Nanjing Iron & Steel Co., Ltd. (NISCO) has commercially deployed the world's first steel model powered by swarm intelligence—the YuanYe Steel Large Model (YuanYe)—across multiple scenarios at scale, in collaboration with Huawei and ecosystem partners. The deployment marks a significant shift in how traditional manufacturers approach digital transformation and cost efficiency, embedding AI into every core stage of production: ironmaking, steelmaking, rolling, and quality inspection.
Unlike general-purpose monolithic models, YuanYe fuses multiple AI models with the steel industry's metallurgical mechanism models to create a dual-drive, multi-model cluster architecture. This architecture features a "right brain" for probabilistic computing and a "left brain" for precise and scientific decision-making. Powered by massive amounts of sensing data aggregated from millions of operational technology (OT) data points across 26 business domains, NISCO has achieved real-time digital twin mapping within its intelligent operations center. As a result, 100% of NISCO's key equipment is now connected and digitally controlled, and its new product R&D cycles have been shortened by 50%, reducing annual costs by over CNY100 million (approx. US$15 million). The company had launched 25 business agents by October 2026, covering scenarios from visual quality inspection to production scheduling.
The impact is particularly notable in four core scenarios. In ironmaking, YuanYe performs real-time dynamic simulations of blast furnaces, enabling operators to predict temperatures and identify issues two hours in advance with over 90% accuracy. This has increased prime-grade molten iron yield by over 10% while significantly cutting unit energy consumption. In rolling, the model's rolling-force prediction achieves millisecond-level response, narrowing thickness tolerance to ±0.2 mm and reducing steel thickness deviation to 1.9%. The rolling-force hit rate has jumped from 88% to 98%, supporting annual output of over 2.5 million tons of steel plates and increasing overall yield rate by 0.5%. In quality inspection, an integrated smart metallographic inspection system has doubled efficiency, raising inspection accuracy to over 95% and reducing per-sample inspection time from 10 minutes to 3–5 minutes. In energy and environmental control, the S-Link Intelligent Agent for Gas and Flue Gas Sulfur Control issues sulfur-risk warnings one hour in advance with over 95% accuracy, slashing natural gas co-firing consumption by nearly 50% and reducing downtime caused by emission exceedance by over 25%, delivering more than CNY600,000 (approx. US$90,000) in annual economic benefits.
Li Jinyan, General Manager of the Digital and Intelligent Integrated Development Department and Director of the AI Research Institute at NISCO, emphasized the human-centric approach: "Digital and intelligent transformation is inherently human-centric. Its purpose is not to replace people, but to unleash human potential." To date, NISCO has deployed over 300 AI models across its plants, alongside more than 1,000 low-code agents and lightweight applications built independently by frontline employees. Cheng Yongfu, Senior Industrial Solution Expert at Huawei's Steel & Non-ferrous BU, noted: "The real value of industrial AI can only be realized when it expands from single-point applications to comprehensive integration with production, operations, and management."
Looking ahead, NISCO plans to explore dedicated data centers for smart steel plants with Huawei and other partners, centered on an AI-native three-tier architecture. The company aims to provide a replicable benchmark for the digital and intelligent transformation of the global steel and manufacturing industries. This announcement, reported via Media OutReach Newswire, underscores the growing role of swarm intelligence in industrial settings and signals a new era of data-driven steel production with significant implications for cost, quality, and environmental performance.
