Adaptive Generative Design Studio for Sustainable Artifacts using AI-Driven Prototyping

Thesis Abstract

This study addresses the growing need for rapid, sustainable artifact design by integrating adaptive generative design processes with AI-driven prototyping to reduce material waste, energy consumption, and time-to-market in contemporary art and design practice. The central aim is to develop and validate an integrated design studio workflow that automatically adapts generative design outputs to sustainability constraints while enabling iterative prototyping through AI-enabled fabrication pipelines. Specific objectives are to (1) engineer a modular generative design framework that encodes material, production, and lifecycle constraints; (2) implement AI-driven prototyping agents that translate digital designs into manufacturable prototypes with minimal waste; (3) evaluate the performance of the studio in producing sustainable artifacts across three case studies representing wood, recycled metals, and bioplastic composites; (4) measure perceptual and functional quality of artifacts against traditional design approaches; and (5) derive design guidelines and ethical considerations for responsible AI in art and design. The methodological approach adopts a mixed-methods design, combining quantitative performance metrics with qualitative insights. The population comprises professional designers and postgraduate students from a renowned art and design faculty, with a purposive sample of 60 participants for the quantitative strand and 24 participants for in-depth qualitative exploration. An experimental design is employed where participants engage with two workflows (a) conventional generative design without sustainability constraints, and (b) adaptive generative design with AI-driven prototyping. Instrumentation includes a standardized artifact evaluation rubric, material usage and energy footprints tracked via embedded IoT sensors, and a 20-item Likert-scale survey assessing perceived usability and creative satisfaction. Data collection also comprises semi-structured interviews and think-aloud protocols to capture cognitive processes during design iterations. Validity is addressed through triangulation, expert panel validation of the constraint encoding, and pilot testing with 10 participants. Reliability is ensured via inter-rater reliability checks on artifact evaluations (Cronbach’s alpha > 0.80) and test–retest reliability for the survey instrument (r > 0.85). Data analysis employs a combination of statistical and thematic techniques. Descriptive statistics summarize design cycle times, material waste, and energy consumption. Inferential analyses include multivariate analysis of variance (MANOVA) to examine differences in artifact quality and sustainability metrics between workflows, followed by regression analyses to identify predictors of waste reduction and production efficiency. Thematic analysis analyzes interview and think-aloud data to surface design rationale, constraints handling, and user experience nuances. A conceptual model is developed to illustrate the relationships among adaptive generative processes, AI-driven prototyping actions, sustainability outcomes, and perceived design quality, drawing on the Theory of Planned Behavior to interpret intention and behavior in design decision-making. Expected findings indicate that the adaptive generative design studio reduces material waste by 28–42% and energy use in prototyping by 15–25% for case-study artifacts without compromising functional performance or aesthetic quality. The AI-driven prototyping agents are anticipated to shorten iteration cycles by 35–50% relative to conventional workflows and to enhance consistency in meeting imposed lifecycle constraints. Qualitative insights are expected to reveal improved designer agency, better transparency in constraint trade-offs, and nuanced considerations related to artifact legibility and cultural value when deploying AI in creative practice. The study contributes to knowledge by operationalizing a pragmatic, scalable framework that couples constraint-aware generative design with intelligent prototyping, offering empirical evidence of sustainability gains in art and design workflows and enriching theoretical discourse on AI mediation of creativity, material economy, and collaborative design. Conclusions suggest that the integrated studio model advances sustainable artifact production while preserving creative agency and perceptual quality. The recommendations emphasize refining constraint encoding, expanding material datasets for lifecycle assessment, developing standardized evaluation protocols for AI-assisted design in art, and aligning practices with ethical guidelines for responsible AI usage in creative domains. Suggestions for further research include longitudinal studies on market impacts, cross-cultural validations, and the integration of responsive feedback mechanisms that adapt to user preferences and evolving sustainability standards.