Sarah Thew

Experience with user-centred requirements engineering

This paper describes the application of human–computer interaction (HCI) principles and methods to requirements engineering in a case study development of a visualisation tool, ADVISES, to support epidemiological research. The development approach consisted of scenario-based design and analysis of the users’ tasks and mental model of the domain. Prototyping and storyboarding techniques were used to explore design options with users as well as specifying functionality for two versions of the software to meet the needs of novice and expert users. Application of HCI functional allocation heuristics to guide system requirements decisions is explained. An evaluation of the prototype was carried out to assess the extent to which the expert model would support public health professionals in their analysis activities. The results of the design exploration requirements analysis study are reported. The implications of scenario-based design exploration, functional allocation and software architecture are discussed.

Requirements Engineering for E-science: Experiences in Epidemiology

The Advises (Adaptive Visualization of E-science) project is developing tools to support geographic visualization in epidemiology and public-health decision making. In this project, a user-centered requirements process focuses on the research questions epidemiologists ask, the language they use, and the tacit knowledge employed in reasoning about epidemiological data. Combining a range of requirements-gathering techniques provides considerable advantages.

User engagement by user-centred design in e-Health

This paper describes the application of user-centred design (UCD) methods and a user engagement (UE) approach to a case study development of a visualization tool (ADVISES) to support epidemiological research. The combined UCD/UE approach consisted of scenario-based design, and analysis of the users’ tasks and mental model of the domain. Prototyping and storyboarding techniques were used to explore design options with users as well as specifying functionality for two versions of the software to meet the needs of novice and expert users. An evaluation of the prototype was carried out to assess the extent to which the expert model would support public health professionals in their analysis activities. The results of the design exploration requirements analysis study are reported. The implications of scenario-based design exploration, participatory design and user engagement are discussed.

Experience in e-Science Requirements Engineering

We describe the experience of using a combination of requirements engineering techniques (scenarios, storyboards, observation and workshops) in an e-science application to develop a geographical analysis tool for epidemiologists. Problems encountered were: eliciting tacit knowledge; and creating new visions and working practices for our users. The combination of techniques worked well, although observation of working practice was not so effective in this scientific domain, where activity is mainly cognitive.

Developing visualization-based decision support tools for epidemiology

The paper describes the application of user-centred design (UCD) methods to a case study of the development of visual decision support tools to support epidemiological research. Understanding the causes of obesity requires analysis of complex medical surveys and geographic information. Translating research on obesity into effective public health measures requires collaboration between medical researchers and public health analysts. The objective of this research is to develop software tools to support medical researchers and public health analysts in collaborative investigation of obesity in children. The UCD approach consisted of scenario-based design, storyboarding and prototyping to explore design options to meet the needs of public health analysts and academic researchers. An evaluation of the prototype was carried out to assess the extent to which the medical researcher model would support public health professionals in their analysis activities. The design and evaluation of the prototype are discussed. A visualization-based research and decision-support system was implemented leading to positive evaluation results from users.

FARSITE: evaluation of an automated trial feasibility assessment and recruitment tool.

Objectives In a review of UK-supported clinical trials more than half of the investigators asked the funding agency for an extension and a third did not hit their recruitment targets [1]. Study feasibility is often assessed on an ad hoc basis by asking clinical staff to estimate how many patients with particular characteristics they might expect to see in a given time period, and over-estimation is common. We have developed FARSITE (Feasibility Assessment and Recruitment System for Improving Trial Efficiency), a system to support the evaluation of trial feasibility by providing accurate assessments of numbers of patients eligible for a particular trial. Furthermore FARSITE automates patient recruitment whilst preserving consent for consent.

Obesity Atlas and Methodbox: Towards an open framework for sharing public health intelligence workflows

The large growth in data sources relevant to public health has not been matched by a growth in human resource for producing intelligence to support decisions or generate new insights. There is a need to bring scarce public health expertise into closer alignment with data and data processing methods to support timely public health analysis. The difficulties of developing and sharing this expertise in large organisations such as the UKs National Health Service have long been recognised. We report findings in this area across two projects Obesity Atlas and Methodbox, which are developing and sharing best practice between Public Health Analysts in England, and we address the relevant generic knowledge management problems in the Public Health community.

Value-based requirements engineering: method and experience

Abstract‘Socio-political’ issues, such as emotions, values and people’s feelings, are often cited as problems in the RE process. A method is described for analysing such issues. The method consists of a taxonomy of stakeholders’ values, motivations and emotions (VME), with process guidance for eliciting and analysing these issues for the RE process and design implications. Values are personal attitudes or long-term beliefs which may influence stakeholder functional and non-functional requirements. Motivations are psychological constructs related to personality traits which may be viewed as stakeholders’ long-term goals in RE. Emotions are cues to stakeholders’ reactions arising from value/motivation conflicts. The method is supported by a website which illustrates the taxonomy with explanations and scenarios describing problems arising from value conflicts, and from poor understanding of stakeholder values. Two method validation studies were undertaken: first, an evaluation of the website and method by novices and RE experts; and second, case study applications of RE value analysis in real-world industrial practice. The method was used by all practitioners, although in different ways, some used it to create an agenda of issues for analysis while others employed the VMEs to interpret stakeholders’ views and manage stakeholder negotiations. The validation studies provide evidence for the acceptability of the method for industrial practitioners, illustrating how value-related problems are identified and analysed effectively by the method. The utility of analysing VMEs is compared to other ‘socio-political issues’-oriented methods in RE and methods which focus on monetized values in product requirements.

The Design Implications of Users’ Values for Software and System Architecture

This chapter argues the case for investigating users’ values as important influences on the design of software architecture. First, definitions of values in the psychological and sociological literature are reviewed, followed by a proposed taxonomy of values that are relevant to software engineering domains, such as trust, cooperativeness, privacy, and creativity. Stakeholder values are related to nonfunctional requirements or system quality attributes. The architectural implications for user values are described by linking values to design for autonomy, monitoring/awareness, and other architectural components. Implications are explained in terms of software and system (i.e., sociotechnical) architecture that relate software components to operations in system design with human factors and organizational concerns. Application of the value-based requirements engineering and architecture design method is illustrated in a case study of a medical research support system. Methods for eliciting user values in the requirements analysis process are explained, and the implications for the future of value-based software architectures are discussed.