制造业本质上是物理变换过程，缺乏对物理环境的感知和执行 能力，AI就无法真正落地。鉴于制造链条涉及多供应商的设备 和机器人，AI需统筹具身智能间的协同，确保顶层集成。 作者： 刘劲等 封图：图虫创意 这两年，汹涌而来的人工智能（AI）浪潮让众多制造业企业家和管理层深感焦虑与迷茫。大家普遍意识到AI的重要性：不拥抱AI，担心被时代抛弃， 失去未来发展的入场券；可当真正拥抱AI时，却发现除了简单的应用或依托成熟技术的场景外，企业往往不知从何入手，难以系统性推进，即便尝试 后，效果也与预期相差甚远。麻省理工学院在2025年的一项调研显示，在众多尝试系统性利用AI的企业案例中，仅有约5%取得了成功。 AI在制造业中的应用现状 在理想状态下，端到端的智慧工厂里，AI将全面取代或主导人类在制造业价值链中的角色。从研发、设计、生产、营销到售后服务，所有环节均由AI 驱动或高度自动化。这不仅是为了提升效率，更是要实现无缝、预测性和自适应生产的全智能状态。 然而，理想愿景虽令人向往，但当前制造业的AI应用还远未达到端到端的智慧水平。大多数企业仍处于"点状智能"阶段，AI主要辅助特定环节，而非 系统性主导。 在销售服务环节，由于 ...

Core Insights - The company has implemented an advanced process control (APC) system integrated with industrial large models to optimize boiler operations, enhancing efficiency and environmental performance [1][2] - The traditional boiler control relied on manual experience, which led to inefficiencies in energy consumption and emissions [1] - The new intelligent control platform allows for data-driven, multi-variable coordination, significantly improving combustion and environmental optimization [2] Group 1: Technological Advancements - The integration of an AI-driven control system represents a significant upgrade from traditional automation, focusing on both economic and environmental benefits [2] - The system employs soft measurement and automatic online optimization mechanisms to create a coal coordination control strategy based on energy balance [2] - Real-time data collection enables dynamic adjustments to coal and air ratios, improving denitrification efficiency and reducing nitrogen oxide emissions [2] Group 2: Environmental and Economic Impact - Since the system's operation, it has reduced carbon dioxide emissions and decreased other pollutants such as nitrogen oxides, sulfur dioxide, and particulate matter by 556.4 kilograms [2] - The company anticipates annual cost savings of 843,000 yuan, achieving a balance between environmental protection and economic efficiency [2]

Market Performance - The three major indices collectively declined, with the Shanghai Composite Index closing at 4075.92, down 0.64% or 26.29 points [2] - The Shenzhen Component Index fell by 1.44%, closing at 13952.71, while the CSI 300 Index decreased by 0.60% to 4670.42 [2] - The ChiNext Index experienced the largest drop, closing at 3260.28, down 1.55% or 51.24 points [2] Sector Rankings - The top-performing sectors included Beauty Care (3.21%), Banks (1.57%), and Food & Beverage (1.31%) [3] - Conversely, the worst-performing sectors were Non-ferrous Metals (-4.57%), Electric Power Equipment (-3.41%), and Coal (-2.22%) [3] Market Outlook - The market showed weakness with all three major indices closing lower, particularly the ChiNext Index [4] - Consumer sectors such as Food & Beverage and Retail saw significant gains, while sectors like Non-ferrous Metals and Electric Power Equipment faced notable declines [4] - The report indicates that the market may experience a phase of oscillation with potential upward movement, while also highlighting the need for caution regarding short-term adjustments and profit-taking risks [6] News and Developments - The Ministry of Industry and Information Technology emphasized breakthroughs in key technologies such as computing power chips and industrial large models [5] - The People's Bank of China held a meeting focusing on building a multi-level financial service system to support key areas like domestic demand and technological innovation [5] - Developments in commercial aerospace are expected to enhance launch efficiency and reduce costs significantly [5]

