The stable semantics has become a prime candidate for knowledge representation and reasoning. The rules associated with propositional logic programs and the stable semantics are not expressive enough to let one write concise optimization programs. We propose an extension to the language of logic programs that allows one to express optimization problems in a suitable well. In earlier work we defined the declarative semantics for partial order clauses. The main contribution of our paper is the following: First, we define the language of our extended paradigm as well as its declarative semantics. Our declarative semantics is based on translating partial order clauses into normal programs and the using the stable semantics as the intended meaning of the original program. Second, we propose an operational semantics for our paradigm. Our experimental results show that our approach is more efficient than using the well known system SMODELS over the translated program.