An evolutionary constraint satisfaction solution for over the cell channel routing

Abstract

A novel evolutionary approach for ordering assignments in combinatorial optimization using constraint satisfaction problem (CSP) modeling is presented. In assignment of values to variables, the order of assignments is determined through evolutionary optimization exploiting problem specific features. No a priori information is available on the order of assignments that is completely determined by evolutionary optimization to produce the best assignment results. Indeed, experimental evaluations show that the proposed method outperforms very well-known approaches for the solution of NP-hard combinatorial optimization problems from VLSI layout design, namely, channel routing and multi-layer over the cell channel routing. © Springer-Verlag 2004.