Ensuring Reliability of Information Provided by Measurement Systems

• Controlling the development of large and complex software is usually done in a quantitative manner using software metrics as the foundation for decision making. Large projects usually collect large amounts of metrics although present only a few key ones for daily project, product, and organization monitoring. The process of collecting, analyzing and presenting the key information is usually supported by automated measurement systems. Since in this process there is a transition from a lot of information (data) to a small number of indicators (metrics with decision criteria), the usual question which arises during discussions with managers is whether the stakeholders can “trust” the indicators w.r.t. the correctness of information and its timeliness. In this paper we present a method for addressing this question by assessing information quality for ISO/IEC 15939-based measurement systems. The method is realized and used in measurement systems at one of the units of Ericsson. In the paper, we also provide a short summary of the evaluation of this method through its use at Ericsson.

Subject headings

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

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

Keyword

software engineering

metrics

reliability

Publication and Content Type

ref (subject category)

kon (subject category)