A Data-Driven and Experience-Driven Iterative Design Method for Intelligent Products

Yiming Zhao

doi:10.54097/9dq0rz24

Authors

Yiming Zhao

DOI:

https://doi.org/10.54097/9dq0rz24

Keywords:

Data-Driven; Experience-Driven; Intelligent Products; Iterative Design; DECI Algorithm; Dual-Modal Fusion; Dynamic Weighting Mechanism; Closed-Loop Iteration.

Abstract

To address the insufficient integration of data and experience-driven approaches in the iteration of intelligent products, this study designs an original DECI algorithm, constructs an integrated iterative design framework, and conducts experimental verification. The DECI algorithm achieves subjective experience quantification and dynamic dual-modal collaboration through three innovations: an experience intent embedding module, an adaptive dynamic weighting mechanism, and a closed-loop iterative update strategy, overcoming the limitations of static fusion. A four-layer progressive framework is built, covering data acquisition, feature processing, fusion iteration, and effect output layers, opening up the dual-modal fusion path and forming a closed-loop iterative chain throughout the entire process. Using intelligent office software and mobile service apps as examples, multi-dimensional experiments were conducted with three comparative algorithms. The results show that the DECI algorithm reduces response time and error rate by 10.1%-40.5% compared to the best comparative algorithm, increases user satisfaction by 4.8%-23.4%, shortens the iteration cycle by 16.1%-23.9%, and controls performance fluctuation in high-concurrency scenarios within 2.1%, with all differences being statistically significant. This study constructs an iterative scheme that balances quantitative accuracy and subjective experience, providing theoretical support and an engineering paradigm for the efficient iteration of intelligent products.

References

[1] Luo, S. Gong, H.B. & Lin, W. Research status and progress of intelligent product interaction design. Journal of Mechanical Engineering (J. Mech. Eng.), Vol. 59(2023) No. 11, p. 1-15.

[2] Lu, W.H. Ni, Y.H. Cai, Z.B. & Liu, R.J. User review data-driven product optimization design method. Journal of Computer-Aided Design & Graphics (J. Comput. Aided Des. Comput. Graph.), Vol. 34(2022) No. 3, p. 482-490.

[3] Zhao, X. Ren, S. Zhang, G. & Zhang, Y.F. Research on a new model of complex product improvement design driven by operation and maintenance data and multimodal knowledge. Journal of Mechanical Engineering (J. Mech. Eng.), Vol. 61(2025) No. 3, p. 91-104.

[4] Lai, X.J. Zhang, C.K. Li, M. Huang, J.Q. Liu, Z.R. & Fu, H. Intelligent cockpit emotional design: a multimodal data-driven method of use-speech-psychology joint analysis. Industrial Engineering (Ind. Eng.), Vol. 28(2025) No. 1, p. 49-60.

[5] Hu, B.T. Feng, Y.X. Liu, J.H. Guo, W. & Tan, J.R. Research on intelligent adaptive design for customized products for "Internet+". Journal of Mechanical Engineering (J. Mech. Eng.), Vol. 59(2023) No. 12, p. 109-125.

[6] Yu, Z.Y. Zhao, W. Zhang, K. Wang, Y. & Yu, M. Knowledge push method to support product iteration design. Journal of Mechanical Engineering (J. Mech. Eng.), Vol. 58(2022) No. 18, p. 292-302.

[7] Lai, H.B. & Yang, Z.C. The impact of design-driven intelligent product innovation on perception, emotion and word-of-mouth communication from the perspective of scenario. Science and Technology Progress and Countermeasures (Sci. Technol. Prog. Countermeas.), Vol. 40(2023) No. 24, p. 31-40.

[8] Tang, M. Research on the application of artificial intelligence data mining in the planning of intelligent vehicle enterprises. Science and Technology Exploration (Sci. Technol. Explor.), Vol. 1(2025) No. 4, p. 20-24.

[9] Guo, X. Huang, Z.C. Wang, J. Zhang, K. & Zhao, W. Construction of Knowledge Graph and Intelligent Reasoning Method for Product Design in Human-Computer Hybrid Interaction. Journal of Mechanical Engineering (J. Mech. Eng.), Vol. 61(2025) No. 3, p. 142-153.

[10] Sun, Z.L. Wang, K.F. & Gu, P.H. Review and Prospect of Design Theory and Method Research. Journal of Mechanical Engineering (J. Mech. Eng.), Vol. 60(2024) No. 13, p. 2-20.

[11] Shao, M.Y. From Craftsmanship to Intelligent Collaboration: Exploration of Product Design Practice Driven by AIGC. Humanities and Social Development (Hum. Soc. Dev.), Vol. 2(2025) No. 4, p. 39-42.

[12] Bai, C.J. Xu, H.Z. & Li, X.L. Embodied Intelligence Driven by Large Models: Development and Challenges. Science in China: Information Science (Sci. China Inf. Sci.), Vol. 54(2024) No. 9, p. 2035-2082.

[13] Zhang, F. & He, J.X. Research on Scene-Driven User Innovation Methods and Paths in the Internet of Things Environment. Science and Technology Progress and Countermeasures (Sci. Technol. Prog. Countermeas.), Vol. 41(2024) No. 6, p. 1-10.