Statechart normalizations

  • Authors:
  • Benjamin De Leeuw;Albert Hoogewijs

  • Affiliations:
  • Ghent University, Department of Pure Mathematics, Gent, Belgium;Ghent University, Department of Pure Mathematics, Gent, Belgium

  • Venue:
  • WSEAS Transactions on Information Science and Applications
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Simplified statecharts are derived by excluding all redundant constructs of the UML (Unified Modeling Language) metamodel on statecharts. In previous papers we introduced this basic concept and pointed out some interesting applications. We transformed and compacted state machines into serializable objects that clearly highlight their basic constructs and showed how a lot of sparse edges were created during this transformation. Sparse edges contain little or no information (e.g. empty transitions). From this compaction we derived a theory in which state machines are reduced to two distinct partial orderings on states. For the sake of convenience to read this paper, we briefly recapture part of this theory and show the exponentially growing complexity of calculating all possible values of variables that appear on state machine edges. In order to arrive at feasible algorithms leading to practical applications of the implicity complex partial ordering relations we need to reduce this complexity by formulating and proving reductions to state machine normal forms. Apart from aligning formal and behavioral equivalence, normal forms allow us to reduce the number of sparse edges and useless states thereby limiting calculational complexity. In this paper, we extend our theory with normal forms and inject them into the theory of simplified statecharts presented in earlier work. Some statecharts are only seemingly different from others if one analyzes the different paths in those statecharts. The UML is designed to allow for this kind of (uncontrollable) flexibility but in mathematical descriptions it has adverse effects. We introduce an equivalence relation on simplified statecharts and derive a normalization procedure which converts a simplified sc to a normalized simplified sc equivalent to the original one. Our formalism allows us to unravel superficial differences between simplified statecharts.