Improving Completeness of Measurement Systems for Monitoring Software Development Workflows

• Monitoring and controlling of software projects executed according to Lean or Agile software development requires, in principle, continuous measurement and use of indicators to monitor development areas and/or identify problem areas. Indicators are specific kind of measures with associated analysis models and decision criteria (ISO/IEC 15939). Indicating/highlighting problems in processes, is often used in Lean SW development and despite obvious benefits there are also dangers with improper use of indicators – using inadequate indicators can mislead the stakeholders towards sub-optimizations/erroneous decisions. In this paper we present a method for assessing completeness of information provided by measurement systems (i.e. both measures and indicators). The method is a variation of value stream mapping modeling with an application in a software development organization in the telecom domain. We also show the use of this method at one of the units of Ericsson where it was applied to provide stakeholders with an early warning system about upcoming problems with software quality.

Subject headings

NATURVETENSKAP  -- Data- och informationsvetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences (hsv//eng)

Keyword

software metrics

Publication and Content Type

ref (subject category)

kon (subject category)