The Emergence of Change at the Systems Engineering and Software Design Interface
An Investigation of Impact Analysis

Year: 2008
Author: KILPINEN, MALIA SOFIA
Supervisor: Clarkson, P. John
Institution: University of Cambridge
Page(s): 270

Abstract

Systems engineers and software designers collaborate throughout product development to delineate and fulfil software requirements. Given the flexibility of software, the effects of software requirement and detail design changes must be understood in order to reduce the risk of ripples of further, unanticipated modifications. As such, these designers both perform impact analysis to estimate the consequences of design changes. This dissertation examines the impact analysis performed by systems engineers and software designers as prescribed by literature and as practised within industry and then nvestigates the implications of impact analysis improvement on the design process.
Literature only mentions the application of impact analysis in practice and often conflicts in describing how impact analysis should be performed. In turn, two extensive empirical studies within two global companies explore and characterise the challenges of implementing prescribed impact analysis techniques. Through the discussion of specific changes and the associated impact analysis applied, interviewees provide suggestions for practical improvement strategies to address these challenges. The effect of impact analysis improvement on the design process through such strategies is then further investigated through modelling and simulation using system dynamics. The simulation of a collaborating company’s design process shows that the time to complete design work can be reduced by about 60% through marginally improving IA and also is about twice as sensitive to IA improvement than other improvement strategies (e.g. optimising design task scheduling or improving requirements management). As a result, heuristics for impact analysis improvement are developed. These heuristics indicate that the quality of impact analysis results and when the analysis is performed during change processes can significantly influence the rate of design project progress. The evaluation of the heuristics by the industry collaborators suggests the utility of strategically reforming impact analysis practice to provide for software design process improvement. Consequently, this dissertation focuses on the effects of impact analysis improvement and argues for elevation of the importance of impact analysis within software design processes.