Output details
11 - Computer Science and Informatics
University College London
Generating and evaluating choices for fixing inconsistencies in UML design models
<07>Success of model-driven engineering will depend on availability of powerful tools for elaborating and managing large-scale models (similar to code IDE). This paper shows, for the first time, how modelling tools can generate concrete propositions for fixing inconsistencies in UML models taking into account inter-dependencies between consistency rules (a harder problem than generating code fixes). Validation of correctness and scalability on more than 30 industrial UML models of up to 12,000 elements. Technique, developed in paper as extension to IBM Rational Rose, has since been ported to Matlab/Stateflow, IBM Rational Software Modeler, and the Dopler suite for product line engineering.