Bradley Camburn

As AI continues to evolve and demonstrate new design capabilities, there is an opportunity to explore its integration with design at a process level. While recent studies have examined the creative potential of AI-based design tools and their use in enhancing individual tasks, there remains an opportunity to understand how the design process might evolve at the end-to-end design workflow macro scale through study of practice-based design. We follow a revised design process workflow, alongside an exemplar implementation using interactive digital workspaces, establishing an applied approach for viewing the design process as a dialogue between humans and AI. Data were collected from three-day continuing education workshops with industry professionals, providing an empirical dataset that offers longitudinal insights into how the proposed workflow performs in real-world contexts. The approach is evaluated through method-level comparisons and a longitudinal study providing process-level evaluation between human-only and AI-augmented workshop flows, comprising four workshops and 20 teams. Two methods, affinity clustering and mind mapping, were sampled for direct comparison. Participants preferred the AI-augmented workflow, and method-level comparisons indicate comparable quality. The primary contribution is a practice-based, longitudinal account of integrating AI across an end-to-end professional design workflow.

We present a multi-agent framework for generating physics simulation code from natural language descriptions, featuring a novel perceptual self-reflection mechanism for validation. The system employs four specialized agents: a natural language interpreter that converts user requests into physics-based descriptions; a technical requirements generator that produces scaled simulation parameters; a physics code generator with automated self-correction; and a physics validator that implements perceptual self-reflection. The key innovation is perceptual validation, which analyzes rendered animation frames using a vision-capable language model rather than inspecting code structure directly. This approach addresses the ``oracle gap'' where syntactically correct code produces physically incorrect behavior--a limitation that conventional testing cannot detect. We evaluate the system across seven domains including classical mechanics, fluid dynamics, thermodynamics, electromagnetics, wave physics, reaction-diffusion systems, and non-physics data visualization. The perceptual self-reflection architecture demonstrates substantial improvement over single-shot generation baselines, with the majority of tested scenarios achieving target physics accuracy thresholds. The system exhibits robust pipeline stability with consistent code self-correction capability, operating at approximately \$0.20 per animation. These results validate our hypothesis that feeding visual simulation outputs back to a vision-language model for iterative refinement significantly outperforms single-shot code generation for physics simulation tasks and highlights the potential of agentic AI to support engineering workflows and physics data generation pipelines.

This pilot study traces students' reflections on the use of AI in a 13-week foundational design course enrolling over 500 first-year engineering and architecture students at the Singapore University of Technology and Design. The course was an AI-enhanced design course, with several interventions to equip students with AI based design skills. Students were required to reflect on whether the technology was used as a tool (instrumental assistant), a teammate (collaborative partner), or neither (deliberate non-use). By foregrounding this three-way lens, students learned to use AI for innovation rather than just automation and to reflect on agency, ethics, and context rather than on prompt crafting alone. Evidence stems from coursework artefacts: thirteen structured reflection spreadsheets and eight illustrated briefs submitted, combined with notes of teachers and researchers. Qualitative coding of these materials reveals shared practices brought about through the inclusion of Gen-AI, including accelerated prototyping, rapid skill acquisition, iterative prompt refinement, purposeful "switch-offs" during user research, and emergent routines for recognizing hallucinations. Unexpectedly, students not only harnessed Gen-AI for speed but (enabled by the tool-teammate-neither triage) also learned to reject its outputs, invent their own hallucination fire-drills, and divert the reclaimed hours into deeper user research, thereby transforming efficiency into innovation. The implications of the approach we explore shows that: we can transform AI uptake into an assessable design habit; that rewarding selective non-use cultivates hallucination-aware workflows; and, practically, that a coordinated bundle of tool access, reflection, role tagging, and public recognition through competition awards allows AI based innovation in education to scale without compromising accountability.

Fiber-reinforcement of structures built with additive manufacturing (AM) has been the focus of many recent innovations in the AM space. These innovations have further enabled AM’s progress from a prototyping technology to a viable manufacturing process for end use parts under normal load conditions. This paper introduces a set of design principles to drive innovation and the creation of novel AM technologies capable of producing fiber-reinforced components. The principles were developed through a comprehensive survey of the literature and patents to identify fundamental design insights that made the published achievements and innovations possible. From the collected body of insights, 23 unique design principles are identified. These principles are formalized and organized within the framework of a matrix representation, connecting established fiber-reinforced composite guidelines to specific sub-functions of AM. Finally, a fiber-reinforced additive manufacturing machine is developed by employing a subset of the identified principles as a demonstration of their utility.

Additive manufacturing and computational modeling approaches have enabled the production of complex and intricate parts. Yet, there has been relatively lesser exploration of how these emerging technologies can support the production of very large parts and assemblies. This paper will explore an approach to segmentation using recursion. Segmentation is the division of a large model into smaller modules using a set of design rules. In recursion, a process is called within a sub-step of the process itself. In this case, a model is segmented, then those initial modules are re-segmented to meet a design criterion through a recursive process. A typical challenge is that printers often have fixed volume print beds. While larger printing systems have been developed or those which are extensible in a single dimension there are opportunities to explore how alternative approaches, such as modular segmentation may support large scale part production with additive manufacture. Previous work has introduced one mechanism for segmentation of large parts via computational algorithm to support large scale prints.

Economic use of early-stage prototyping is of paramount importance to companies engaged in the development of innovative products, services, and systems because it directly impacts their bottom line. There is likewise a need to understand the dimensions, and lenses that make up an economic profile of prototypes. Yet, there is little reliable understanding of how resources expended and views of dimensionality across prototyping translate into value. To help practitioners, designers, and researchers leverage prototyping most economically, we seek to understand the tradeoff between design information gained through prototyping and the resources expended prototyping. We investigate this topic by conducting an inductive study on industry projects across disciplines and knowledge domains while collecting and analyzing empirical data on their prototype creation and test processes. Our research explores ways of quantifying prototyping value and reinforcing the asymptotic relationship between value and fidelity. Most intriguingly, the research reveals insightful heuristics that practitioners can exploit to generate high value from low and high fidelity prototypes alike.

Prototyping is interwoven with nearly all product, service, and systems development efforts. A prototype is a pre-production representation of some aspect of a concept or final design. Prototyping often predetermines a large portion of resource deployment in development and influences design project success. This review surveys literature sources in engineering, management, design science, and architecture. The study is focused around design prototyping for early stage design. Insights are synthesized from critical review of the literature: key objectives of prototyping, critical review of major techniques, relationships between techniques, and a strategy matrix to connect objectives to techniques. The review is supported with exemplar prototypes provided from industrial design efforts. Techniques are roughly categorized into those that improve the outcomes of prototyping directly, and those that enable prototyping through lowering of cost and time. Compact descriptions of each technique provide a foundation to compare the potential benefits and drawbacks of each. The review concludes with a summary of key observations, highlighted opportunities in the research, and a vision of the future of prototyping. This review aims to provide a resource for designers as well as set a trajectory for continuing innovation in the scientific research of design prototyping.