Better algorithms for analyzing and enacting declarative workflow languages using LTL

  • Authors:
  • Michael Westergaard

  • Affiliations:
  • Department of Mathematics and Computer Science, Eindhoven University of Technology, The Netherlands

  • Venue:
  • BPM'11 Proceedings of the 9th international conference on Business process management
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Declarative workflow languages are easy for humans to understand and use for specifications, but difficult for computers to check for consistency and use for enactment. Therefore, declarative languages need to be translated to something a computer can handle. One approach is to translate the declarative language to linear temporal logic (LTL), which can be translated to finite automata. While computers are very good at handling finite automata, the translation itself is often a road block as it may take time exponential in the size of the input. Here, we present algorithms for doing this translation much more efficiently (around a factor of 10,000 times faster and handling 10 times larger systems on a standard computer), making declarative specifications scale to realistic settings.