Abstracts and contact information are available for the following
21 Doctoral Thesis Research Topics:
If your thesis work falls within the domain of requirements engineering, send your thesis title, a 200 word abstract and your contact information to Annie I. Antón (aianton@eos.ncsu.edu).
Abstract: The objective of this work is to develop, validate and assess a goal-based method, which provides procedural support, for the identification, elaboration, refinement, and organization of goals in the specification of software-based information systems. The two main contributions of this work to current goal-refinement requirements methods are:
Existing goal-based methods stress the need to characterize, categorize, decompose and structure goals. However, they fail to offer strategies for the initial identification and construction of goals. It is assumed that the goals already exist in some artifact. However, if goals have not been previously specified, how does one go about specifying them, and how does one know when all of the goals have been completely specified? Strategies are needed to resolve these issues. Our research has enabled us to develop a method which offers a straightforward, methodical approach to identifying system (and enterprise) goals and requirements. It suggests goal identification and refinement strategies and techniques by including a set of guidelines and recurring questions types. Use of this method will produce a software specification of the functional requirements in the form of goal schemas showing behaviors in terms of goals, their relationships, constraints, and obstacles, as well as agent responsibilities.
Email: aianton@eos.ncsu.edu
Thesis Co-Advisors: Dr.
Peter A. Freeman and Dr.
Colin Potts
Graduation Date: June 1997
Thesis: Available from http://www.csc.ncsu.edu/faculty/anton/pubs/thesis/thesis.html
Abstract: The KAOS approach, on which the thesis relies, allows requirements to be captured rigorously. A KAOS specification models a domain application and its environment in terms of goals, constraints, actions, entities, relationships, agents and events, and the relationships over them. Our main contribution has been to develop techniques to help analysts specify hierarchies of KAOS goals that are complete, that are proved correct and that integrate alternatives. To this aim, a library of goal reduction patterns has been designed; a reduction pattern links abstract assertions through a reduction relationship. The reduction patterns are related with each other through weakening/strengthening relationships. Domain-dependent and independent reduction tactics have been introduced in addition for guiding the selection of reduction patterns. The library of reduction patterns has been used to produce operationalization patterns that allow goals to be operationalized into constraints assignable to individual agents. It has also been used to develop complete, abstract and reusable KAOS models based on the concept of resources requested by agents.
Process modelling aims to capture the activities ruling the construction of software and thus in particular the activities ruling the construction of a first specification (requirements model). The ICARUS process meta-model, on the definition of which we have worked, has been used to formalize the KAOS goal acquisition process. A library of development operators for goal acquisition has been developed and specified formally in an algebraic language with loose semantic. Given an application domain, the application of development operators during the analysis allows goals to be elicited, possibly formalized, and decomposed into a hierarchy of subgoals. Operators that reduce goals reduce them according to reduction patterns and tactics described above. The ICARUS process meta-model has been extended to allow tactics (and particularly reduction tactics) to be modelled. Other technical extensions are also proposed to improve the way processes are modelled.
Email: rd@cediti.be
Thesis Advisor: Prof.
A. van Lamsweerde
Graduation Date: June 23, 1995
Thesis: Available in postscript.
Abstract: The requirements engineering activity, which concerns the elicitation, the modelling, the analysis, and the validation the customers' needs, is recognized as a crucial activity in the software lifecycle. Specification languages supporting this activity for complex systems, like real-time systems, are expected possess two main qualities: (i) expressiveness so that requirements can be modelled in a natural way and (ii) formality allowing automated reasoning on a specification document in order to discover potential incompletenesses and/or inconsistencies. The Albert II language has been designed with these objectives in mind. In this thesis, the Albert II language is presented intuitively and formally, and is illustrated through the handling of a real-size case study. Various issues around the language are discussed, e.a. methodological issues, availability of tools, link with other phases of the software lifecycle,...
Email: pdu@info.fundp.ac.be
Thesis Advisor: Prof.