Core Insights - The launch of the "Action Plan for the Integration of Industrial Internet and Artificial Intelligence" and the "Action Plan for Promoting High-Quality Development of Industrial Internet Platforms (2026-2028)" marks a significant shift in China's industrial internet sector, moving from basic connectivity to data-driven and intelligent decision-making [1] - Kaos IoT Technology Co., Ltd. has submitted its prospectus to the Hong Kong Stock Exchange, aiming to become the first publicly listed company in the "AI + Industrial Internet" space, reflecting a response to the evolving industry landscape [1] Business Structure - Kaos is transitioning from scale expansion to high-quality growth, with notable improvements in profitability and financial structure [2] - The company reported a net loss in 2023 but turned a profit of 65.14 million yuan in 2024, with net profit rising to 176 million yuan in the first nine months of 2025, indicating a significant value reconstruction [2] Revenue Drivers - Revenue is driven by two main engines: data intelligence solutions and IoT solutions, with the former's contribution increasing from 18.3% in 2023 to 29.0% in the first nine months of 2025 [3] - The green manufacturing solutions segment saw a remarkable growth, with revenue in the first nine months of 2025 exceeding 15 times that of the entire year of 2023, aligning with national "dual carbon" strategies [3] Profitability Improvement - The data intelligence solutions segment maintains a gross margin above 35%, indicating a shift towards high-value software, platform, and service areas [4] - The relationship with the controlling shareholder, Haier Group, has provided Kaos with valuable industrial application scenarios, which are crucial for its growth [4] Market Expansion - Revenue from independent third-party clients increased from 27.2% in 2023 to 41.1% in the first nine months of 2025, with over 9,500 paying clients across various industries [5] - Kaos ranks first in the Chinese market for platform-based industrial data intelligence solutions, providing a strong foundation for future expansion [5] Long-term Strategy - Kaos is building three differentiated "moats" to ensure sustainable profitability: platform capabilities for scale and efficiency, integration of industrial models with scenario-based intelligence, and a focused deepening strategy in key industries [6][7][8][9] - The platform's standardization and modularization enhance delivery efficiency and reduce marginal costs, crucial for profitability [7] - The integration of AI with specific industrial processes creates unique customer value, supporting ongoing margin improvements [8] - The company plans to use fundraising for potential investments and acquisitions, allowing for rapid capability enhancement and market entry [9] Conclusion - Kaos's journey towards public listing reflects a pivotal transition in the Chinese industrial internet sector, showcasing improvements in profitability and business structure [11] - The ability to convert technological expertise into scalable and replicable business success will be a key measure of the industrial internet's impact on the real economy [11]

Core Viewpoint - Kaos, the industrial internet platform under Haier Group, has officially submitted its application for listing on the Hong Kong Stock Exchange, aiming to become the ninth listed company in the Haier ecosystem, following a failed attempt to list on the A-share market [1][2]. Group 1: Company Overview - Kaos was established in April 2017 and has developed the COSMOPlat industrial internet platform, which ranks first in China's industrial data intelligent solutions market based on revenue in 2024 [2]. - The company has been preparing for its IPO for over five years, with a significant focus on industrial internet solutions while Haier Group concentrates on smart home businesses [2][5]. Group 2: Financial Performance - In 2023 and 2024, Kaos reported revenues of 4.994 billion yuan and 5.070 billion yuan, with net profits of -82.721 million yuan and 65.136 million yuan, respectively. For the first nine months of 2025, revenue reached 4.42 billion yuan with a net profit of 176 million yuan [3]. - The revenue from the IoT solutions segment was 3.14 billion yuan, a year-on-year increase of 10.86%, while the data intelligence solutions segment generated 1.28 billion yuan, up 59.64% year-on-year [3]. Group 3: Revenue Sources - Haier Group is the largest customer and supplier for Kaos, contributing 3.607 billion yuan, 3.421 billion yuan, and 2.549 billion yuan to Kaos's revenue in 2023, 2024, and the first nine months of 2025, respectively [3]. - Government subsidies have been a significant source of income for Kaos, amounting to 97.6 million yuan, 79.7 million yuan, and 50.5 million yuan in 2023, 2024, and the first nine months of 2025 [4]. Group 4: IPO Plans and Use of Proceeds - The funds raised from the IPO will primarily be used to enhance core platform capabilities, develop industrial models, expand market reach, and explore potential investment opportunities [5][6]. - The decision to shift from A-share to Hong Kong listing was influenced by regulatory uncertainties in the A-share market and the company's global expansion strategy [2][4].

