A graph-based aspect interference detection approach for UML-based aspect-oriented models

Aspect Oriented Modeling (AOM) techniques facilitate separate modeling of concerns and allow for a more flexible composition of these than traditional modeling techniques. While this improves the understandability of each submodel, in order to reason about the behavior of the composed system and to detect conflicts among submodels, automated tool support is required. Current techniques for con°ict detection among aspects generally have at least one of the following weaknesses. They require to manually model the abstract semantics for each system; or they derive the system semantics from code assuming one specific aspect-oriented language. Defining an extra semantics model for verification bears the risk of inconsistencies between the actual and the verified design; verifying only at implementation level hinders fixing errors in earlier phases. We propose a technique for fully automatic detection of conflicts between aspects at the model level; more specifically, our approach works on UML models with an extension for modeling pointcuts and advice. As back-end we use a graph-based model checker, for which we have defined an operational semantics of UML diagrams, pointcuts and advice. In order to simulate the system, we automatically derive a graph model from the diagrams. The result is another graph, which represents all possible program executions, and which can be verified against a declarative specification of invariants. To demonstrate our approach, we discuss a UML-based AOM model of the "Crisis Management System" and a possible design and evolution scenario. The complexity of the system makes conflicts among composed aspects hard to detect: already in the case of two simulated aspects, the state space contains 623 different states and 9 different execution paths. Nevertheless, in case the right pruning methods are used, the state-space only grows linearly with the number of aspects; therefore, the automatic analysis scales.