four-phase cycle was adopted as the broad TQC methodology. In the thirties, this was developed as the Shewart wheel (illustrated), after an American physicist/ statistician; when promoted by Deming in the late 1940s / early 1950s it assumed his name; and is

now commonly known in the west as the PDCA cycle, and in Japan as the kanri (management) cycle.
The content of the various phases is varied to suit the particular application - the kanri cycle takes the form:
1) Planning: Establish plans for accomplishing the purpose of business.
2) Doing: execute plans according to programs.
3) Checking: check whether or not the results satisfy the initial purpose.
4) Action: correct or eliminate the problems found in the checking stage.

Similarly, QC circles label the stages according to their particular problem solving approach - for example:

PLAN: state problem, set target, identify activity framework

DO: gather data, analyse, consider countermeasures, plan trials

CHECK: assess effectiveness of countermeasures

ACTION: apply new standard; consider further improvement

Note the term 'countermeasure' rather than 'solution'. In continuous improvement we accept that 'there is always a better way', hence the cycle, rather than a finite, linear approach.
Tools
Given a methodology, various tools and techniques were found to be useful in the gathering and analysis of data which teams used as the basis for problem solving. 
The following 'Seven QC Tools' were seen as the most useful for Quality Circles

The Check Sheet (or tally chart): 

A tool for gathering data. Anticipates possible variations / attributes and categorises them - for example into 'what', 'where', 'when'. Makes recording data, and subsequent analysis easy.

Graph

Plots variation over time

Scatter Chart

Establishes if there is a correlation between two variables

Histogram 

Shows the extent of variation of a particular variable (how heavy / light? ...hot / cold? ...long / short?..)

Pareto Diagram (or ranked frequency diagram) Segregates the 'significant few' from the 'trivial many'.

Cause & Effect Diagram (aka 'Fishbone' or 'Ishikawa' Diagram)

Used as a framework to accommodate speculation from brainstorming. What possible causes are there for this problem? Categorise them in like groups (often the 4Ms) ...then test the possible causes to see what produces the effect. 

There are many variations of Cause & Effect diagrams ...also the reverse, the Solution Effect Diagram.

Control Charts

When a capable process has been established, but setting is required to control / compensate for some variable, control charts are useful. Plotted data enables predictions to be made and manufacturing out-of-tolerance to be avoided.

Although 'The Seven Statistical Tools' or 'The Seven QC Tools' are frequently referred to, the identity of the seven often varies! 'New Tools' were developed in the 1970s, primarily for the use of management/engineers or in specialist areas: 

     - Multivariate analysis, and statistical analysis of reliability
     - Quality table or quality deployment, FMEA (Failure Mode and Effects Analysis), and FTA     (Fault Tree Analysis)

     - 7 Management Tools for QC: Affinity diagram, relation diagram, tree diagram, matrix diagram, arrow diagram, PDPC (Process Decision Program Chart) and matrix-data analysis.

 

TQC Development
Quality circles had beneficial effects not only on product quality, but also on the way people worked together and the way they managed their work environment and processes. The same approach spread from the shop floor into other areas of business - Support departments, Administration, Design, ... in fact everybody in every function that has a customer to satisfy! The 'Total' was added to 'Quality Control' to signify the company-wide approach. (sometimes 'CWQC')
Given a basic grounding in QC practices to assure product quality, develop the people and strengthen the business, other areas of specialisation emerged - TPM (Total Productive Maintenance)  to assure facility integrity, hence delivery; JIT on the optimisation of material flow to produce at minimum cost.
With quality and delivery taken as 'givens', efforts focus on draining
waste, wherever it may be found in the system, to reduce cost.

Western adoption of this approach suffered some compromises: 'Quality Control' had bad connotations - so became called 'Quality Assurance'; "We'll tell people they're responsible for their own quality, but not develop their understanding ...and we'll only focus on the shop floor!" QC Circles were introduced piecemeal, often with emphasis on results alone, without understanding the processes through which results are achieved. Manufacturing at minimum cost has been sought without undertaking the intermediate steps.
Where TQC has been effective, it has often been under the 'Total Quality Management' banner (avoiding the 'control' word!). Such companies have been prepared to invest time and money in developing their people, and the organisational framework within which they work, in order to establish paradigms for the 21st Century - then tackling the practical aspects.
In Japan, no conflict is seen between the various philosophies, tools, techniques, systems, etc.: they are integrated into the way the company, as a whole, conducts its business. Having applied TQC, TPM, JIT and all, and tuned them to their own requirements, they are often labelled 'The XXX Production System'. Genba Kanri is the generic term for this.

 

'Six Sigma'

Current efforts in the spirit of TQC are being promoted under the banner of 'Six Sigma'. Sigma, the 18th letter of the Greek alphabet, is often used to denote the standard deviation, a statistical measure of distribution. It is a measure of how consistent individual pieces are 

6s
from one particular point of view. Using data from a sample, the standard deviation can be calculated so as to determine the scatter of the 'population' as a whole about the mean.
It is self-evident that in any distribution 50% of the population will be more than average; 50% will be lower. The standard deviation tells us more: it is known that 68.26% of the population (from which the sample is derived) lies within +/-1s of the mean; 95.44% within +/-2s; 99.73% within +/- 3s.
Traditionally, this last figure has been known as 'six sigma', or '6s', the benchmark of acceptability for a process (individual machines +/-4s: 99.994%) when assessing capability. It accepts that one part in 400
will be outside tolerance. Even +/-4s accepts that 6,000 parts per million will be outside tolerance, and will have to be 'inspected out' before reaching the customer - or the costs of customer dissatisfaction paid! Much filtering can be achieved through human or mechanical means - but this still costs money.
Nowadays, in World Class companies, no more than 20 ppm defects are acceptable, and the objective of zero defects is being aimed at through achieving process capabilities of +/-6s. No part of the process is being ignored: Design, Procurement, Manufacturing, Logistics, Service ...every aspect of the business joins in the effort. Whilst consistent with the ideals of JIT, TPM, TQC and all, in achieving high quality and rapid delivery at minimum cost, the banner of Six Sigma has been adopted for the approach by some, most notably Motorola and General Electric in the USA.             
Six Sigma can be as effective as any other rigorously-applied improvement theme. Experience shows that the three main factors to ensure success are:

1) Top management support

2) Process management support  

3) Selection of projects where the cost of poor quality has hitherto been invisible

Although most companies applying it would not claim six sigma capability, it provides a very clear objective.

In recent years, interpretations of Six Sigma have diversified, with the result that we see banners such as 'Six Sigma Plus' and 'Seven Sigma' appearing. Companies originally embarking on the statistical approach have come to recognise the value of:

1. The 'soft' people factors. Relationships,  will, buy-in, ownership, transactional aspects, etc.

2. Establishing the basics. There is no point in making sophisticated measurement and analysis to  unstable processes. 

In essence, many Six Sigma programmes are now retrospectively dipping into the Genba Kanri toolbox!

RETURN TO MAIN PAGE

The web site dedicated to improving manufacturing competitiveness  

TQC in Japan

genba-kanri.com