TY - GEN
T1 - Debugging of inconsistent UML/OCL models
AU - Wille, Robert
AU - Soeken, Mathias
AU - Drechsler, Rolf
PY - 2012
Y1 - 2012
N2 - While being a de-facto standard for the modeling of software systems, the Unified Modeling Language (UML) is also increasingly used in the domain of hardware design and hardware/software co-design. To ensure the correctness of the specified systems, approaches have been presented which automatically verify whether a UML model is consistent, i.e. free of conflicts. However, if the model is inconsistent, these approaches do not provide further information to assist the designer in finding the error. In this work, we present an automatic debugging approach which determines contradiction candidates, i.e. a small subset of the original model explaining the conflict. These contradiction candidates aid the designer in finding the error faster and therefore accelerate the whole design process. The approach employs different satisfiability solvers as well as different debugging strategies. Experimental results demonstrate that, even for large UML models with up to 2500 classes and constraints, the approach determines a very small number of contradiction candidates to be inspected.
AB - While being a de-facto standard for the modeling of software systems, the Unified Modeling Language (UML) is also increasingly used in the domain of hardware design and hardware/software co-design. To ensure the correctness of the specified systems, approaches have been presented which automatically verify whether a UML model is consistent, i.e. free of conflicts. However, if the model is inconsistent, these approaches do not provide further information to assist the designer in finding the error. In this work, we present an automatic debugging approach which determines contradiction candidates, i.e. a small subset of the original model explaining the conflict. These contradiction candidates aid the designer in finding the error faster and therefore accelerate the whole design process. The approach employs different satisfiability solvers as well as different debugging strategies. Experimental results demonstrate that, even for large UML models with up to 2500 classes and constraints, the approach determines a very small number of contradiction candidates to be inspected.
UR - http://www.scopus.com/inward/record.url?scp=84862108901&partnerID=8YFLogxK
U2 - 10.1109/date.2012.6176655
DO - 10.1109/date.2012.6176655
M3 - Conference contribution
AN - SCOPUS:84862108901
SN - 9783981080186
T3 - Proceedings -Design, Automation and Test in Europe, DATE
SP - 1078
EP - 1083
BT - Proceedings - Design, Automation and Test in Europe Conference and Exhibition, DATE 2012
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 15th Design, Automation and Test in Europe Conference and Exhibition, DATE 2012
Y2 - 12 March 2012 through 16 March 2012
ER -