To establish a collaborative design methodology based on swarm intelligence theory, and to overcome the limitations of traditional linear design models. By leveraging intelligent technologies, it seeks to achieve exponential improvements in design efficiency and promote the digital transformation and value chain restructuring of the custom furniture industry, with a focus on custom wardrobes.

A progressive research approach of “theoretical modeling, system construction, and empirical validation” is adopted, concentrating on custom wardrobe application scenarios. A three-layer collaborative architecture based on swarm intelligence is established. The design tasks are decomposed into three core modules: 1) user data acquisition and demand visualization module; 2) design knowledge function module; and 3) user collaborative design automatic generation module.

The swarm intelligence collaborative design system effectively addresses the contradiction between processing massive personalized design demands and delivering efficient service supply, confirming the feasibility of empowering the digital transformation of traditional manufacturing through intelligent technologies. The layered and decoupled architecture successfully achieves the integrated innovation of three key technologies: structured design knowledge (knowledge graph), intelligent demand analysis (deep learning), and automated solution generation (GAN algorithm).

At the theoretical level, a “demand-knowledge-solution” ternary collaborative model is proposed, expanding the application paradigm of swarm intelligence theory in the field of industrial design. At the practical level, the modular technical architecture developed for custom wardrobes offers a reusable solution for the industry's digital transformation. This design provides a new technological pathway for the intelligent upgrading of the home furnishing industry. To further promote industrial collaboration towards a higher dimension, efforts should focus on constructing cross-enterprise collaborative ecosystems and innovating cloud-based intelligent service models.

To accurately capture the needs of elderly individuals regarding cabinet usage by constructing a structural equation model to analyze the key factors influencing the design of age-friendly cabinets. The results provide important data support for the design of age-friendly cabinets and the formulation of corresponding product marketing strategies.

This study focused on elderly individuals aged 60 and above, using a questionnaire survey method. A total of 457 valid questionnaires were collected to gather data on the independent variables affecting the design of age-friendly cabinets. The Likert 5-point scale was used to analyze and quantify the impact of each indicator on the design of age-friendly cabinets. Then, 36 design indicators were selected from six aspects: materials, dimensions, layout, functionality, color, and intelligence. Innovation was considered as a mediating variable, and consumer willingness as the dependent variable. A partial least squares structural equation model (PLS-SEM) was employed to empirically analyze the key factors influencing the design of age-friendly cabinets.

1) In the reliability and validity test of the scale, the cronbach's alpha values of the eight dimensions-layout, materials, dimensions, functionality, color, intelligence, innovation, and consumer willingness were all above 0.86, indicating strong internal consistency among the measurement items. The composite reliability values exceeded 0.8, further confirming the scale's high reliability. The average variance extracted (AVE) values were all above 0.7, demonstrating good convergent validity. The measurement model of aging cabinet design constructed in this study has good performance in terms of reliability and validity; 2) The scale was verified using AMOS software, and the square roots of AVE for each dimension were higher than the square values of its correlations with all other variables, proving good discriminant validity; 3) In the PLS-SEM model, the factor loading coefficients of the dimensions were relatively high, but there were differences in the loading coefficients of various indicators within their respective dimensions. Indicators with higher factor loading coefficients should be prioritized, as they provide stronger explanatory power for latent variables; 4) Hypothesis testing revealed that the path coefficients of H1-H6 passed the significance test (P < 0.05), and the indirect effects of H7-H12 on different paths passed the significance test (P < 0.05), indicating that the previous hypothesis was established at a certain significance level.

A series of crucial influencing factors for the design of age-friendly cabinets were identified. Among these, functionality had the highest influence on purchase willingness, while the influence of intelligent features was relatively lower. The product's functional characteristics emerged as one of the most critical factors in the design of age-friendly cabinets, providing a theoretical basis for the design and marketing strategies of age-friendly cabinets.