Designing the Future: Big Data and AI Revolutionize Product Innovation

By Dr Silpaja Chandrasekar, PhDReviewed by Susha Cheriyedath, M.Sc.Aug 22 2023

In a paper published in the journal Applied Science, researchers discussed how the increased living standards had led to a heightened desire for customized products. Limited by subjectivity and inadequate data, traditional design techniques had proven insufficient. However, recent strides in big data and artificial intelligence (AI) introduced an innovative product design paradigm.

This novel approach harnessed the variety of data types to foster intelligent design solutions, effectively mitigating historical constraints. The present study offered a comprehensive investigation of this method, underscoring the significant potential of big data and AI-infused design while delineating directions for future research endeavors.

Background

The development of competitive products that exceed consumer expectations is central to enterprise activities. Successful products enhance user satisfaction, stimulate purchases, and drive market competitiveness. Over the past few decades, product design has evolved from functional to emotion-centered and user-centric approaches, spurred by technological advancements. However, market competition demands next-generation innovation driven by big data, deep learning, and AI algorithms. These advances revolutionize understanding, pattern recognition, and generative design across various data types.

The focus of the consumers on personalized experiences has led to increased attention to product appearance alongside functionality. Traditional design methods face challenges in gathering user needs and effectively assessing product performance through Big Data and AI. Manual surveys consume much time, while relying on intuition is subjective. Recent research suggests that big data and AI can address these limitations, especially with AI-Generated Content (AIGC). The Internet's growth has generated a vast amount of data within the product lifecycle, offering insights into user preferences and market trends. Processing this data presents challenges but also opportunities for better product design.

Big data in product lifecycles encompass diverse data types and high volumes, best processed by AI algorithms like Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs). The intersection of big data and AI-driven product design is gaining traction. The present paper reviews the applications of big data in product design, offering theoretical foundations for future research. It aims to streamline product advancement, improve user contentment, and foster automation and intelligence in design.

Review of Product Design Approaches and Technologies

Crucial Stages in Product Design

The product design process involves nine key tasks: product vision definition, market research, competitive research, user research, idea generation, feasibility analysis, sketching, prototyping, and scheme evaluation. Product vision sets the overall goal, while market research explores trends and user demand. User research, supported by data, captures preferences and satisfaction using Natural Language Processing (NLP) and emotion recognition. Idea generation involves logical and intuitive thinking, while feasibility analysis considers technical, economic, and environmental aspects. Sketches and prototypes visualize ideas aided by AI-driven methods. Evaluation gathers user feedback for refining and verifying the design.

Methods in User-Centric Design

Several user-centric design methods are important in product innovation. Kansei engineering shifts focus to sensory aspects in product design, involving collecting and evaluating user emotions related to product attributes. The Kano Attractive Quality Model (KANO) categorizes user requirements based on satisfaction levels, while Quality Function Deployment (QFD) translates these needs into technical attributes using the House of Quality. The Theory of Inventive Problem Solving (TRIZ) generates innovative solutions by addressing contradictions. These methods offer valuable insights into user needs and innovative product development.

Utilizing Big Data and AI for Product Design

Product design benefits from structured, semi-structured, and unstructured big data. Structured data adheres to predefined models, while semi-structured data offer flexibility. Unstructured data, like images and text, have no fixed structure. Image data inspires and generates designs through techniques like GAN and neural style transfer. GAN creates designs by competing generators and discriminators. Neural style transfer combines content and style. Challenges include quality control and innovation. A balance of AI and traditional design knowledge is crucial for effective product design.

Audio data is valuable for product design but lacks sufficient literature. AI technologies like speech recognition, speaker identification, and emotion recognition can be useful. Speech recognition converts speech to text, while speaker identification identifies individuals based on their voice. Emotion recognition detects emotions expressed in speech. These technologies face challenges like variability in voiceprints and noise interference. Video data is also valuable, enabling insights into product usage. Intelligent video analysis aids motion detection, summarization, retrieval, and more. Challenges include complex activity analysis and consistent data acquisition from multiple cameras. Overall, while audio and video data hold potential for product design, they are limited by processing and data quality challenges.

Conclusion

In summary, in the knowledge-driven economy, rapidly changing customer demands and shorter product lifecycles require innovative approaches to product design. Big data and AI offer unprecedented potential for driving innovation throughout the lifecycle. This survey provides a comprehensive overview of how these technologies impact product design. By leveraging NLP, image processing, audio, and video data, product design benefits from improved user understanding, creative inspiration, user behavior observation, and virtual displays. This approach harnesses collective intelligence and advanced technologies to enhance product development efficiency, resulting in more competitive and intelligent innovations.