Core Viewpoint - Kaos IoT Technology Co., Ltd. has submitted an application for listing on the Hong Kong Stock Exchange, aiming to enhance its global expansion strategy and strengthen its core platform capabilities [1][5]. Group 1: Company Overview - Kaos was established in April 2017 and is an industrial internet platform under Haier Group, having developed the COSMOPlat platform to assist enterprises in digital transformation [1]. - The company has achieved large-scale commercial application of its products and solutions across various verticals, serving over 9,500 paying customers [3]. Group 2: Financial Performance - Kaos's revenue is projected to increase from RMB 49.94 billion in 2023 to RMB 50.69 billion in 2024, with a revenue of RMB 44.21 billion reported for the first nine months of 2025 [3]. - The company has turned a net loss of RMB 82.72 million in 2023 into a profit of RMB 65.14 million in 2024, with a significant increase in net profit to RMB 17.6 million in the first three quarters of 2025, representing a 393% year-on-year growth [3][4]. - The net profit margin is expected to improve from -1.7% in 2023 to 4.0% by the end of September 2025, attributed to increased gross profit and operational efficiency [3]. Group 3: Market Strategy and Global Expansion - Kaos has shifted its listing strategy from A-shares to Hong Kong due to regulatory uncertainties in the A-share market and aims to leverage Haier Group's international network for global expansion [5]. - The company plans to enhance its market penetration in key industries and regions, focusing on emerging markets in the Americas, Southeast Asia, and the Middle East [5]. - As of September 2025, Kaos's solutions have been promoted in over 20 countries, serving more than 50 overseas enterprises [6]. Group 4: Shareholding Structure - Haier Group is the controlling shareholder of Kaos, holding approximately 78.04% of the voting rights, with direct ownership of 10.83% and additional stakes through subsidiaries [3]. Group 5: Industry Context - The listing of Kaos on the Hong Kong Stock Exchange is expected to expand the "Haier system," which currently includes multiple listed companies under Haier Group [7].

本报讯 （记者冯思婕）1月30日，杭州海康威视数字技术股份有限公司（以下简称"海康威视"）公众号 发布消息称，公司新上岗了一位包装配件"AI质检员"——海康观澜工业大模型产线场景方案。依托海康 观澜工业大模型能力，该方案即使面对工厂配件包装体量大、规格多变的复杂场景，也能精准识别配件 错放、漏放问题，并实时拦截风险。 据介绍，海康观澜工业大模型产线场景方案不需要复杂调试，就能直接上产线，实现分钟级部署。同 时，只要产线调整，它也会自动切换检测模型，实现"零延迟"换产，高效适配柔性产线。 以海康威视智造工厂为例，这里生产节奏快，工人平均每10秒就要完成一组包装；但频繁切换的产线， 又让配件规格、组合处于持续变化中；即便是最熟练的工人，也很难在高强度重复作业与多变的订单需 求交织下，始终确保配件不错漏。 为了守住配件齐套这道质量防线，以往，海康威视工厂采用"元素管控"方式，工人事先按每种配件10个 为一组分装；包装时按组取用，一组用完若有剩余，再开箱倒查。但这种方式既耗人力、又费时间，错 漏问题难以快速追溯。 但传感器只以"取件次数"判定包装是否完成，无法识别取件品类、数量是否匹配单套产品要求，且计数 未达标时无 ...

