Quality Systems Frameworks September 6, 2004 Swami Natarajan RIT - PowerPoint PPT Presentation

Quality Systems Frameworks September 6, 2004 Swami Natarajan RIT Software Engineering

Some Major Quality Frameworks • ISO 9000 family of standards – A general international standard for organizational quality systems • Oriented towards assessment and certification • Malcolm-Baldrige assessment discipline – A set of criteria for the (US) Malcolm-Baldrige quality award • Designed to encourage and recognize excellence • SEI CMM (Capability Maturity Model) – A software-specific model for improving the maturity of software development practices • Oriented towards self-assessment and improvement • Total Quality Management (TQM) – A philosophy and practices for improving quality • Focuses on building an organizational quality culture September 6, 2004 Swami Natarajan RIT Software Engineering

Quality vs. Quality frameworks • A major point to note is that all these are about Quality Systems, and not directly about the actual quality of the product – The difference between excellence in quality control for an assembly line car and producing a handmade Rolls-Royce (work of art)! • The principle is that an organization with a culture of focusing on quality and on continuous improvement will consistently produce good output – Remember also “continually optimize achievement of multiple objectives” – that is what systems do September 6, 2004 Swami Natarajan RIT Software Engineering

“Systems don’t produce quality, People do” • Would you agree? • What does this say about the need for quality systems? September 6, 2004 Swami Natarajan RIT Software Engineering

Value of the frameworks • “Optimize across all project and organizational objectives” is too open-ended – Frameworks provide models of what needs to be addressed • Primary value from these frameworks is – Defining the specific set of areas to address – Defining specific criteria for determining whether the areas are being addressed well – Providing basic structures to ensure continuing focus • Defining appropriate processes and metrics • Mechanisms for continuous improvement, so that processes keep improving and evolving as needs change • Assessment mechanisms, to check that all this is happening September 6, 2004 Swami Natarajan RIT Software Engineering

Which framework to use? • Different frameworks address different needs – Also, there are many other frameworks, and many additions/variations to each • Organizations design their own quality management approaches (or it just evolves without design!), possibly using one or more frameworks as a starting point – Anyway, frameworks only supply goals, and suggest some ways to achieve goals – Each organization needs to adapt the framework(s) to their needs, and decide how to achieve the goals • It is common for organizations to go through multiple different assessments, for different purposes • If used well, any of the frameworks are good enough • If used poorly, none of them will help! (In fact, they will hurt) “Just another tool” September 6, 2004 Swami Natarajan RIT Software Engineering

Why do you care about all this? • General knowledge – As a software engineer, people will expect you to know a little bit about these models • Understanding the big picture helps, before we start to focus on specific metrics and techniques • Understanding the philosophy and limitations helps you to get a more balanced picture of the quality area • And yes, it will be on the exam! – But only the concepts, not the specifics of each model September 6, 2004 Swami Natarajan RIT Software Engineering

Malcolm-Baldrige • An award for companies that excel in quality management and quality achievement • Broad areas, organization-level focus • Leadership • Information and Analysis • Human resource utilization • Quality results • Strategic quality planning • Customer satisfaction • Quality assurance of products and services • Looks at: Approach, Deployment, Results • The value is in its assessment of quality focus and excellence in all aspects of an organization – Not just engineering September 6, 2004 Swami Natarajan RIT Software Engineering

Do you think it helps to have such awards? Preparing for a Malcolm-Baldrige evaluation takes a company dozens, maybe hundreds of staff months. Is it energy well-spent? September 6, 2004 Swami Natarajan RIT Software Engineering

ISO 9000 • A standard for certifying that organizations follow procedures for ensuring quality • Heavy focus on processes and evidence of compliance (documentation) • Some focus on statistical techniques and processes for improvement • ISO 9000 is about procedures – “assure minimum standards of operation” – “existence of quality systems and commitment to them” • Complementary to other quality management frameworks – limited value in itself September 6, 2004 Swami Natarajan RIT Software Engineering

What are the advantages & disadvantages of a quality system that is heavily procedures and documentation-oriented? September 6, 2004 Swami Natarajan RIT Software Engineering

TQM • Not really a framework, more a philosophy and set of practices • Very strongly oriented towards building a “quality culture” • Major points of emphasis – Human aspects: • “Quality is the responsibility of every employee” • Empowering employees to take actions to improve quality • Management commitment – Metrics & analysis – Continuous improvement of process & quality – Customer needs, customer satisfaction • Most successful quality management programs incorporate all these aspects September 6, 2004 Swami Natarajan RIT Software Engineering

The culture approach is not really about “compliance” at all It is about building an environment where focusing on project and organizational objectives (“Q”) is part of what it means to “get work done” and to “deliver” • Is such an approach good? Is it workable? • What are the barriers? September 6, 2004 Swami Natarajan RIT Software Engineering

CMM • Created by the SEI (Software Engg Institute) to drive process improvement in software – Defines a series of “maturity levels” so that organizations can incrementally put software quality systems in place – Self-assessment – the idea is to help organizations create the most appropriate quality system for themselves • Focused (relatively) narrowly on the software development process • Defines a set of “Key process areas” and goals for each area • There is now also a CMMI, that expands CMM to cover “system engineering” aspects – Useful when project scope includes system design September 6, 2004 Swami Natarajan RIT Software Engineering

Frameworks as Knowledge Bases • (Look at the CMM 1.1 doc on the course website) • It is really a knowledge base – “What areas do we need to address if we want projects to be successful?” – “How do we keep everyone aware of good ways to accomplish tasks?” ( define processes and practices !) – “What are common sources of problems? What structures can we put in place to reduce the chance that they will occur?” – “What structures do we need to ensure that the organization will keep trying to improve its processes and practices?” – “How do we ensure that good processes lead to good results?” ( define appropriate metrics ) – “How we can we figure out when things aren’t working and how to fix them?” ( metrics interpretation, causal analysis, prevention ) • An organization’s quality management system is its own knowledge base of the best answers to these questions!) September 6, 2004 Swami Natarajan RIT Software Engineering

Limitations of CMM • It is heavily oriented towards “optimization of repetitive tasks”, especially at higher maturity levels – NOT appropriate if projects differ significantly from one another • There is a fundamental assumption that being highly structured is a good thing – But structure is not free! (Effort, time, flexibility) • This is where agile development methodologies came in – It might run counter to the desired organizational culture • Organizations need to decide how much maturity is right for them in each process area – The more recent “continuous” model addresses this • It only works well if the organization has an underlying commitment to quality and structure as the road to results September 6, 2004 Swami Natarajan RIT Software Engineering

Assessments • Assessments are massive exercises – Value: Feedback on what’s working, opportunities for improvement – Cross-fertilization of ideas • Problems – Easy to “create evidence for the assessment” – Passing means at best that systems are in place, not that results are superior – Assessments easily become exercises in PR (public relations) – Over-focus on “avoiding mistakes” can take energy away from excellence • It would be a mistake to read too much into the results – Being assessed at high maturity levels or receiving a quality award does NOT guarantee that the organization will be more successful or produce better products – it just means that they have structures in place to keep trying to do better September 6, 2004 Swami Natarajan RIT Software Engineering