Degree Type

Dissertation

Date of Award

2019

Degree Name

Doctor of Philosophy

Department

Computer Science

First Advisor

Carl K. Chang

Abstract

The establishment of smart environments, Internet of Things (IoT) and socio-technical systems has introduced many challenges to the software development process. One such main challenge is software requirements gathering which needs to address issues in a broader spectrum than traditional standalone software development. Consideration of bigger picture that includes software, its domain, the components of the domains and especially the interactions between the software and the surrounding domain components, including both human and other systems entities, is essential to gathering reliable requirements. However, most of the traditional Requirements Engineering approaches lack such comprehensive overlook of the overall view.

The main objective of this work is to introduce a human-centered approach to Requirements Engineering in order to push the boundaries of traditional concepts to be more suitable for use in the development of modern socio-technical systems in smart environments. A major challenge of introducing a human-centered approach is to effectively identify the related human factors; especially, since each individual has unique desires, goals, behaviors. Our proposed solution is to use the observational data sets generated by smart environments as a resource to extract individual's unique personalities and behaviors related to the software design. The concept of situations defined in our earlier study is used to represent the human and domain related aspects including human desires, goals, beliefs, interactions with the system and the constrained environment.

In the first stage of this work, a computational model called situation-transition structure is developed to understand the discrete factors and behavior patterns of individuals through the observational data. During the second stage, the information mined from the situation transition structure is applied to propose new human-centered approaches to support main Requirements Engineering concepts: requirements elicitation, risk management, and prioritization. The pertinence of the proposed work is illustrated through some case studies. The conclusion asserts some of the future research direction.

Copyright Owner

NimanthiAtukorala

Language

en

File Format

application/pdf

File Size

139 pages

Share

COinS