E. DUBOIS
Graduation Date: September 19, 1995
Abstract: A requirement that is difficult or impossible
to satisfy can lead to one of two extremes: (1) an unsatisfied,
impractical requirement that is too strong, or (2) a requirement
that has been unnecessarily weakened or abandoned. Traditional
software engineering methods are not well suited to addressing
problems caused by requirements that are too weak or too strong
because such methods focus on changing specifications, not requirements.
The premise of this work is that changing requirements is an important
alternative to changing specifications. My thesis is that knowledge-based
analysis of requirements and environmental constraints supports
requirements change by allowing a program to find the strongest
requirement satisfied by a specification. This dissertation makes
two contributions. First, it defines a relation called IS-STRONGER-THAN
that allows a program to compare the strength of alternative requirements.
A program called GIRAFFE uses a set of transformations based on
the IS-STRONGER- THAN relation to incrementally strengthen and
weaken requirements. Second, the dissertation describes a method
for finding general scenarios. GIRAFFE uses general scenarios
to determine when requirements changes are appropriate. GIRAFFE's
method for finding scenarios allows it to use partially-specified
initial states and find scenarios that include multiple paths
and more general object types. GIRAFFE uses the IS-STRONGER-THAN
relation and general scenarios to find the strongest requirement
satisfied by a specification. The program thus helps an analyst
change requirements and so avoid the problems of unsatisfied,
impractical requirements and unnecessarily weak requirements.
Email: bdurney@uvsc.edu
Thesis Advisor: Dr.
Steve Fickas
Graduation Date: June 12, 1994
Abstract: Telecommunications has brought about a tremendous change in the last 150 years, and is still one of the most rapidly developing technologies. Advances in mobility and quality over long dis-tances have led to a highly complex communication medium, which contains equipment of differing capabilities and age, produced by a variety of vendors, in a worldwide distributed network. Nowa-days, one of the greatest challenges that telecommunication operators face is the development and deployment of new services. Many service ideas exist; however, their implementation has persis-tently proved to be difficult. One of the most problematic tasks lies in understanding the require-ments and correctly transforming them into code. Each introduction of a new service causes fears as to how the service will interact with already existing services. Traditional approaches have failed to deliver high-quality service software; new approaches have to be found. The objective of this research was therefore to investigate the use of advanced software engineering, requirements engi-neering and artificial intelligence technologies for telecommunications service design. The outcome of this is the Requirements Assistant for Telecommunications Services (RATS), which consists of the RATS service development methodology and its implementation in the RATS tool. The meth-odology is based on a novel three-dimensional framework for requirements engineering, and pro-vides, with the help of a specialised use-case design process, a smooth transition to currently exist-ing development life cycles which employ formal methods. The guidelines of the methodology contain in a concise way the steps to be performed during service design. The RATS tool is an expert system which advises the service developer during all stages of the development process, at different levels of abstraction. It provides requirements management facilities, traceability, impact analysis and document generation. Some of the features are illustrated using examples from the Universal Personal Telecommunication (UPT) service.
Email: eberlein@enel.ucalgary.ca
Thesis Advisors: Fred Halsall
Graduation Date: January 6, 1998
Thesis: Available
from: http://www.enel.ucalgary.ca/People/eberlein/publications/thesis.zip
Abstract: Presents an information process model for design, and an associated methodology for automating the evaluation of partial designs using black-box testing techniques. The methodology generates black-box evaluation testsets using a novel semantic graph data model which maintains the relationships between design and requirements data. The testset is used to simulate the design module and the related requirements, thus generating a set of expected and a set of actual results suitable for analysis.
