Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Creating a code inspection model for simulation-based decision support

 
: Neu, H.; Hanne, T.; Münch, J.; Nickel, S.; Wirsen, A.

Portland State University; Fraunhofer USA Center for Experimental Software Engineering Maryland:
ProSim Workshop '03. Software Process Simulation Modeling
Portland, 2003
10 S. : Ill., Lit.
Software Process Simulation Modeling Workshop (ProSim) <2003, Portland/Or.>
International Conference on Software Engineering (ICSE) <25, 2003, Portland/Or.>
Englisch
Konferenzbeitrag
Fraunhofer IESE ()
decision support; discrete-event simulation; inspection; process optimization

Abstract
Due to the fact that organizations developing software face ever increasing challenges to the quality, costs, and duration of software development, process models are used to understand, document, visualize, plan, and improve the development process. Usually, these models show the static structure of the processes, while the dynamic behavior is difficult to predict. Simulation models of software development processes can help to determine the process dynamics. While most of the simulation models proposed up to now are based on the system dynamics approach, we develop a discrete-event simulation model focusing on the inspection process that allows a more detailed representation of organizational issues, products and resources. In contrast to a system dynamics approach in a discrete-event simulation model, software products (such as code components) and resources (such as developers) are explicitly represented with attributes (e.g., size, skills).
In this article we sketch the development of the model, its structure, and the results of selected experiments with the model. The model aims at supporting decision making for introducing and tailoring inspections. The question of which artifacts should be inspected and how inspection activities should be staffed will be discussed. The model considers influences of varying project and context conditions and can therefore be used as a basis for the adaptation to different development environments.

: http://publica.fraunhofer.de/dokumente/N-17112.html