Model checking based service delivery planning

Abstract

Service delivery planning may often involve plan specific ordering or sequencing especially in non-euclidean transit environments. Common approaches for vehicle routing fall short in adequately handling compound ad-hoc plan dependent constraints. We address multi-agent service provisioning by starting from different initial nodes of shared irregular transport graphs. We employ concurrent and interleaved dynamics composition to capture agent based service delivery problems as transition systems and related goal state properties that can be the subject of symbolic model checking. Formally verifying the negation of the goal state property allows to generate a counter example such that, if it exists, the route allocation planning problem finds a solution. The state explosion is mitigated by bounded model checking that has the additional advantage of providing counter examples of minimal length which correspond to minimized agent tours under a cost bound.