Multiagent Shape Grammar Implementation: Automatically Generating Form Concepts According to a Preference Function

Abstract

In new product development, quickly generating many product form concepts that a potential consumer prefers is a challenge. This paper presents the inaugural multiagent shape grammar implementation (MASGI) to automatically generate product form designs according to a preference function that can represent designer or consumer design preference. Additionally, the multiagent system creates a flexible shape grammar implementation that enables modifications to the shape grammar as the form design space changes. The method is composed of three subprocesses: a shape grammar interpreter that implements the shape grammar, an agent system that chooses which shape grammar rules to implement and the parametric design choices according to a preference function, and a preference investigator that determines the preference function, which constraints the automated form design process.