Abstract

Abstract. Reformulation into SAT is one of the main approaches to deterministic planning. The main idea is to successively check the satis-fiability of each of the propositional formulas in a sequence φ0, φ1, φ2,... where φi encodes the feasibility of a plan of length i. For real-life applications, resources such as time, distances, or money must often be considered. Therefore, in the context of plannning with resources, a nat-ural choice is to consider satisfiability modulo theories (SMT) instead of plain SAT. In this paper we consider the application of off-the-shelf SMT solvers to the Petrobras domain, an abstraction of a real-life prob-lem of resource-efficient transportation of goods from ports to petroleum platforms. We consider different encodings into SMT and compare with state of the art alternative approaches. 1