Interactive configurator with FO(.) and IDP-Z3
This addresses the problem of difficult implementation and maintenance of configurators for engineers in industrial settings, but it appears incremental as it builds on existing knowledge-based approaches.
The paper tackled the challenge of building interactive configurators for constraint solving problems in industry by introducing IDP-Z3, a reasoning engine for the FO(.) knowledge representation language, which enables automatic generation of configurators from a knowledge base, though no concrete performance numbers are provided.
Industry abounds with interactive configuration problems, i.e., constraint solving problems interactively solved by persons with the assistance of a computer. The computer program, called a configurator, needs to perform a variety of reasoning tasks with the (often incomplete) information that the user provides. Imperative programming approaches make such systems difficult to implement and maintain. Knowledge-based configurators have been proposed to help engineers solve such problems, but many challenges remain. We present IDP-Z3, a new reasoning engine for the FO(.) KR language, and we report on its use for building configurators automatically from a knowledge base.