Development of an artificial-intelligence-driven product design evaluation model using multi-modal data

Abstract

This thesis systematically explores the potential and practical applications of artificial intelligence (AI) in enhancing product design evaluation. As AI technology advances, its integration with design processes offers new approaches to improve efficiency, innovation, and decision-making in product design. This study focuses on developing and validating a multimodal AI-assisted product design evaluation system, leveraging deep learning algorithms that can process and analyze complex multimodal data, images and text, to provide comprehensive evaluation metrics. The contribution of this study is the establishment of a framework that integrates AI with traditional product design evaluation practices. The research is organized around five interrelated studies, each of which addresses a specific application of AI in product design--from data collection and model development to practical validation and human-machine collaboration. The results show that AI can significantly improve the objectivity and efficiency of design evaluation, especially when dealing with large-scale and multidimensional datasets. AI's ability to synthesize large amounts of design-related information has been shown to significantly enhance the decision-making process, allowing for faster iterations and more informed adjustments during the design phase. However, the study also acknowledges the challenges of AI in its application, especially in dealing with highly subjective design aspects such as aesthetics and user experience, where human insight remains indispensable. In addition, this paper proposes a human-machine collaborative model as an ideal approach for design evaluation, combining the analytical advantages of artificial intelligence with human creativity and critical judgment. This model not only improves the reliability of design evaluation, but also promotes innovation by integrating different perspectives. In terms of practice, the developed AI-assisted evaluation system has been tested and validated through empirical methods, showing usability and effectiveness results in real-world design scenarios. This study contributes to the theoretical and practical understanding of the application of artificial intelligence in design and suggests avenues for future innovation and ethical considerations for the integration of artificial intelligence technology in the design industry. Overall, this study not only highlights the transformative potential of artificial intelligence in product design evaluation, but also lays the foundation for future advances to promote smarter, more efficient, and user-centered design practices.