更重要的是，这套系统部署灵活，可以适配不同形态配件的识别需求： 如果配件形态简单，比如螺栓、支架，只需要 拍摄1-8张图片，无需额外训练，分钟级实现配件注册和算法部署。 即使是形态不规则的配件，比如塑料袋螺丝包，也 无需担忧——只需要拍摄少量样本照片启动大模型训练，系统就能自动捕捉配件特征，不管角度怎么变，都能精准识 别！03堆叠场景巧应对复杂工况稳拿捏真实的产线现场，情况往往更复杂。在海康威视网络工厂，网络摄像机的配件包 装并不是规整平铺，而是直接堆叠放进包装盒里。 面对这类配件堆叠的包装场景，大模型虽然不能"一眼看穿"，但会换个思路"解题"：通过内置的高精度预训练手部检测 大模型，精准识别工人的手部是否进入指定料盒区域，以及是否完成有效取件动作。一旦拿取次数不足或拿错料盒区 域，系统就会触发NG警报，督促工人复核。 凭借灵活的检测逻辑，这套检测系统让海康威视工厂配件包装实现100%全 检，质量效率得到双提升；更支持全流程可视化追溯，让质量问题有据可查。目前其已广泛覆盖汽车零部件、电子产品 组装、家电产品总装、医药用品分拣等场景。 财经频道更多独家策划、专家专栏，免费查阅>> 配件包装看似简单，但以柔性产线 ...

大小通吃，这碗饭谁都能吃 这风口一开，产业链彻底沸腾。天眼查数据显示，2025年新注册的工业互联网企业暴涨了27.8%，大家都想分一杯羹。浪潮云洲的董事长看得很透：大企业 要搞专用大模型，玩的是"高定"；中小企业玩不起高科技？那就推"模型即服务MaaS"，像点外卖一样用AI，门槛直接踩到底。 网络要"神经"，落地得"抠细节" 最近的政策圈有点"忙"，关于工业互联网的利好跟连珠炮似的往外冒。这不是简单的喊口号，而是实打实的"路线图"——到2028年要搞出450多个有头有脸 的工业互联网平台，还要拉着不少于5万家企业搞"新型工业网络"大改造。这一套组合拳下来，目标很明确：让AI给工业互联网装上"大脑"，到2025年核心 产业规模要冲过1.6万亿元大关。这哪是规划？分明是给制造业泼天的富贵！ 告别"连得上"，我们要"想得明白" 以前咱们聊工业互联网，顶多是"连得上、看得见"，像鞍钢那位大佬说的，那是跑分阶段。现在生成式AI和工业大模型成熟了，直接跨越到"认知理 解"和"智能决策"。啥概念？就是以后的工厂不光能传送数据，还能自己琢磨怎么省电、怎么排产。目前这网已经罩住了41个工业大类，2.3万个"5G+工业互 联网" ...

Core Insights - China's manufacturing industry is at a historical intersection of the "14th Five-Year Plan" and a deep restructuring of the global industrial landscape, with smart manufacturing as a key driver for transformation and upgrading [1] - The "14th Five-Year Plan" emphasizes the need for technological upgrades, promoting digital transformation in manufacturing, and developing smart, green, and service-oriented manufacturing [1] Group 1: Smart Manufacturing Trends - The "Smart Manufacturing Technology 50" selection aims to identify leading companies in the smart manufacturing sector, covering areas such as industrial IoT, smart manufacturing, intelligent robotics, and specialized "little giant" enterprises [3] - The evaluation process involves a rigorous assessment system and multiple rounds of expert reviews to select the most innovative and growth-oriented companies [3] Group 2: Industry Dynamics - Over 70% of the selected companies are from the smart manufacturing and intelligent robotics sectors, with nearly half being growth-stage companies established 6-10 years ago [4] - The report highlights a shift in the manufacturing industry towards high efficiency, intelligence, and sustainable development, particularly in the automotive sector, which is experiencing deep integration of electrification and intelligence [4] Group 3: Future Projections - The report identifies six major trends for smart manufacturing development by 2030, including the role of industrial large models as a core engine for intelligent transformation and the emergence of a "human-machine symbiosis" ecosystem [4] - The Chinese industrial robot market remains the largest globally, with collaborative robot shipments exceeding 40,000 units, indicating a shift towards more complex operational environments [5] - Companies that can integrate perception, decision-making, execution, and feedback into a cohesive system will define the competitive landscape of smart manufacturing by 2030 [5]