en.Wedoany.com Reported - Recently, the State Administration for Market Regulation of China organized leading scientific research forces to tackle key technologies, achieving a breakthrough in building a knowledge base for quality safety risk control of intelligent products. This research addresses industry challenges such as the complex risk mechanisms of intelligent products and the lack of basic knowledge bases, establishing a quality safety analysis framework for intelligent interconnected products to provide data-driven support for risk identification, mechanism analysis, and regulatory decision-making.
The research team incorporated "information dimensions" and "intelligent logic" into the product quality safety analysis process, proposing a "personnel-product-environment-information" four-in-one framework. Compared with the traditional "human-machine-environment" model, the new framework further covers risk sources in data interaction, algorithm decision-making, network connectivity, and usage scenario changes of intelligent products, making it more suitable for handling complex safety issues of products such as smart homes, intelligent terminals, and interconnected devices.
In terms of the underlying construction of the knowledge base, the team independently developed an iterative knowledge extraction framework based on AI large language model technology. This framework has constructed a quality safety knowledge graph covering 10 typical types of intelligent products, over 50,000 basic data entries, 16,000 nodes, and 32,000 relationships, and has structurally depicted the causal evolution chain between "hazard sources—safety incidents—control measures."
The key value of this achievement lies in transforming scattered product failure information, risk scenarios, control measures, and regulatory knowledge into a computable, traceable, and expandable knowledge system. For the quality safety supervision of intelligent products, the knowledge base can not only be used for risk attribution but also support risk correlation analysis across different levels of products, components, and system architectures.
Currently, the knowledge base system has functions such as risk visualization and intelligent decision-making Q&A, capable of supporting technical research and development in areas like failure mechanisms and fault-tolerant control of intelligent interconnections, and health and safety evaluation of smart homes. The system can extend from underlying key components to the overall architecture, assisting in identifying potential quality safety risks and providing tool support for enterprises to conduct preemptive warnings and rapid responses.
The research results have entered the pilot application stage. As intelligent products shift from standalone devices to networked, platform-based, and algorithm-driven systems, quality safety supervision also needs to gradually extend from post-event sampling inspection to risk modeling, dynamic identification, and process control. Going forward, whether this knowledge base can be expanded to more categories of intelligent products, enterprise quality management systems, and regulatory application scenarios will determine its actual scope of support for the safety governance of intelligent products.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
