Fast exact minimization of BDDs

We present a new exact algorithm for finding the optimal variable or deringfor r educ edordered Binary Decision Diagrams (BDDs). The algorithm makes use of a lower bound technique known from VLSI design.Up to now this technique has been use donly for the oreticalconsider ationsand it is adapte dhere for our purp ose. Furthermore, the algorithm supports symmetry aspects and makes use of a hashing b aseddata structure. Exp erimental esults are given to demonstrate the efficiency of our appr oach. We succeeded in minimizing adder functions with up to 64 variables, while all other pr eviously pr esente d appr oaches fail.