Email: frezza@ee.pitt.ed
Thesis Advisors: Steven P. Levitan
Graduation Date: August 1995
Abstract: The thesis of this work is that computer-based systems design can benefit from lessons learned in other disciplines, to include how to work with users, objects, commercial components, and previous projects. The importance of improving requirements engineering approaches has been recognized as a critical problem for the 1990s, as reflected by the start of the International Symposium on Requirements Engineering (RE) in 1993. Selection of meta questions for requirements engineering comparison across disciplines is based on a proposed unified requirements engineering process model. The research approach also involves case study analysis of successes and failures in six disciplines. Recommended requirements engineering process patterns will be documented and used to extend the requirement exploration framework of Gause and Weinberg (1989).
Email: marilyn.gaska@lmco.com
Thesis Advisor: Professor Donald C. Gause
Graduation Date: 1997
Abstract: This thesis is concerned with the problem of requirements traceability (RT). In the context of systems and software engineering, RT refers to the ability to describe and follow the life of a requirement in both a forwards and backwards direction. Despite being highlighted as an area in need of improvement, being introduced as a mandatory activity in development standards, and a recent surge in dedicated support, RT remains cited as a dominating problem by industry.
In recognising that RT problems are multifaceted in nature, and that there is unlikely to be an all-encompassing solution, we attribute their persistence to the lack of any thorough problem analysis. Prevailing approaches to RT are inadequate because, being primarily solution-driven, they fail to account for the underlying issues which perpetuate RT problems in practice.
Based upon findings from an empirical analysis of the RT problem, we suggest that the crux of the RT problem revolves around the paucity of information about the human sources of requirements and requirements-related information. We found that practitioners predominantly claim to have experienced RT problems when, being unable to retrieve information they want about requirements, they have further been unable to identify those in a position to supply it. We therefore propose that, by making the social structure that gave rise to requirements explicit and open to interrogation, a suitable basis would be provided for tackling this central issue.
To provide such a basis, this thesis presents an approach through which the contribution structure underlying requirements artifacts can be defined and maintained. This enables conventional forms of artifact-based RT to be extended with contribution structures to accommodate the diverse forms of personnel-based RT that practitioners were found to need. Experience with a prototype tool and a case study further demonstrates how the approach provides both a feasible and practical solution to the crux of the RT problem and, in so doing, corroborates the position of this thesis.
Email: olly@soi.city.ac.uk
Thesis Advisor:
Prof. Anthony Finkelstein
Graduation Date: 1995
Abstract: Available on his homepage.
Email: haumer@informatik.rwth-aachen.de
Thesis Advisor: Prof.
Dr. Matthias Jarke
Expected Graduation Date: June 2000
Abstract: A study of the quality of specifications and their correlation with project results, compared with other aspects on the development process .
Email: hoek@dsv.su.se
Thesis Advisor:
Expected Graduation Date:
Abstract: The objective of this research is to provide a framework for reusing requirement models by analogical reasoning. In particular, the reuse of goal reduction and constraint operationalization requirements will be studied in the context of Kaos, a requirements acquisition framework previously defined.
Email: phm@cediti.be
Thesis Advisor: Prof. Axel van Lamsweerde
Expected Graduation Date:
Abstract: The objective of this thesis is to present
a methodology and a prototype system to support requirements traceability
(RT) during the development of an information system and the ERA
process and product. Requirements Traceability has been defined
as the process of describing and following the lifecycle of a
requirement in both forwards and backwards direction. Our intention
is to trace requirements back to their origin as well as follow
them to the requirement specifications and their subsequent deployment
into an information system. This thesis describes the SERT (Supporting
Enterprise Requirements Traceability) methodology developed to
assist the process of tracing requirements to their origin by:
The SERT process is based on:
The SERT approach is driven by the view that in order to facilitate
change management, designers need to understand and model the
environment within which requirements are established. This view
is the by-product of the need to capture and represent the situation
from which requirements are generated. Situations represent the
context in which changes and problems occur. Only by explicitly
representing them is it possible to reason about the changes in
an enterprise.
