Model-based Planning and Control for Balancing Functionality and Quality in Iterative Software Construction
Date
2013-05-28
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Software project management is the art and science of planning and controlling software projects. Software project management incorporates activities that ensure software is delivered on time, within budget, and in accordance with the requirements. Regarding challenges such as growing size, complexity, and continuous change, projects must be managed effectively; otherwise the results can be destructive for the organization.
The focus of this thesis is on a decision support system for planning and controlling an iterative software development process as a novel and evolutionary method. The main contribution lies in the design, application, and validation of a decision support system through modeling the process while integrating static defect prediction model and dynamic SD simulation models. On the other end of capabilities, this decision support system conducts the analysis to cope with balanced functionality and quality during the iterative development process. In this thesis effort was made to model certain aspects of the development processes and to integrate all available data, knowledge, and experience with a sound methodology to provide the foundation for making effective project decisions. To reach the defined objectives the major key contributions are:
(i) Modeling a real-world iterative software development process with the system dynamics modelling approach to support project management decision scenarios to control the balanced functionality and quality during the construction phase.
(ii) Designing and evaluating an attribute weighting technique using sensitivity analysis based on a neural network and employing it in an analogy-based defect prediction model.
(iii) Providing customization support for developing a CBR-based defect prediction model with a rule set defined by the source code and project information.
(iv) Refining the models through process progression, in addition to employing results from the prediction model, iteratively, to update built-in predicted parameters of the SD model.
(v) Employing a bi-criteria decision support method during planning to provide trade-off scenarios for balanced functionality and quality, considering the risk and value of features, according to the premise of providing a target quality level.
(vi) Prototype DSS development, incorporating a knowledge base and dashboard, and initial evaluation for the case study project.
Description
Keywords
Computer Science
Citation
Paikari, E. (2013). Model-based Planning and Control for Balancing Functionality and Quality in Iterative Software Construction (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/26933