Keywords requirements engineering, process modelling
Start Date: 01-MAY-92 / Duration: 31 months
[ contact / participants ]
The goal of the F3 project is to produce a set of models and techniques designed to improve requirements acquisition and its transformation into validated system specification. This set will be supported by an RE workbench, a working environment and general facilities for managing the RE process and reusing earlier work.
The techniques involved include the simultaneous use of formal and informal notations with conversion both ways, the explicit modelling of the environment of the computerised system, the ability to integrate compatible viewpoints and to negotiate incompatible ones, the tracking of RE decisions and assistance in analysing their consequences, and the validation of the requirements specification by techniques such as rapid prototyping of the various layers of an application (database access, dialogue, human computer interface) in a multi-system environment.
The conceptual core of F3 framework has been elaborated. It consists of a set of interrelated models that can be used for representing the various parts of the knowledge needed by Requirement Engineers, either on the enterprise environment or on the target Information system (data, dialogue, HCI, and technical non functional Requirement modelling. Linkages between sub models are provided so checking consistency of Requirement choices is made possible. IS modelling facilities include formal and executable languages designed to be directly interpreted by rapid prototyping tools. A generic decision trace mechanism is described which allow Requirement engineers to record significant decisions which ever part of the RE process is concerned. A model for identifying reusable Requirements Engineering components is also provided.
An architecture for RE case tool interoperability has been defined, based upon Software Bus (R) Technology (cf. Eureka project: ESF European Software factory); it will provide a working environment able to integrate multiple tools and supporting common facilities (session management, cross-referencing, decision trace...), in such a way integrated views of Requirement components are made available.
First demonstrators have been achieved, especially a knowledge based system for analyzing conflicting requirements, and prototyping tools (data intensive system prototyper, User interface generator).
Fuzziness of requirements represents a major issue both for software houses and firms developing software for themselves. F3 method and tools are expected to provide European actors a competitive edge by an improved quality and a reduction of software time to market. Key success factors of F3 framework are its acceptability by Requirement engineers and the continuity with existing methods especially standards such as SSADM and MERISE. During phase 2 of the project, those issues will be specifically addressed: communication supports (F3 book, Hypertext and slide presentations) will be specified and realised). Special attention will be paid on the adaptation of existing software development methods through the EUROMETHOD framework.
Acceptance of the F3 framework by Requirement Engineers will also be facilitated by the flexibility of its tool support. F3 tools are designed as communicating and "pluggable components". They will be eligible to be incorporated in any case workbench confirming to ESF Software Bus (R) or CORBA standard.
Mr B. Guthmann
SEMA GROUP SA
16, rue Barbès
F - 92126 Montrouge Cedex
tel: + 33 / 1-4092-4288
fax: + 33 / 1-4656-9653
SEMA GROUP SA - F - C
BRITISH AEROSPACE (MILITARY
AIRCRAFT) LTD - UK - P
INISEL - E - P
RUTHERFORD APPLETON LABORATORY - UK - P
TXT SPA - I - P
SISU - S - P
GOETHE UNIVERSITÄT FRANKFURT - D - A
POLITECNICO DI MILANO - I - A
SEMA GROUP SAE - E - A
UNIVERSITÉ DE PARIS I SORBONNE - F - A
UMIST - UK - A
F3 - 6612, December 1993
please address enquiries to the ESPRIT Information Desk
html version of synopsis by Nick Cook