Network Programme - Section 3

Approach

We will exploit the breadth of skills and backgrounds within RENOIR to develop a "broad spectrum" approach to requirements engineering combining experimental, conceptual, formal and observational methodologies. RENOIR is uniquely placed to foster such an approach. We will place particular emphasis on developing balanced assessments of the merits and demerits of different approaches and on the contexts in which they can best be applied. We will emphasise the development and use of, industrially relevant, shared exemplars and focus problems.

By focusing on research areas and tasks there is a danger that insufficient emphasis is placed on the requirements engineering process as a whole. A major theme in RENOIR, cutting across all the research areas will be the requirements engineering process and process improvement. RENOIR will pay specific attention to the means by which individual contributions in specific research areas can be assembled into a coherent tool-supported method (using that term loosely) sensitive to the evolutionary and incremental character of the systems development process.

RENOIR will, in addition to building on research in software engineering and information systems, draw together research with relevance to requirements engineering originating in the areas of human factors, systems science, management and the social sciences.

RENOIR provides a framework for research coordination; research training; infrastructure; industrial liaison and technology transfer. This is reflected in our programme of activities. It provides a means for overcoming practical problems associated with a distributed and heterogeneous research community. These problems are evident in all the areas which RENOIR addresses but is most immediately obvious in the areas of infrastructure and in industrial liaison and technology transfer. The difficulty of access to existing, but somewhat fragmented expertise, excludes research groups and industry (particularly SMEs) and has inhibited technology transfer and process improvement in the area of requirements engineering.

There is substantial value added by coordination over and above the benefits which accrue from surmounting the practical problems discussed above. RENOIR will provide an opportunity for the key European research groups to work together and to develop joint research; it will improve the dissemination of the products of research and will foster a critical research and technology development community; it will allow economy of effort through shared infrastructure and research training; it will bring together application, technology and domain expertise to focus on common problems.

Our analysis of the major unsolved problems in requirements engineering can be linked to the framework outlined above. Some specific "hot" questions which we anticipate will be considered within RENOIR are:

• establishing the relation between requirements engineering and incremental development;
• requirements engineering in the context of development for markets;
• reengineering contract and procurement processes;
• extending work on psychology, selection and training of programmers to the participants in requirements engineering;
• the use of design experience to inform bounding decisions;
• postmortem analysis of system failures, methods for carrying out such analyses and recording and deploying the resulting knowledge;
• organised guidance to assist in identifying stakeholders and means for reasoning about, and drawing consequences from, stakeholder analysis;
• empirical studies of group work in requirements engineering;
• selection and combination of information gathering techniques;
• application of multi-criteria decision making techniques to support negotiation and trade-offs;
• techniques for goal identification;
• abstractions underlying modelling languages;
• scenarios to support the identification of domain objects;
• integrating task models with other models used in requirements engineering;
• testing requirements engineering reuse approaches through the development of significant case bases and generic domain models for realistic domains;
• techniques for generating interesting or relevant questions to ask of the products of requirements engineering;
• generation of testing strategies;
• how to guide exploration and organise feedback;
• automated support for formal reasoning and proof in the context of requirements engineering;
• consistency management and toleration of inconsistency;
• development of groupware support for inspection;
• building support for comprehensive data collection into requirements engineering methods and tools;
• video and sound records of technical meetings and document annotations;
• information presentation and extraction from large requirements engineering repositories;
• establishing advantages and drawbacks of different schemes for recording rationale;
• supporting "pre-requirements" traceability; evolution of requirements engineering standards and codes of practice.

Back to title Next section Previous section Contents page

---

Last up-date: 30 July 1998

---