No description available.

Risk management in software engineering has become a recognized project management practice but it seems that not all companies are systematically applying it. At the same time, agile methods have become popular, partly because proponents claim that agile methods implicitly reduce risks due to for example, more frequent and earlier feedback, shorter periods of development time and easier prediction of cost. Therefore, there is a need to investigate how risk management can be usable in iterative and evolutionary software development processes. This research work aims to answer this need by building an appropriate and realistic model of risk management and to support this with a tool for managing risk in agile projects. The approach can be characterized as lightweight risk management which provides the needs of risk management but limits the human effort expended. This is achieved by using software agents to carry out risk identification, risk assessment and risk monitoring, the agents making use of data collected from the project environment. This thesis describes a new solution approach supported by an Agile Risk Tool (ART) which includes a model of the risk environment and support for risk management in agile development environments. In the approach used, the project manager has to define these elements: project goals, problem scenarios, consequences, risk indicators, project environment data as well as specifying risk rules using a predefined 'Rule template'. Therefore risk can be explicitly managed in the early phase of the project, leaving the designated software agents to monitor the rest. The ART model and tool support is evaluated using two case studies, both from student projects. Evidence is therefore provided for the feasibility and applicability of the approach. Overall, the research contributes a new method for risk management in agile software processes, the necessary tool support to demonstrate the method in practice as well as providing evidence to support the efficacy of the approach. In addition, an example is given of the use of software agents as a potential means to reduce the burden of risk management in software projects.

No description available.

No description available.

This thesis examines the success and failure factors in ICT projects. The low success rate of software projects in terms of reliability, meeting due dates and working within assigned budgets is widely recognised and topical. International as well as Dutch publications and the procedures in Tarek Abdel-Hamid's work on Software Project Management/Dynamics are discussed. A SUFFI Chart (SUFFI = SUccess and Failure Factors in ICT projects) is developed. The management of a portfolio of projects is compared with the SUFFI Chart. A number of Dutch projects with which the author was directly involved are examnined to show how they compare with the factors identified from the literature. These do show considerable correlation between important SUFFI factors and project success. The portfolio consists of nine ICT projects and four ICT project audits. Projects such as SAP, RBAC, EAI, charging method, PABX, financial building adrnmistration, information retrieval, book reservations, traffic data collection, introduction of the Internet functionality and SOX, for different companies/organisations (Delft University of Technology, National Police Services Agency, KPN - Dutch Telecom Company, University of Amsterdam, government, banking). This work shows that for a successful project 4 of the 5 most important SUFFIs have to be absent.

Developing correct concurrent software is a difficult task, due to the inherently non-deterministic nature of thread interactions. Traditional testing techniques typically perform an explicit exploration of the possible program executions, and are thus inadequate for concurrent software. Symbolic verification techniques for concurrent programs are therefore desirable. Sequentialization has become one of the most promising symbolic approach for the verification of concurrent programs in recent years. However, current efficient implementations still struggle with concurrent programs that contain rare bugs, and their purposes is restricted to bug-finding. In this thesis, we advance sequentialization to provide pragmatic and scalable verification approaches for concurrent programs, aiming at finding bugs and proving correctness. Concerning finding rare bugs in concurrent programs, we present our work on optimising Lazy-CSeq sequentialization using abstract interpretation. We empirically demonstrate that this procedure, which is implemented in the tool called Lazy-CSeq+ABS, can lead to significant performance gain for very hard verification problem. Furthermore, we propose a “swarm” verification approach that can enable existing tools to find rare concurrency bugs which were previously out of reach. We implement the approach in VeriSmart, as a extension of Lazy-CSeq, and empirically demonstrate that VeriSmart can spot rare bugs considerably faster than Lazy-CSeq tool can. With regard to proving correctness, we develop a novel lazy sequentialization for unbounded concurrent programs and implement the corresponding schema in a tool named UL-CSeq based on the CSeq framework. Empirical experiments show that our new schema is efficient in both proving correctness and finding bugs on concurrency benchmarks in comparison with state-of-the-art approaches.

Recognising implemented instances of Design Patterns (DPs) in software design discloses and recovers a wealth of information about the intention of the original designers and the rationale of their design decisions. Because it is often the case that the documentation available for software systems, if any, is poor and/or obsolete, recovering such information can be of great help and importance for maintenance tasks. Since DPs are abstractly and vaguely defined, a set of software classes with exactly the same relationships as expected for a DP instance may actually be only accidentally similar. On the other hand, a set of classes with relationships that are to an extent different from the typically expected can still be a true DP instance. The deciding factor is mainly whether or not the set of classes is actually intended to solve the design problem addressed by the DP, which makes the intent a fundamental and defining characteristic of DPs. Discerning the intent of potential instances requires building complex models that cannot be built using the information known about DPs. So, a paradigm shift in DP recognition to fully machine learning based approaches is required. The problem is that there exists no accurate and sufficiently large DP datasets and it is difficult to manually construct one. Also, there is a lack of research on the feature set that should be used in DP recognition. The main aim of this thesis is to enable the paradigm shift by laying down an accurate, comprehensive and information-rich foundation of feature and data sets. To achieve this aim, a large set of features is developed to cover a wide range of design aspects, with a particular focus on the design intent. This set serves as a global feature set from which different subsets can be objectively selected for different DPs. A new and feasible approach for DP dataset construction is designed and used to construct training datasets. The feature and data sets are then used experimentally to build and train DP classifiers. The results demonstrate the accuracy and utility of the sets introduced, and show that fully machine learning based approaches do provide the appropriate and well-equipped solutions to the problem of DP recognition.

