Quality Management

Adaptive Quality Management

First I would like to say thank you for having time, now, to look at quality. Quality is about saving time and money tomorrow and forever; but at the expense of time today and I appreciate how pressing today is.

The directives of Adaptive Quality Management are;

• to introduce disciplines that lead to improvements in the quality of work, and
• to counteract freedoms, that lead to work degradation, but also.
• to counteract disciplines, that lead to degradation in the quality of work, and
• to introduce freedoms, that lead to work improvement.

The balance between disciplines and freedoms is a delicate one and going too far in either direction can lead to problems in quality, efficiency and the pleasure in work that we all hope for.

Adaptive Quality Management

A Common Language

In the Adaptive Quality Management System (written documents called) standards form a common language in which certain routine activities of the organisation can be;

• remembered,
• communicated,
• recorded and
• improved.

Standards define sets of terms that must or should be met. Standards can control the way a routine is done or control the condition of some object through a checking or test routine performed upon it.

The people who are responsible for performing the routine activities of a standard are called the performers of the standard. The person who is responsible for appointing performers and improving the routines contained in the standard, is called the supervisor of the standard.

Adaptive Quality Management is not intended to regulate every task by a standard nor turn every task into a routine, nor to turn your organisation into a routine organisation rather than an innovative and adaptive one.

W E Deming says "We must use standards as the liberator, that relegates the problems, that have already been solved to the field of routine, and leaves the creative faculties free, for the problems that are still unsolved".

Guided by Effectiveness

Adaptive Quality Management has many applications and it will, with your participation, evolve to fit your organisation through people determining what is, or is not effective.

• AQM recognises - that not all processes of the organisation are routine and that not all routines of the organisation should be turned into standards. Standards are only needed where they lead to improvement.
• AQM is focussed - in that people only need to understand the standards that apply to their own work rather than having a huge quality manual, that all employees are expected to read, and most forget. People find it a lot easier to grasp standards which they help create in relation to their own work.
• AQM is flexible - in that no rules are in stone, but where rules have to be broken they must always be subsequently reviewed and probably changed or removed from the standard's routines. This is adaptation.

Reviews pick up on defective items that have managed to get past the standards. Again leading to changes in the standards and possibly new rules, to avoid repeating the defects. This is adaptation.

Use

Standards can be used to embody, the requirements of ISO9000, legal requirements, and the organisations routine activities such as assembly, inspection and testing.

Standards can take you through the five levels of maturity of the Capability Maturity Model which has heavily influenced other quality standards.

Learning and Application

The use of a standard should not require more effort than it saves;

• Most of the effort associated with using standards comes at the beginning, writing the standard and then learning to work to that standard. The cost of this effort is fairly clear and there is a tendency to avoid it, to save time and money now.
• Most of the effort associated with not using standards for routine activities comes at the end, usually because the necessary routine hasn't been performed correctly, because it was not remembered or not communicated properly. The possible cost of these mistakes is not always clear and there is a tendency to ignore them, because they aren't now.

What Is Quality?

Dr David Garvin of Harvard Business School (in 1987) Defined quality from the client's point of view. He splits the client's perception of quality into eight dimensions;

• Performance - The primary operating characteristic of the product or service.
• Features - The secondary characteristics of the product or service.
• Reliability - The frequency with which the product or service fails.
• Conformance - The match with specification or pre-established standards.
• Durability - Product life.
• Serviceability - The speed, courtesy and competence of repairs.
• Aesthetics - Fit and finish.
• Brand - Reputation of supplier.

Quality Isn't a Bolt On

Quality isn't a bolt on, it isn't the case that you do everything the way you did it before and then just do this extra bit to keep your boss, your customer or the government happy. Quality is part of the way we do things better. You must make quality work for you, not against you.

Sources

The Travel Terminal software on which NAL started was an unreliable beast written by a collection of four of us including myself and Patrick in the space of a couple of months. I was not used to this kind of thing as my previous programming work was reliable. The following two years saw me re-writing the whole thing and so software quality was an issue for me as I discovered the good habits that make good ‘C’ code. I was on the look out for anything about software quality. I grabbed what I could on the subject of safety critical software.

I saw a copy of DO178 "Software Considerations in airborne systems and equipment certification" and this to some extent answered the question as to how aircraft software was written but using such an approach would be far to slow for us.

On 9th June 97 there was a free software quality presentation as part of SCATE a European initiative and it was at this that I heard about the Capability Maturity Model (CMM) and its use in the development of the space shuttle software. I went straight out to buy the book "The Capability Maturity Model (Guidelines for Improving the Software Process)". It was too complex for us to apply directly but served as an inspiration to me. It is a very thorough and comprehensive system which can be taken beyond its original intended use.

I got the opportunity to seriously start work on quality in November 1999. ISO9001:2000 attempts to be all things to all men. It has merged some of the original ISO900X standards to form ISO9001:2000 which although it deals with some important issues the relevance of each part of it depends on the business you are in and there is a great deal that is simply not covered at all.

A quality system for NAL needed to exceed the requirements of ISO9001:2000, focus on the particular quality issues of NAL and take us up through the five levels of maturity defined by the CMM. I wanted a strong practical focus, a minimum of paperwork and a good introductory path.

I was inspired by a specification published by Motorola for the VME bus (a bus for use with the 680XX range of microprocessors). In it very clear language was used to define rules for developers wishing to develop hardware to connect to the VME bus.

Most of the quality systems discussed in books talked about defining processes. Yet there was considerable evidence to suggest that rules embodied in standards were a better basis for forming a quality control system. Any process could be standardised through a written standard which defined it, while any standard could not so easily be defined in terms of a process. There is a British Standards Institute not a British Process Institute. An internal standards based quality control system is a flexible system of rules that can provide disciplines and freedoms as necessary within an institution. I started work using this approach in June 2001.

Naturally I began with the standard for standards and the system grew from there. The first pilot of standards was the “network cabling standard” that resulted in tidy and orderly under floor network cabling in May 2001. The quality system has gone on to include standards for, business cases, projects, requirements, health and safety, specifications, filing, interviewing, purchasing, and supply. (There are many more specifically related to the work of NAL.)

NAL was awarded ISO9001:2000 on 6th March 2002 as a result of a tender requiring it ISO, I had six weeks to get us through, this was the usual just in time approach which I do not like. NAL has passed the assessment in each subsequent year. However many of the standards were not adhered to, to the extent that they should have been and the inspectors seemed lenient to say the least. In my view ISO9001:2000 has become more of a political standard than a practical one. In October 2003 I introduced an assessment system to ensure standards were being followed or changed as appropriate.

During the following year I pushed to get NAL to follow some kind of standardised business process and to share information across the business in a standard way. In February 2005 Individualised quality manuals were presented to supervising staff and I then handed off the quality system to an employee.

Many parts of the system are in use including the Health and Safety standards, however it seems that there is still no work being done based on any business cases, instead less formal approaches are preferred which in my view precludes NAL from benefiting from the wider markets.

Documentation for the resulting quality system is owned by me and used by NAL. I am developing this quality system further and am interested in other businesses interested in using it.

Recommended Reading

• "The Capability Maturity Model (Guidelines for Improving the Software Process)" by various people at Carnegie Mellon University Software Engineering Institute published by Addison Wesley.
• "Applying Total Quality Management to Systems Engineering" by Joe Kasser (1995) published by Artech House
• "Juran's Quality Handbook" 5^th edition by Joseph M. Juran and A. Blanton (1999) Godfrey published by McGraw-Hill
• "Out of the Crises" by W. Edwards Deming published by The MIT Press