Email: mokitimi@sna.co.umist.ac.uk
Thesis Advisor: Prof. Peri Loucopoulos
Expected Graduation Date: July 1998
Abstract: My thesis will present a framework for representing and dealing with performance requirements during the development of information systems. It helps a developer organise knowledge about performance, development and domains, in order to produce solutions which address the particular needs of the particular system being developed. It provides a structured, goal-oriented process for stating, refining and evaluating performance requirements during system development It adapts and integrates work done on non-functional (quality) requirememnts, performance of information systems, and the design and implementation of object-based databases. To examine the applicability of the framework, studies have been conducted on a variety of information systems, and the studies are being reviewed by industrial domain experts. In addition, a prototype tool is under development, to aid use of the framework.
Email: nixon@ai.toronto.edu
Thesis Advisor:
Prof. John Mylopoulos
Graduation Date: mid-1995
Topics: Requirement Engineering, Viewpoints, Perspectives, Inter-perspective relationships, Distributed Artificial Intelligence, Temporal logic.
Abstract: The requirement engineering activity (for composite systems) implies the existence of several viewpoints on the software to be designed. The specifications will be issued from those different perspectives which complement but also contradict each other. Notably each perspective may be specified using a particular language. Consequently the central problem is to express links between those perspectives, i.e. the specification fragments.My work outlines a multi-agent system for expressing the inter-perspective relationships. Each perspective is associated with an agent. The distributed artificial intelligence (DAI) framework allows to represent links between perspectives and to represent the dynamics of those links. Notably, my multi-agent system can represent temporal constraints on the dynamics of the relationships between the agents and then between the perspectives. To this end I characterize the inter-perspective relationships. Next a logical approach of these relationships is defined.
Email: perussel@cix.cict.fr
Thesis Advisors: Pierre-Jean Charrel and Bernard Rothenburger
Graduation Date: 1997
Abstract: During system testing, determining if the observed behaviour of a real-time system is consistent with its requirements specification can be difficult. I propose that a system to check the behaviour against the specification, a monitor, be automatically derived from the requirements documentation. The monitor would model the system requirements as a modified finite state automaton in which the states represent equivalence classes of system histories and transitions are labelled with predicates such that it accepts only executions representing acceptable system behaviour. Investigation into the design of such a monitor, and the process for automatically generating it from reviewable requirements documentation is on-going.
Email: peters@mcmaster.ca
Thesis Advisor:
Dr. David L. Parnas
Graduation Date: Spring '98
Abstract: A justification of a proposed software-based system is an argument given in order to increase confidence that the system, once implemented, will be successful in its environment. The work reported in this thesis is motivated by two problems with formal justification in the requirements engineering process:
- the need to reason about models of the proposed system, assumptions concerning its environment, and the joint properties, or requirements, which the system should achieve with its environment
- the need to carry out the re-justification of proposals as changes are made to the various models involved
Solutions to both of these problems are found in the application of the assumption/commitment approach to modelling, originally developed for reasoning about interference in concurrent systems. The approach is applied in this thesis through the provision of a highly modular style of object description, based on temporal logic, along with the use of a compositional reasoning method. The work is evaluated through a case study of a requirements engineering process.
Email: S.R.Smith@durham.ac.uk
Thesis Advisor: Dr. D.G. Bridge
Graduation Date: May, 1996
Thesis: Available from the Technical
Report Site of the Department
of Computer Science at the University
of York.
Abstract: The objective of this work is to develop and validate a method to guide the definition of requirements of a software-based system that provides support to structural, social, political and symbolic dimensions of work. The main contributions of this thesis to current methods for requirements' definition are:
Other methods assume that work reality in organizations has a subjective nature and stress the importance of considering the organizational culture that provides the foundation for the work design. However, they do not explicitly relate the assumed paradigm for knowledge creation with the structuring of the process of requirements' definition. They also fail to provide guidelines for the characterization of the structural, social, political and symbolic dimensions of work and for the definition of requirements of a system that supports/changes the studied aspects.
