Computer-Aided Systems Engineering Methodology Supportand Its Effect on the Output of Structured Analysis

  • Authors:

  • David Jankowski

  • Affiliations:

  • Center for High Technology Management, California State University, San Marcos, San Marcos, CA 92096-0001. E-mail: doctor@csusm.edu

  • Venue:
  • Empirical Software Engineering
  • Year:
  • 1997

Quantified Score

Hi-index 0.02

Visualization

Abstract

The information systems development literature indicatesthat there is no conclusive, empirical evidence that CASE improvesthe quality of system specifications or the resulting informationsystems. One role of a CASE tool is to serve as a methodologycompanion—to assist an analyst in the creation of documentationpassed to succeeding phases of the life cycle, and to guide theanalyst through a particular systems development methodology.A framework for comparing the level of methodology support providedby a CASE tool is proposed and applied to 27 structured analysismethodology rules. The framework contains seven levels of ruleenforcement, ranging from real-time enforcement of a rule tothe absence of enforcement for a rule. The methodology enforcementframework is then applied to two popular, commercial CASE tools.Each CASE tool was used by eight project teams over a two monthperiod to construct a functional specification for a hotel informationsystem. The goal of the study was to examine the influence onthe functional specification of the level of methodology supportprovided by the CASE tool for each structured analysis methodologyrule. Methodology errors in the system specification were notedfor each structured analysis methodology rule. An analysis ofthe frequency of errors indicates that internal consistency rulesare easily adhered to regardless of the level of methodologysupport provided by the CASE tool, while hierarchical consistencyrules are adhered to more frequently in the presence of rigorousmethodology support.