On the morning of September 6, the parallel session dedicated to the equipment manufacturing industry took place at the 2026 Lean Digital Innovation Conference and Lean Digital and Intelligent Transformation Innovation Ecosystem Expo. Liu Mingzhong, First Vice President of the China Enterprise Reform and Development Society and former Party Secretary and Chairman of China First Heavy Industries, along with Academician Ling Wen of the Chinese Academy of Engineering, and guests from benchmark enterprises including Tiandi Shanghai Mining Equipment, Wafangdian Bearing Group, China Railway Construction Heavy Industry, Jiangsu Nuclear Power, and Xi'an XD Transformer, gathered to share cutting-edge insights and frontline practices around the theme "How AI and Intelligence Can Truly Land in Manufacturing," exploring pathways to elevate capabilities across the entire value chain of the equipment manufacturing industry.
In his opening remarks, Liu Mingzhong outlined a strategic blueprint for the high-quality development of the equipment manufacturing industry from the perspective of enterprise reform and governance. He stated that the high-quality development of manufacturing cannot be achieved without reform-driven momentum and institutional innovation, emphasizing that lean management serves as the foundational framework for the effective implementation of AI.
Academician Ling Wen delivered a keynote speech titled "Practices in Multi-Agent Collaborative Technology," systematically elaborating on the latest advancements in this field. He pointed out that industrial AI is transitioning from single-point breakthroughs to systematic collaboration, with the key lying in building an open innovation platform for industrial intelligent agents that features "full-stack open source and low-threshold sharing." By establishing a national standard for intelligent agents that integrates "perception, decision-making, and execution" and developing a full-cycle automated testing and evaluation system, institutional barriers preventing enterprises from adopting these technologies can be dismantled, accelerating the deployment of multi-agent systems in complex industrial scenarios.
Benchmark practices: Presenting a real, measurable account of implementation results
Multiple leading enterprises shared data-backed digital and intelligent practices across the entire value chain. Jiang Shi, Deputy General Manager of Tiandi Shanghai Mining Equipment, presented "Lean Digital and Intelligent Implementation Practices in Large-Scale Equipment Manufacturing." He emphasized that "lean management is the prerequisite and foundation for digital and intelligent transformation." The company has built an intelligent factory with a total investment of RMB 569 million, comprising six digital workshops, 11 intelligent production lines, and 23 intelligent units, including the first automated roller welding line and the industry's first "lights-out" inspection line in China. Measured results show a 35% improvement in typical roller welding efficiency and a 90% welding reachability rate. Through continuous lean improvements, spare parts on-time delivery rates rose by 63%, production plan achievement rates increased by 22%, and production cycles for major components shortened by 15%, cumulatively generating over 3,500 improvement proposals and more than 400 institutional standards.
Liang Shuang, Director of the Lean Office at Wafangdian Bearing Group, focused on "Decision Models and Dynamic Scheduling of AI in Integrated Linkage Planning," sharing how the company restructured order fulfillment business flows and uncovered hidden waste through data and rule governance, integrated planning linkage, and AI collaboration. The presentation illustrated AI's evolution from a "support tool" to a "decision engine."
Wang Xiaoteng, Dean of the Intelligent Manufacturing Research and Design Institute at China Railway Construction Heavy Industry, shared "Experience in Intelligent Manufacturing for Underground Engineering Equipment." Addressing the challenges of tunnel boring machines characterized by "one machine, one strategy, and simultaneous surveying, design, and manufacturing," the company, in partnership with ABERY, advanced three phases of lean projects to build an AI-driven parallel collaborative intelligent manufacturing model for critical components. AI-based process design reduced process generation time by over 90%, intelligent welding equipment efficiency improved by 30%, and the first-pass qualification rate for key welds reached over 98%. The first domestic flexible assembly line for main drives shortened delivery cycles by 35.56% and reduced quality issues per unit by 47.9%, and was included as a typical use case in the ISO 23247-6 international standard. After three phases, the average achievement rate for first-level key nodes reached 95.3%, and the total manufacturing overhead rate declined by 6.66%
In the energy equipment sector, Zhang Xianggui, Deputy General Manager of Jiangsu Nuclear Power, shared "CNNC: Equipment Operations and Maintenance × AI." As the world's largest nuclear power base by total installed capacity, the Tianwan nuclear power base has achieved "one-chart sensing" of equipment status through digitalized flowcharts (structuring over 20,000 flowcharts), digitalized procedures, intelligent patrol inspections, digitalized isolation, and 4D maintenance of critical equipment. AI-powered smart site management achieves a violation detection accuracy rate of over 95%. He stressed that, given the high-consequence and strictly regulated nature of nuclear power, AI must adhere to the "safety bottom line" and maintain "human-in-the-loop" principles, advancing with a cautious approach.
Wang Zhong, Deputy General Manager of Xi'an XD Transformer, presented "Lean Management Enhancement in Power Transmission and Transformation Equipment." The company adopted a balanced, pull-based integrated linkage plan as its core approach, building continuous-flow production lines in its ultra-high-voltage workshop with targets of 20% capacity increase, 20% production cycle reduction, and 20% reduction in work-in-process capital, providing a model for lean digital and intelligent transformation in discrete heavy equipment manufacturing.
Methodology and talent: Solidifying the "foundation" for AI scale-up
Liu Zechen, Technical Director of the Equipment Division at ABERY Technology, delivered a keynote on "Lean Full-Value-Chain Management Empowering AI Transformation and Upgrading in Equipment Manufacturing." He argued that the difficulty in implementing AI in manufacturing does not lie in technology but in the simultaneous obstruction of four breakpoints: "data × standards × processes × talent." He noted, "Buying the most expensive AI while running on the most chaotic processes—AI does not eliminate chaos; it amplifies it." He proposed a pathway of "lean sets standards, AI enhances efficiency," where lean provides AI with learning objects and AI accelerates lean, achieving systematic implementation through scenarios across the full value chain, including strategic deployment, agile R&D, OTD fulfillment, procurement and supply chain, cost control, and site standardization. In cases shared from China Railway Construction Heavy Industry and Dalian Wafangdian Bearing, the order fulfillment cycle for boring machines was shortened by 15.7%, and annual cost savings at Wafangdian Bearing reached RMB 45 million.
Cai Jin, Senior Consultant at ABERY Talent Development Institute, discussed "Cultivating AI Composite Talent." He noted that the success of an AI strategy fundamentally comes down to the success of talent and organization. He introduced a five-in-one talent cultivation paradigm of "assessment, learning, training, practice, and evaluation," along with the new role of "Lean AI Application Specialist"—a bridge between business and technology—serving both as a "translator" who understands business needs and as an "actuary" who precisely identifies high-value scenarios, thereby building a talent engine for enterprise-wide AI transformation at scale.
In closing, Wang Ping, Chief at ABERY Technology, reviewed and commented on the day's case studies, synthesizing the common patterns: lean is the foundation, and AI is the engine. When lean management meets artificial intelligence, the equipment manufacturing industry is witnessing a profound efficiency revolution. Attendees noted that the forum, through the deep convergence of academician-level frontier insights and real-world cases from central state-owned enterprises, provided a replicable methodology and roadmap for the genuine implementation of AI and intelligence in equipment manufacturing, collectively advancing the industry's high-quality development in the AI era.