Background: Requirements engineering (RE) is the process of identifying, defining, documenting, and validating requirements. However, RE approaches are usually not applied to AI-based systems due to their ambiguity and is still a growing subject. Research also shows that the quality of ML-based systems is affected due to the lack of a structured RE process. Hence, there is a need to apply RE techniques in the development of ML-based systems.  Objectives: This research aims to identify the practices and challenges concerning RE techniques for AI-based systems in autonomous driving and then to identify a suitable RE approach to overcome the identified challenges. Further, the thesis aims to check the feasibility of the selected RE approach in developing a prototype AI-based classification system for road markings.  Methods: A combination of research methods has been used for this research. We apply techniques of interviews, case study, and a rapid literature review. The case company is Scania CV AB. A literature review is conducted to identify the possible RE approaches that can overcome the challenges identified through interviews and discussions with the stakeholders. A suitable RE approach, GR4ML, is found and used to develop and validate an AI-based classification system for road markings.  Results: Results indicate that RE is a challenging subject in autonomous driving. Several challenges are faced at the case company in eliciting, specifying, and validating requirements for AI-based systems, especially in autonomous driving. Results also show that the views in the GR4ML framework were suitable for the specification of system requirements and addressed most challenges identified at the case company. The iterative goal-oriented approach maintained flexibility during development. Through the system's development, it was identified that the Random Forest Classifier outperformed the Logistic Regressor and Support Vector Machine for the road markings classification.  Conclusions: The validation of the system suggests that the goal-oriented requirements engineering approach and the GR4ML framework addressed most challenges identified in eliciting, specifying, and validating requirements for AI-based systems at the case company. The views in the GR4ML framework provide a good overview of the functional and non-functional requirements of the lower-level systems in autonomous driving. However, the GR4ML framework might not be suitable for validation of higher-level AI-based systems in autonomous driving due to their complexity.

The agile principles applied to software engineering include iterative and incremental development, frequent releases of software, direct stakeholder involvement, minimal documentation and welcome changing requirements even late in the development cycle. The Agile Requirements Engineering applies the above mentioned principles to the Requirements Engineering process. Agile Requirements Engineering welcomes changing requirements even late in the development cycle. This is achieved by using the agile practice of evolutionary requirements which suggests that requirements should evolve over the course of many iterations rather than being gathered and specified upfront. Hence, changes to requirements even late in the development cycle can be accommodated easily. There is however, no real process to the agile approach to Requirements Engineering. In order to overcome this disadvantage, we propose to adapt the Requirements Generation Model (a plan-driven Requirements Engineering model) to an agile environment in order to structure the Agile Requirements Engineering process. The hybrid model named the Agile Requirements Generation Model is a soft-structured process that supports the intents of the agile approach. This model combines the best features of the Requirements Generation Model and Agile Software Development. / Master of Science

In modern software development, the gap between software requirements and implementation is not always conciliated. Typically, for Web services-based context-aware systems, reconciling this gap is even harder. The aim of this research is to explore how software reengineering can facilitate the reconciliation between requirements and implementation for the said systems. The underlying research in this thesis comprises the following three components. Firstly, the requirements recovery framework underpins the requirements elicitation approach on the proposed reengineering framework. This approach consists of three stages: 1) Hypothesis generation, where a list of hypothesis source code information is generated; 2) Segmentation, where the hypothesis list is grouped into segments; 3) Concept binding, where the segments turn into a list of concept bindings linking regions of source code. Secondly, the derived viewpoints-based context-aware service requirements model is proposed to fully discover constraints, and the requirements evolution model is developed to maintain and specify the requirements evolution process for supporting context-aware services evolution. Finally, inspired by context-oriented programming concepts and approaches, ContXFS is implemented as a COP-inspired conceptual library in F#, which enables developers to facilitate dynamic context adaption. This library along with context-aware requirements analyses mitigate the development of the said systems to a great extent, which in turn, achieves reconciliation between requirements and implementation.

Today, software has become deeply woven into the fabric of our lives. The quality of the software we depend on needs to be ensured at every phase of the Software Development Life Cycle (SDLC). An analyst uses the requirements engineering process to gather and analyze system requirements in the early stages of the SDLC. An undetected problem at the beginning of the project can carry all the way through to the deployed product. The Requirements Traceability Matrix (RTM) serves as a tool to demonstrate how requirements are addressed by the design and implementation elements throughout the entire software development lifecycle. Creating an RTM matrix by hand is an arduous task. Manual generation of an RTM can be an error prone process as well. As the size of the requirements and design document collection grows, it becomes more challenging to ensure proper coverage of the requirements by the design elements, i.e., assure that every requirement is addressed by at least one design element. The techniques used by the existing requirements tracing tools take into account only the content of the documents to establish possible links. We expect that if we take into account the relative order of the text around the common terms within the inspected documents, we may discover candidate links with a higher accuracy. The aim of this research is to demonstrate how we can apply machine learning algorithms to software requirements engineering problems. This work addresses the problem of requirements tracing by viewing it in light of the Ant Colony Optimization (ACO) algorithm and a reinforcement learning algorithm. By treating the documents as the starting (nest) and ending points (sugar piles) of a path and the terms used in the documents as connecting nodes, a possible link can be established and strengthened by attracting more agents (ants) onto a path between the two documents by using pheromone deposits. The results of the work show that ACO and RL can successfully establish links between two sets of documents.