Email: isantos@mail.telepac.pt
Thesis Advisor: Prof. João Alvaro Carvalho, Prof
Aníbal Alves and Daniel
M. Berry
Graduation Date: September 2000
Abstract: Normally, in a contracted software development, the requirements are captured in a number of documents, including a Software Requirements Specification document (SRS), a prototype of the system user interface, and a document called Occupation Analysis Document (OAD) written by the human engineering group; this last document resembles a draft version of a tutorial and user's manual. In a typical requirements engineering situation these days, a prototype may be developed in order to answer questions, primarily about user interfaces. Once the prototype is developed and agreed upon, how is the information that it contains transmitted to the programmers? Part of the information appears explicitly in the SRS, part is expressed indirectly by other statements in the SRS, part appears in the OAD, and part does not appear anywhere even though it is known, understood, and agreed upon by all the people involved in the project, by virtue of their having worked together to produce the prototype. Typically, the information implicit in the prototype is left in the prototype simply because a lot of information in it is not known until the prototype is used to answer questions. The proposed research is an attempt to get a good answer to this question.
Email: ravid@techunix.technion.ac.il
Thesis Advisor: Daniel
M. Berry
Graduation Date: who knows, but > 1997!
Thesis: More information avaialable at: http://www.cs.technion.ac.il/~dberry/reqs.prototype.html
Abstract: The aim of our research is to investigate whether a requirements engineer can actively influence the maintainability of an Information System(IS), and if so, to what extent and through which mechanisms. We will call these mechanisms `maintainability factors'. Our approach is of a highly empirical nature and can be divided up in two phases. In the first phase, a list of potential maintainability factors is constructed. This list is based both on literature and experiments generating alternative correct conceptual models for a set of requirements. In the second and most important phase, experiments are used to evaluate to which degree each factor influences maintainability. Ideally, the results of these experiments will enable us to formulate guidelines on how the requirements engineer can `build' maintainability into a future IS.
Email: verelst@ruca.ua.ac.be
Thesis Advisors: Prof. Dr. G. Thiers
Expected Graduation Date: 1999
Abstract: Available on her homepage.
Email: lorna@isi.edu
Thesis Advisors:
Dr. W. Lewis Johnson
Graduation Date: 1995
Abstract: This thesis defines and evaluates a logic-based framework for modelling, reasoning about, and managing changing software requirements, while paying particular attention to the issues of completeness and consistency.
Requirements modelling is considered as the construction of logical theories of some nonmonotonic logic. Requirements evolution then involves the mapping of one such theory to another. Recognising the need for nonmonotonic reasoning approach in the process of requirements evolution and establishing the connection between this process and that of belief revision, this thesis demonstrates the utility of default reasoning (a specific form of nonmonotonic reasoning) in the requirements evolution process. Exploiting the deductive power of the theory of belief revision and default reasoning, both originally developed within the discipline of artificial intelligence, this thesis defines a formal model of changing requirements by developing two operators for the revision and contraction of requirements models (represented as default theories). This model also provides a simple yet complete framework for describing the requirements engineering process itself.
The research demonstrates, through working examples, how the representation of requirements models as default theories in the proposed framework can formally and accurately capture intuitive aspects of managing changes to requirements models, while maintaining their completeness and consistency. As a proof of concept, the architecture of a prototype environment, Computer-Assisted Requirements Evolution Toolset (CARET), based on the proposed formal framework is sketched and the reasoning component of such an environment is implemented.
The major contributions of this thesis are a formal framework for the effective management of changing software requirements and new methods for treating completeness and handling inconsistency in evolving models of requirements. This research thus offers a rigorous approach to reasoning about requirements evolution and an important starting point for defining semantically well-founded methods and tools for the effective management of changing software requirements.
Email: didar@ics.mq.edu.au
Thesis Advisors: Professor Ray Offen, Dr. Pavlos Peppas
and Dr. Aditya Ghose
Graduation Date: June 1999