A Symbolic OBDD-Based Algorithm for Assembly Sequence Planning

Abstract

Algorithms based on timed Petri net are competitive for solving the problem of assembly sequence planning (ASP). In order to alleviate the state-space explosion problem which is caused by the Petri net-based representation of assembly sequences and to improve the efficiency of planning algorithms, an approach based on ordinary binary decision diagrams (OBDDs) is presented in this paper. On the one hand, all the timed transitions in Petri nets are substituted by some technically designed timed transition structures; on the other hand, Petri nets used in algorithms of assembly sequence planning are represented symbolically as OBDDs. As a result, a novel assembly sequence planning algorithm based on symbolic OBDDs is presented. With the help of OBDDs, both the marking vectors and the states of Petri nets are formulated more compactly; at the same time, the searching processes can be manipulated implicitly. Therefore, compared with original algorithms, both the time performance and the space performance are improved.