Core Insights - The strength of domestic large models has increased, showcasing competitive barriers in "local capability SOTA" and cost-effective computing power utilization [3] - The rapid iteration of models enhances reasoning capabilities, significantly benefiting upstream infrastructure, with cloud services and core computing components being the main beneficiaries [3] - The industrial software sector possesses a deep moat, as it requires a collaborative computing network across cloud, edge, and terminal, with higher security demands [3] - The synergy between computing power and electricity policies is driving the growth of intelligent power scheduling and trading services, while also promoting the export of tokens and energy [3] - Investment recommendations focus on midstream domestic large models, upstream cloud service providers, and downstream industrial AI solution providers [3] Section Summaries 01 Strengthening of Domestic Large Models - Domestic large models have entered a dual-driven phase of "agent engineering" and "native multimodal" [12] - Major companies like Alibaba and ByteDance have released flagship models that enhance agent capabilities, indicating a shift from mere conversational tools to execution agents [14] - Domestic large models have shown significant improvements in long context, multimodal, and collaborative capabilities, with many models achieving context window lengths of 256K and supporting native multimodal understanding [19] 02 Explosion of Token Demand - The daily average token usage in China is projected to increase from 100 billion in early 2024 to 100 trillion by the end of 2025, with a significant surge to 140 trillion in March 2026 [26] - The transition to Agentic AI has led to a paradigm shift in token consumption, with active agents expected to grow from approximately 28.6 million in 2025 to 2.216 billion by 2030, reflecting a compound annual growth rate of 139% [26] - The demand for reasoning computing power is structurally exploding, with significant capital expenditures from overseas firms projected to continue rising [27] 03 Industrial AI and Computing Power Synergy - The integration of computing power and electricity is expected to optimize energy consumption and stabilize power systems, with a strong growth outlook under the "East Data West Computing" initiative [3] - The domestic computing power landscape is shifting towards increased localization, with significant percentages of domestic chip usage in various AI computing centers [34] - The demand for optical communication components is expected to rise sharply, driven by the need for high-speed interconnects in AI computing clusters [50]

Investment Rating - The report rates the computer industry as "Positive" [1] Core Insights - Domestic large model capabilities are strengthening, with Chinese manufacturers demonstrating competitive barriers in "local capability SOTA" and cost-effective computing power utilization [3] - The rapid iteration of large models enhances reasoning capabilities, significantly benefiting upstream infrastructure, with cloud services and core computing components being the main beneficiaries [3] - The industrial software sector possesses a deep moat, as it requires a collaborative computing network across cloud, edge, and terminal, with higher security demands that general large models cannot meet [3] - The synergy between computing and electricity policies is accelerating the growth of intelligent power scheduling and trading services, while also promoting the export of tokens and energy [3] - Investment recommendations focus on midstream domestic large models, upstream cloud service providers, and hardware companies capable of domestic substitution, as well as downstream industrial AI solution providers [3] Summary by Sections 01 Strengthening of Domestic Large Models - Domestic large models have entered a dual-driven phase of "intelligent agent engineering" and "native multimodal" [12] - Major companies like Alibaba and ByteDance have released flagship models that enhance agent capabilities, marking a shift from mere conversational tools to execution agents [14] - Domestic large models have shown significant improvements in long context, multimodal, and collaborative capabilities [19] 02 Explosion of Token Demand - The daily average token usage in China is projected to increase from 100 billion in early 2024 to 100 trillion by the end of 2025, with a current surge to 140 trillion [26] - The transition to Agentic AI has led to a structural explosion in reasoning power demand, with active agents expected to grow from approximately 28.6 million in 2025 to 2.216 billion by 2030 [26][36] - The demand for reasoning power is expected to drive significant increases in cloud computing prices, with major cloud providers already implementing price hikes [51] 03 Industrial AI and Computing-Electricity Synergy - The integration of computing and electricity is expected to optimize energy consumption and stabilize power systems, with a focus on achieving carbon neutrality [3] - The report highlights the importance of industrial AI solutions that can provide intelligent scheduling and trading services in the context of computing-electricity synergy [3]