Despite existing work on dealing with security and usability concerns during the early stages of design, there has been little work on synthesising the contributions of these fields into processes for specifying and designing systems. Without a better understanding of how to deal with both concerns at an early stage, the design process risks disenfranchising stakeholders, and resulting systems may not be situated in their contexts of use. The research problem this thesis addresses is how techniques and tools can be integrated and improved to support the design of usable and secure systems. To develop this understanding, we present IRIS (Integrating Requirements and Information Security) --- a framework for specifying usable and secure systems. IRIS considers the system design process from three different perspectives --- Usability, Security, and Requirements --- and guides the selection of techniques towards integrative Security, Usability, and Requirements Engineering processes. This thesis claims that IRIS is an exemplar for integrating existing techniques and tools towards the design of usable and secure systems. In particular, IRIS makes three significant contributions towards the stated research problem. First, a conceptual model for usable secure Requirements Engineering is presented, upon which the IRIS framework is founded; this meta-model informs changes to elicitation and specification techniques for improved interoperability in the design process. Second, several characteristics of tool-support needed to elicit and specify usable and secure systems are introduced; the CAIRIS (Computer Aided Integration of Requirements and Information Security) software tool is presented to illustrate how these characteristics can be embodied. Third, we describe how the results of applying IRIS can be used to improve the design of existing User-Centered Design techniques for secure systems design. We validate the thesis by applying the IRIS framework to three case studies. In the first, IRIS is used to specify requirements for a software repository used by a UK water company. In the second, IRIS is used to specify security requirements for a meta-data repository supporting the sharing of medical research data. In the final case study, IRIS is used to analyse a proposed security policy at a UK water company, and identify missing policy requirements. In each case study, IRIS is applied within the context of an Action Research intervention, where findings and lessons from one case study are fed into the action plan of the next.

The contextual evaluation framework (CEF) is a requirements engineering technique that incorporates a particular sociological orientation, Ethnomethodology, in the development of a rigorous and systematic approach for requirements elicitation. This qualitative approach examines how well a system may be aligned with the endogenous organisation of work within a community of practice. Assessing how well a system supports the knowledge, skills and practices that already exist within a community is equally as important as developing solutions that will eventually reconfigure those practices, create new ones and extend modes of collaboration. The aim of this thesis is to address the absence of a systematic approach to quasi-naturalistic prototype evaluation which may be useful to a broader community such as requirements engineers, computer scientists and others not familiar with the details of sociological approaches. Such an aim is in line with the ways in which prototype evaluation approaches, particularly in HCI, have successfully been disseminated throughout the computer science research community - with the provision of guidelines. Likewise, the CEF is conceived of to be implemented in a similar manner. Its focus is on the analysis of a prototype’s relevance as a tool that is in some manner familiar to those who might use it. Specifically, professionals within a discipline share complex skills and knowledge where they learn to use similar tools, instruments and processes necessary for their work. Implicit in these social practices, practitioners gradually acquire the knowledge and skills necessary to become full members of a community of practice. In this way, the processes, objects and artefacts of practice come to possess specific meaning and significance. The CEF examines how this complex architecture of meaning is supported, constrained or transformed when using a prototype and makes possible an assessment of the ways in which participants interpret its usefulness and usability.

Complex defense acquisition programs, such as the Global Positioning System program, have many requirements engineering challenges to overcome in order to deliver capabilities to customers and satisfy other stakeholders. To meet these challenges and stay within cost and schedule constraints, engineers and managers need system requirements information organized in a clear, complete, and efficient manner to support decision making. An effective methodology tailored to the needs of the program decision makers will ensure that important information is correct, organized, and readily accessible. The GPS program is implementing a methodology that includes standardized processes across its segments. However, the GPS program refrained from implementing a better requirements engineering approach and using its current requirements engineering tool to take advantage of this approach.

Advanced Surface Movement Guidance and Control Systems (A-SMGCS) help to further improve safety and efficiency of the traffic on the aerodrome surface. The current A-SMGCS sensor technologies have certain operational and functional limitations. A new and unprecedented sensor technology is being tested as a pilot project. This unique sensors is called MagSense®. It works based on the principle of detecting the influence of ferromagnetic materials on earth’s magnetic field. For applications in the aviation environment, learning processes are necessary which are generally based on the graphical depiction of stored sensor data and features to analyze the graphs. For this purpose a visualization and analysis tool is needed. In order to create an adequate tool to allow for depicting stored sensor data and the peaks caused by ferromagnetic objects in aircraft and vehicles, a requirements engineering process will be conducted wherein the requirements of the various stakeholders will be identified and harmonized. In general, the appropriate RE approach will ensure mutual agreement among the stakeholders and a set of requirements for the first edition of the tool without contradictions. The harmonized package of requirements will then be used as the starting point for a software development process, after which the tool will be produced as specified and validated as a part of this Master’s Thesis. This Master’s Thesis puts a special focus on the choice of a suitable method in Requirements Engineering and Requirements Management, adequately adapted to the project size and its quality. The selection of appropriate elements from the methodology as well as the outcomes from applying them on a specific software production project are at the core.

