||A Multi-Modal Transport Net: Using a Systematic Approach to Software Development
||Cour, Peder D. la
||Kristensen, Jens Thyge (Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
Bjørner, Dines (Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
||Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark
||The thesis takes as its main example that of multi-modal, hence abstracted transportation nets. Transportation nets can contain roads, railways, shipping lanes and air corridors. In steps of development via requirements the thesis then shows how to project and instantiate the domain description onto requirements for such seemingly diverse applications as (a) road maintenance, (b) travel planning [logistics] in road and rail transportation nets, and (c) the animation of train traffic. Any kind of transportation nets could be used for the kinds of systems in (a), (b) and (c) (e.g. one might wish a system for travel planning using roads, trains, ship and airplanes), the examples mentioned here are the ones used in this thesis.
The thesis of this project include the claims that to professionally develop trustworthy software, in one or more versions, one is well advised in:
1. Following a method from domain engineering via requirements engineering to software design.
2. To present both domain descriptions and requirements prescriptions in both informal, narrative form, and in modern mathematical form. The informal form is understandable by all stakeholders, and it helps to ensure that one arrives at the right software. The mathematical form helps to better ensure that we arrive at software that is right (correct).
3. To apply a number of algebra-like techniques for ''deriving'' domain requirements prescriptions in a systematic manner from domain descriptions.
4. To ensure that when making two or more applications within roughly the same domain one must apply a careful analysis of how the domain description imposes the necessary and sufficient so-called ''fitting'' requirements. Domains do not change that often. Requirements are often claimed to change very often for and within the same application. We dispute that claim - showing that the analysis and synthesis techniques of ''fitting'' - in a large measure - helps to avoid the syndromes of ''changing requirements''.
||Thesis not public available.
||Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU : DK-2800 Kgs. Lyngby, Denmark
Creation date: 2006-10-06
Update date: 2012-12-18