Proving preservation of transitivity invariants in model transformations

This paper develops methods to reason about graph transformations, and in particular to show that transitivity and reachability invariants are preserved during transformations. In our approach, graph transformations consist of a pattern defining an applicability condition, and an operational description of the desired transformation. Whereas previous work was restricted to Boolean combinations of arc expressions as patterns, we extend the approach to patterns containing transitive closure operations, which implicitly denote an unbounded number of nodes. We show how these can be internalized into a finite pattern graph so that model checking techniques can be applied for verification.