The challenges of developing the information systems (IS) that support modern enterprises are becoming less about engineering and more about people. Many of the technical issues of the past, such as hardware size and power, connectivity, and robust software, are engineering problems that have largely been solved. In the next stage of computing, the human factor will be far more important than it has been in the past: the colors of an interface or the shape of an icon are the engineering problems of the past, and the availability and usefulness of such basic solutions is rapidly coming to a close. A new paradigm is needed that provides a roadmap of higher level conceptions and values, one about humane computing. A part of this older, mechanistic approach are quantitative, economic values whose impact on IS are readily visible and acknowledged within software engineering. However, qualitative values, and in particular consumer preferences, have been researched to a lesser degree, and there has been very little direct application.  To create the next-generation information systems, requirements engineers and systems developers need new methods to capture the real preferences of consumers, conceptualize these abstract concepts, and then relate such preferences to concrete requirements for information systems. To address this problem, this thesis establishes a conceptual link between the preferences of consumers and system requirements by accommodating the variations between them and expressing them via a conceptual model. Modeling such preferences and values so that they can be used as requirements for IS development is the primary contribution of this work. This is accomplished via a design science research paradigm to support the creation of the works’ primary artifact—the Consumer Preference-aware Meta-Model (CPMM). CPMM is intended to improve the alignment between business and information systems by capturing and concretizing the real preferences of consumers and then expressing such preferences via the requirements engineering process, with the eventual output being information systems. CPMM’s development relies on theoretical research contributions within three areas in information systems—Business Strategy, Enterprise Architecture, and Requirements Engineering—whose relationships to consumer values have been under-researched and under-applied. The case studies included in this thesis each demonstrate the significance of consumer preferences to each of these three areas.  In the first, a set of logical mappings between CPMM and a common approach to business strategy (strategy maps/balanced scorecards) is produced. In the second, CPMM provides the conceptual undergirding to process a massive amount of unstructured consumer-generated text to generate system requirements for the airline industry. In the concluding case, an investigation of foreign and domestic students at Swedish universities is structured through CPMM, one that first discovers the requirements for a consumer preference-based online education and then produces feature models for such a software product line-based system. The significance of CPMM as a lens for discovering new concepts and highlighting important information within consumer preference data is clearly seen, and the usefulness of the meta-model is demonstrated by its broad and beneficial applicability within information systems practice and research.

The systematic interaction of software developers with the business domain experts that are usually no software developers is crucial to software system maintenance and creation and has surfaced as the big challenge of modern software engineering. Existing frameworks promoting the typical programming languages with artificial syntax are suitable to be processed by computers but do not cater to domain experts, who are used to documents written in natural language as a means of interaction. Other frameworks that claim to be fully automated, such as those using natural language processing, are too imprecise to handle the typical requirements documents written in heterogeneous natural language flavours. In this thesis, a framework is proposed that can support the specification of business rules that is, on the one hand, understandable for nonprogrammers and on the other hand semantically founded, which enables computer processability. This is achieved by the novel language Adaptive Business Process and Rule Integration Language (APRIL). Specifications in APRIL can be written in a style close to natural language and are thus suitable for humans, which was empirically evaluated with a representative group of test persons. A useful and uncommon feature of APRIL is the ability to define reusable abstract mixfix operators as sentence patterns, that can mimic natural language. The semantic underpinning of the mixfix operators is achieved by customizable atomic formulas, allowing to tailor APRIL to specific domains. Atomic formulas are underpinned by a denotational semantics, which is based on Tempura (executable subset of Interval Temporal Logic (ITL)) to describe behaviour and the Object Constraint Language (OCL) to describe invariants and pre- and postconditions. APRIL statements can be used as the basis for automatically generating test code for software systems. An additional aspect of enhancing the quality of specification documents comes with a novel formal method technique (ISEPI) applicable to behavioural business rules semantically based on Propositional Interval Temporal Logic (PITL) and complying with the newly discovered 2-to-1 property. This work discovers how the ISE subset of ISEPI can be used to express complex behavioural business rules in a more concise and understandable way. The evaluation of ISE is done by an example specification taken from the car industry describing system behaviour, using the tools MONA and PITL2MONA. Finally, a methodology is presented that helps to guide a continuous transformation starting from purely natural language business rule specification to the APRIL specification which can then be transformed to test code. The methodologies, language concepts, algorithms, tools and techniques devised in this work are part of the APRIL-framework.