Core Viewpoint - The collaboration between industrial giants, cloud computing platforms, and robotics signifies a shift of AI from the internet realm to the industrial era, marking the beginning of a new phase in AI development focused on physical world applications rather than just virtual ones [1][3][16]. Group 1: Collaboration and Integration - The partnership between Siemens, Alibaba Cloud, and Yuzhu Technology represents a significant step towards achieving a "closed-loop" system in industrial AI, integrating software, cloud computing, and robotics [5][6]. - Siemens provides industrial software and automation systems, serving as the "operating system" for manufacturing, while Alibaba Cloud offers scalable computing power and infrastructure, addressing concerns about data security and deployment costs [6][9]. - Yuzhu Technology introduces humanoid robots that enhance the execution layer of AI, enabling adaptability in unstructured environments and solving the challenge of transitioning from simulation to real-world application [7][9]. Group 2: Advantages of the Chinese Market - China is identified as the optimal environment for the implementation of industrial AI due to its comprehensive manufacturing ecosystem and the willingness of local factories to adopt new technologies for efficiency gains [9][10]. - Local cloud providers like Alibaba have developed mature capabilities in data processing and service responsiveness, which are crucial for addressing the specific needs of Chinese enterprises [10][11]. - The rapid advancement in robotics within China, exemplified by Yuzhu Technology, is narrowing the gap with international competitors, making the commercialization of "robot + AI" more feasible [10][11]. Group 3: Investment Implications - The collaboration indicates a paradigm shift in the industrial landscape, where the value chain is transitioning from traditional manufacturing to a focus on data, models, and execution capabilities [13][14]. - The importance of ecosystem collaboration is emphasized, as no single company can cover the entire industrial AI chain, necessitating a focus on connectivity and integration among cloud, models, and hardware [13][14]. - The efficiency gains from AI in manufacturing are expected to be exponential, with robots capable of managing multiple processes and continuously learning, which will significantly impact profit margins [15][16].

Market Overview - In February, A-shares and Hong Kong stocks showed a divergence, with small-cap stocks outperforming large-cap growth and Hong Kong tech stocks experiencing significant adjustments[4] - The geopolitical risks, particularly in the Middle East, have led to a rise in international gold and oil prices, impacting the cyclical sectors in A-shares and Hong Kong stocks[4] Investment Strategy - The report suggests focusing on strategic resources and cyclical recovery sectors, particularly industrial metals like copper, precious metals, and energy metals due to supply constraints and geopolitical tensions[4] - Emphasis on technology self-reliance and new productivity sectors, particularly AI computing and military industries, is recommended as the market anticipates policy support[4] Key Stock Recommendations - Zijin Mining (601899.SH) is projected to have an EPS of 3.37 yuan in 2026, with a PE ratio decreasing from 32.66 in 2024 to 11.70 in 2026, indicating strong growth potential[6] - New Fengming (603225.SH) is expected to benefit from seasonal demand, with an EPS forecast of 0.99 yuan in 2026 and a PE ratio of 21.44[21] - Baosteel (600019.SH) is highlighted for its significant market share in high-end products, with an EPS of 0.55 yuan in 2026 and a PE ratio of 13.29[30] Financial Projections - The projected revenue for Zijin Mining is expected to grow from 303.64 billion yuan in 2024 to 423.24 billion yuan in 2026, with a net profit increase from 32.05 billion yuan to 89.51 billion yuan during the same period[19] - New Fengming's revenue is projected to increase from 67.09 billion yuan in 2024 to 73.29 billion yuan in 2026, with a net profit growth from 11 billion yuan to 15.16 billion yuan[28] Risk Factors - Potential risks include unexpected policy changes, underperformance in commercialization, and geopolitical uncertainties affecting market stability[4]

