CEC Theses and Dissertations

Campus Access Only

All rights reserved. This publication is intended for use solely by faculty, students, and staff of Nova Southeastern University. No part of this publication may be reproduced, distributed, or transmitted in any form or by any means, now known or later developed, including but not limited to photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the author or the publisher.

Date of Award


Document Type

Dissertation - NSU Access Only

Degree Name

Doctor of Philosophy in Computer Information Systems (DCIS)


Graduate School of Computer and Information Sciences


Frank Mitropoulos

Committee Member

Sumitra Mukherjee

Committee Member

Gregory Simco


Agile software development methodologies, such as Scrum, have gained tremendous popularity and proven successful in quickly delivering quality Functional Requirements (FRs). However, agile methodologies have not adequately identified, modeled, and linked Non-Functional Requirements (NFRs) with FRs in early development phases. Researchers agree that NFRs have been generally ignored in conventional methodologies, especially ignored in agile environments.

This dissertation develops a conceptual framework for NFR modeling in agile processes. The proposed Non-functional Requirements Modeling for Agile Processes (NORMAP) Methodology investigated the feasibility of identifying, linking, and modeling Agile Loose Cases (ALCs) with Agile Use Cases (AUCs) and Agile Choose Cases (ACCs). AUCs are newly proposed hybrid of use cases and agile user stories. ALCs are proposed—loosely—defined agile NFRs. ACCs are proposed potential solutions (operationalizations) for ALCs. A lightweight adapted version of the NFR Framework was developed including 25 important NFRs selected out of 161 for this study. Further, an enhanced risk-driven agile requirements implementation sequence (NORPLAN) was developed and visualized as a tree-like view (NORVIEW).

The NORMAP Methodology was validated through developing NORMATIC--a Java-based agile visual modeling simulation tool and two case studies. NORMATIC utilized Natural Language Processing (NLP) tools to parse requirement sentences and identify potential ALCs. The first case study utilized the Predictor Models in Software Engineering (PROMISE) dataset used in NFRs classification. NORMAP successfully parsed and classified ALCs for 529 out of 607 (87.15%) independent user requirements. The second case study utilized the European Union eProcurement System’s 26 functional requirements. NORMAP successfully parsed and classified ALCs for 50 out of 57 sentences that included possible ALCs (87.71%). Furthermore, requirements quality and project management metrics were used to calculate a risk-driven requirements implementation sequence using three priority schemes.

Results showed that Riskiest-Requirements-First priority scheme planned requirements in 17 sprints--two months earlier than the Highest-Business-Value-First scheme (21 sprints) and one month earlier than the Riskiest-Requirements-Last scheme (19 sprints). Agile communities can potentially benefit from the NORMAP Methodology by utilizing a systematic and risk-driven lightweight engineering process to visually model and plan NFRs as first-class artifacts in agile environments.

To access this thesis/dissertation you must have a valid nova.edu OR mynsu.nova.edu email address and create an account for NSUWorks.

  Contact Author

  Link to NovaCat