Representing Configuration Knowledge with Weight Constraint Rules

Answer Set Programming: Towards Efficient and Scalable Knowledge Representation and Reasoning

Volume

Papers from the 2001 AAAI Spring Symposium

Contents

The paper demonstrates that product configuration applications fit naturally the framework of answer set programming. It is shown that product configuration knowledge can be represented systematically and compactly using a logic program type rule language such that the answers of a configuration task, the configurations, correspond to the models of the rule representation. The paper presents such a rule-bised for realization of a unified configuration ontology using a weight constraint rule language. The language extends normal logic programs with cardinality and weight constraints which leads to a compact and simple formalization. The complexity of the configuration task defined by the formalization is shown to be NP-complete.