Updated 2015-02-16


Home Page


Software test

Requirements   engineering

People issues

Iterative development

Improvement models

Change strategies

Experience-based   improvement
  Problem diagnosis
  Diagnosis applied at B&K
  Seminars / Presentations
  Reports / Papers

Defect analysis

Configuration management

Contact details


Reports and Papers

1. BOOTSTRAP 3.0: Software Process Assessment Methodology (1998), Bicego A., M. Khurana, P. Kuvaja, Proceedings of the SQM'98.

2. Capability Maturity Model for Software, Version 1.1, 93-TR-024 (1993), Paulk M.C., B. Curtis, M.B. Chrissis, and C.V. Weber, Software Engineering Institute, Pittsburgh, Pennsylvania.

3. Experience-Based Approaches to Process Improvement (2000), Vinter O. - Proceedings of the Thirteenth International Software Quality Week, Software Research, San Francisco, USA. (Presentation slides)

4. Improving Software Organizations: From Principles to Practice (2002), Mathiassen L. e.a. eds. Addison-Wesley Crystal Series. Foreword by Bill Curtis.

5. Using Defect Analysis as an Approach to Software Process Improvement (2001), Vinter O. - Otto Vinter, Software Engineering Mentor, 2630 Taastrup.

Experience-based Software Process Improvement

At Brüel & Kjær I was responsible for a number of software process improvement projects. Focus in these was how to make software development processes more effective with the aim to reduce the number of errors and improve the predictability in software development.

The improvement projects were all based on the knowledge and experiences already present in the organisation. The improvement results were validated through BOOTSTRAP [1] assessments. The principles in this experience-based approach to software process improvement will be described below.

Problem Diagnosis

The approach that most companies choose to base their software process improvements (SPI) on is to adopt one of the well-known models of software maturity (BOOTSTRAP [1] or CMM [2]). Improvement actions follow a predefined model where a company's processes are improved on a broad scale, with the aim to move the company to higher levels of maturity.

A fundamentally different approach to SPI is an experience-driven one, based on knowledge and experience that is already available in the organisation. The approach is based on various "problem diagnosis" techniques, which aim at identifying the most important (process) issues as perceived by the organisation. Focused improvement actions to remove these issues are then planned and implemented in close collaboration with the powerful actors in the organisation.

Examples of such problem diagnosis techniques are: Analysis of error reports (defect analysis), project progress reports, or post-mortem reports, as well as focused (self-)assessments, and structured interviews with different types of persons in the organisation (project managers, developers, testers etc.).

For companies at the lower maturity levels in the assessment-based maturity models, we have found, that experience-based improvement techniques are both simpler to perform and more effective as motivators in the execution of process improvement activities. When you compare the problem diagnosis techniques and their results with assessments we conclude the following:

  • The primary problem areas uncovered by the problem diagnosis techniques are also found by the assessment-based methods
  • When the focused improvement activities have been completed, the problem areas found previously are now removed from the lists over primary problem areas, both in problem diagnosis and in assessments

Return to top of page.

Diagnosis Applied at B&K

At Brüel & Kjær we developed and applied such an experience-based approach to SPI based on the following problem diagnosis techniques:

  • Analysis of error reports from previous projects (defect analysis). The improvement technique was applied as the central element in the software test improvement project PET and in the requirements engineering improvement project PRIDE. These projects were so-called Process Improvement Experiments funded by EU under the ESSI programme. ESSI (European System and Software Initiative) aims at improving Software Best Practice in Europe under the EU 4th Framework (ESPRIT).

  • Structured interviews with project managers. This improvement technique was applied as the central element in Brüel & Kjaer's participation in the large Danish SPI project Center for Software Process Improvement (CSPI). During this project dissemination took place of the requirements engineering techniques from the PRIDE project, as well as improvement work on different iterative development models, and configuration management.
We have found that experience-based improvement techniques are both simple and effective. They can easily be tranfered to other organisations. We have validated the improvement techniques and their results through BOOTSTRAP [1] assessments before and after the improvement activities.

A description of the approach and the results we obtained at Brüel & Kjær was presented in its first version at EuroSPI'98, later updated as a keynote at QWE'99, and in its final form as a Quick Start Tutorial at SQW´00: Experience-Based Approaches to Process Improvement [3]. The problem diagnosis techniques and our results are also described in chapters 4, 8 and 18 in: "Improving Software Organizations: From Principles to Practice" [4].

Return to top of page.

Seminars and Presentations

Performing a problem diagnosis is a process that takes time and requires experience. I can assist companies in performing defect analysis, help identify major issues in the company's processes, and initiate possible improvement actions. The defect analysis approach is very briefly described in: Using Defect Analysis as an Approach to Software Process Improvement [5].

I give a half day tutorial titled: Experience-Based Approaches to Process Improvement [3], that introduces the problem diagnosis techniques and describe the results we have achieved using these as the way to initiate and perform process improvements.

I also give presentations on this subject varying from 45 minutes to 2 hours depending on your needs.

Return to top of page.

Home Page -
Hjemmeside - mail@ottovinter.dk