INTRODUCTION. This research delivers a new complete and prescriptive software development framework, known as the Quality Framework for Software Development (QFSD) for immediate use by software development practitioners. Whilst there are a number of existing methodologies available, and many software development standards they fail to address the complete development lifecycle. A review of current literature supports this assertion. AIMS AND OBJECTIVES. The overall aim of the research is to create a new software development framework, applying it to a substantial number of real-world software projects in two different industrial software development environments and thereby demonstrating its effectiveness. METHODS. Based on a review of the available research approaches and strategies, the researcher selected 'pragmatism' as the most suitable for this research. This selection was driven by two contributory factors. The first was that in order to conduct the research the researcher would have active participation in the majority of the research activities. The second was that the deliverables from the research should be immediately useable for the benefit of software practitioners and hence not be regarded as a theoretical framework. The approach was further refined by adopting Action Research and Case Study strategies. The research was divided in to stages each of which was executed within separate companies. The companies were very different in terms of their business areas, culture and views on quality and specifically quality of software deliverables. RESULTS. The research findings provided a strong indication that a holistic software development framework does provide an improvement in software project deliverables quality and repeatability in terms of schedules and quality. In the case of Fisher-Rosemount it enabled them to attain ISO 9000/Ticket accreditation. In addition, by providing all processes and tools in a single web based environment the adoption by software developers, project managers and senior management was very high.

During the last several decades, many kinds of population based Evolutionary Algorithms have been developed and considerable work has been devoted to computational methods which are inspired by biological evolution and natural selection, such as Evolutionary Programming and Clonal Selection Algorithm. The objective of these algorithms is not only to find suitable adjustments to the current population and hence the solution, but also to perform the process efficiently. However, a parameter setting that was optimal at the beginning of the algorithm may become unsuitable during the evolutionary process. Thus, it is preferable to automatically modify the control parameters during the runtime process. The approach required could have a bias on the distribution towards appropriate directions of the search space, thereby maintaining sufficient diversity among individuals in order to enable further ability of evolution. This thesis has offered an initial approach to developing this idea. The work starts from a clear understanding of the literature that is of direct relevance to the aforementioned motivations. The development of this approach has been built upon the basis of the fundamental and generic concepts of evolutionary algorithms. The work has exploited and benefited from a range of representative evolutionary computational mechanisms. In particular, essential issues in evolutionary algorithms such as parameter control, including the general aspects of parameter tuning and typical means for implementing parameter control have been investigated. Both the hyperheuristic algorithm and the memetic algorithm have set up a comparative work for the present development. This work has developed several novel techniques that contribute towards the advancement of evolutionary computation and optimization. One such novel approach is to construct a mixed strategy based on the concept of local fitness landscape. It exploits the concepts of fitness landscape and local fitness landscape. Both theoretical description and experimental investigation of this local fitness landscape-based mixed strategy have been provided, and systematic comparisons with alternative approaches carried out. Another contribution of this thesis is the innovative application of mixed strategy. This is facilitated by encompassing two mutation operators into the mixed strategy, which are borrowed from classical differential evolution techniques. Such an improved method has been shown to be simple and easy for implementation. The work has been utilised to deal with the problem of protein folding in bioinformatics. It is demonstrated that the proposed algorithm possesses an appropriate balance between exploration and exploitation. The use of this improved algorithm is less likely to fall into local optimal, entailing a faster and better convergence in resolving challenging realistic application problems.

This thesis presents an ethnographic process for identifying the needs for collaborative tools within an organisation and a participatory design method for matching these to existing off the shelf technologies. This process is developed through, and illustrated with, a case study looking at tools to support collaborative software development teams at Airbus. The thesis begins by exploring the background to systems development processes as well as methods for identifying needs and understanding the context of use. It then presents an in depth description of ethnographic studies carried out at Airbus in which the needs of software development teams are identified, along with high-level requirements for supporting these. These focus on ways to aid traceability of design decisions in a lightweight and flexible manner. A description of the process used to develop a participatory design workshop that sought to bridge the gap between the output of the ethnographic work and the selection of potential solutions is provided. This includes three pilot studies used to refine the method and a final workshop held with stakeholders at Airbus. The thesis then describes the evaluation of two potential solutions, the Livescribe Echo Smartpen and a Motorola Xoom tablet computer with a number of apps installed. Through evaluations it was possible to identify the Smartpen as being the most feasible technology, although some barriers to use still existed. Finally the thesis concludes with a summary of the process developed, reflections on the methods used, and an outline of the case study. Limitations and recommendations for the future are also outlined.