工业软件行业丨研究报告 摘要： 工业软件具有发展的紧迫性和必要性，且当前处于政策红利带的有利时间窗口期。 当前，我国工业和经济达到分水岭，经济体发展需要创新驱动，而工业软件作为工业知 识 的 载体，既是新型工 业化的核心生产资料和关键生产力，又是工业大脑和数字基 石，其 自主可控意义深远。不同于 国外工业软件是 先工业后软件的自然生长 ，我国的工业 软件先是用市场换 效 率，后是工业 和软件同步的压缩 式 发展，现在是追赶核 心技术可控，保障供应链安全。因此，当前工业软件既有发展的必要性，又有发展的紧迫性。 工业软件是一个慢行业，发展需要耐心和长期主义思想，同时在变化与重构中，也为企业带来机遇与挑战。 中国的工业软件市场是千亿的大盘子， 2024 年市场接近 3000 亿，市场增长稳健， 但 核心技 术空心、产业结构失衡等 问题凸显。当前，研发设计类工业软件是卡脖子最为 严重，其本质 是与数学 与基础 学科相关的根技术缺乏海量真实工业场景试错进行工程优化，表现为实体就是核心组件 / 引擎层受限。值得注意的是，根技术的突破没有捷 径可走，只能死磕 。 工业软件产业处于动态发展中，未来产业、市场、产品的发展方向值 ...

Core Viewpoint - The industrial software industry is at a critical juncture, necessitating innovation-driven development, with significant policy support and market opportunities ahead [1][4][12]. Group 1: Industry Background - Industrial software is essential for innovation and transformation in the economy, serving as a core production material and a digital foundation for new industrialization [4][7]. - Since 2018, China's per capita GDP has reached $10,000, marking a shift towards a more developed economy that requires innovation for further growth [4]. - The transition of control from hardware to software is evident, with industrial knowledge increasingly encapsulated in software code [4][9]. Group 2: Market Dynamics - The Chinese industrial software market is projected to approach 300 billion yuan in 2024, indicating robust growth despite challenges such as core technology gaps and structural imbalances [1][17]. - The market is characterized by a significant demand for domestic alternatives, particularly from leading clients, while mid-tier clients can form communities and long-tail clients can expand revenue opportunities [2][14]. Group 3: Product Evolution - Current industrial software primarily involves purchasing products, but the future will shift towards selling "intelligence," leading to the emergence of industrial intelligent agents [3][48]. - The evolution of industrial software is moving from tools to systems, platforms, and eventually to a genetic level, focusing on data value and efficiency [48][52]. Group 4: Driving Factors - Policy support and technological advancements, particularly in AI and large models, are accelerating the development and application of industrial software [12][19]. - The demand from enterprises emphasizes practical utility and domestic alternatives, while government initiatives focus on top-level planning and industry promotion [14][50]. Group 5: Challenges and Opportunities - The industry faces systemic issues related to technology, scenarios, ecosystems, and business models, with a significant focus on overcoming core technology deficiencies [26][39]. - The market is experiencing a shift towards platformization and the integration of data services, which presents opportunities for companies that can effectively leverage data [30][52]. Group 6: Future Directions - The industrial software sector is expected to evolve into a phase where products are not just tools but intelligent systems capable of autonomous task execution [52][54]. - Companies are encouraged to target head clients for technological breakthroughs while also engaging mid-tier and long-tail clients to enhance cash flow and industry integration [50][52].

