A Two-Tiered Grammatical Approach for Agent-Based Computational Design

Abstract

Early stages of the engineering design process are vital to shaping the final design; each subsequent step builds from the initial concept. Innovation-driven engineering problems require designers to focus heavily on early-stage design generation, with constant application and evaluation of design changes. Strategies to reduce the amount of time and effort designers spend in this phase could improve the efficiency of the design process as a whole. This paper seeks to create and demonstrate a two-tiered design grammar that encodes heuristic strategies to aid in the generation of early solution concepts. Specifically, this two-tiered grammar mimics the combination of heuristic-based strategic actions and parametric modifications employed by human designers. Rules in the higher-tier are abstract and potentially applicable to multiple design problems across a number of fields. These abstract rules are translated into a series of lower-tier rule applications in a spatial design grammar, which are inherently domain-specific. This grammar is implemented within the HSAT agent-based algorithm. Agents iteratively select actions from either the higher-tier or lower-tier. This algorithm is applied to the design of wave energy converters, devices which use the motion of ocean waves to generate electrical power. Comparisons are made between designs generated using only lower-tier rules and those generated using only higher-tier rules.

An Adversarial Agent-Based Design Method Using Stochastic Stackelberg Game Conditions

“…We anticipate the work presented can apply to any configuration design problem. Previous work has shown similarities in the behavior and characteristics of problem-solving spaces for a variety of configuration problems including wave energy converters, HVAC, and fluid networks [30,31]. To extend this method to these problems, defining adversary actions against the solution that the designer is configuring, which may include component failures or abnormal conditions, will be necessary.…”

Mapping artificial intelligence-based methods to engineering design stages: a focused literature review

“…In the conceptual design stage, researchers often used NLP (8 out of 28 papers) and deep learning (11 out of 28 papers) to translate design requirements to specifications, ideate designs and generate alternative design concepts for further analysis. In addition to these methods, we observed: three agent-based modeling methods applied to facilitate ideation (Maier et al 2020), to transfer design strategies from human to computer , and to generate concepts from standard grammatic representations and abstract designer heuristics (Puentes et al 2018); three machine learning methods for recommending design features (Yao et al 2017), evaluating design features (Vasantha et al 2022), and…”