Core Insights - The article emphasizes the significance of industrial software as the "brain and nerve" of the industrial system, facilitating the digitalization of industrial knowledge, processes, and workflows across various stages including R&D, manufacturing, and management [1][48]. Group 1: Industrial Software Landscape - Industrial software encompasses various types such as CAD, CAE, EDA for product design, ERP for business management, PLM for product lifecycle management, and MES for production execution, forming the core digital foundation of manufacturing [5][48]. - The industrial software industry is structured in three layers: foundational support, core R&D, and industry applications, with strong policy support and the integration of AI technologies driving growth [6][10]. Group 2: Market Opportunities and Growth - The Chinese industrial software market is projected to grow from CNY 31.86 billion in 2024 to CNY 76.5 billion by 2029, with a compound annual growth rate (CAGR) of approximately 19.1% [6][24]. - The integration of AI into industrial software is expected to see a CAGR of 41.4% from 2024 to 2029, significantly outpacing the overall industrial software growth rate [9][50]. Group 3: Investment Value of the National Industrial Software Index - The National Industrial Software Theme Index (980034.CNI) was launched on February 17, 2015, and includes 50 stocks that reflect the overall performance of the Chinese industrial software industry [2][29]. - The top ten stocks in the index account for 56.37% of its weight, with a notable increase in R&D investment from 4.31% in 2022 to 9.22% by the third quarter of 2025 [2][30]. Group 4: Profitability and Growth Potential - The index is expected to see a net profit growth rate of 59.22% in 2025 and 30.12% in 2026, with earnings per share growth rates of 59.93% and 59.32% respectively, indicating a positive outlook for profitability and growth [3][37]. - The index includes 34 specialized and innovative companies, representing 68% of its constituents, which is significantly higher than mainstream broad-based indices [40][50]. Group 5: Long-term Risk-Return Profile - The index has shown an annualized return of 2.52% from 2017 to February 2026, with a Sharpe ratio of 0.23, outperforming other indices such as the CSI Computer and CSI Software Service indices [45][50]. - The industrial software sector is entering a golden development period, driven by accelerated domestic substitution, AI integration, and robust policy support [50].

Quantitative Models and Construction Methods - **Model Name**: CNI Industrial Software Theme Index **Model Construction Idea**: The index aims to comprehensively reflect the overall performance of China's industrial software industry by selecting 50 stocks from companies involved in industrial R&D design software, industrial production information software, industrial enterprise business management software, and industrial automation software[2][32][33] **Model Construction Process**: 1. **Sample Space Definition**: Includes A-shares and depositary receipts of red-chip enterprises meeting the following conditions: - Non-ST or *ST securities - Listed for over 1 year on the STAR Market or Beijing Stock Exchange, and over 6 months for other securities - No major violations or financial issues in the past year - No abnormal operations or significant losses in the past year - No abnormal price fluctuations during the observation period - Business involvement in industrial software-related fields such as R&D design, production information, business management, and automation software[33] 2. **Selection Method**: - Calculate the average daily trading volume and market capitalization over the past six months for eligible securities - Exclude the bottom 10% of securities by average daily trading volume - Rank the remaining securities by average daily market capitalization and select the top 50 as index components. If fewer than 50 securities meet the criteria, all eligible securities are included[33] 3. **Weighting Method**: The index uses the Paasche weighting method and adjusts weights to ensure that the weight of industrial R&D design software is not less than 50%[33] 4. **Periodic Adjustments**: The index is adjusted semi-annually on the trading day following the second Friday of June and December[33] - **Model Evaluation**: The index is highly focused on the industrial software sector, with a strong emphasis on innovation and small-cap characteristics, making it a valuable tool for capturing growth in this industry[32][34][39] --- Model Backtesting Results - **CNI Industrial Software Theme Index**: - Annualized Return: 2.52% - Annualized Sharpe Ratio (IR): 0.23 - Annualized Volatility: 30.01% - Maximum Drawdown: -61.56%[56] --- Quantitative Factors and Construction Methods - **Factor Name**: R&D Intensity Factor **Factor Construction Idea**: Measures the innovation capability of index components by calculating the proportion of R&D expenses to operating revenue[36] **Factor Construction Process**: - Formula: $ R&D\ Intensity = \frac{R&D\ Expenses}{Operating\ Revenue} $ - Data: - 2022: 4.31% - 2023: 8.64% - 2024: 8.71% - 2025 (Q3): 9.22%[36][37] - **Factor Evaluation**: The continuous increase in R&D intensity highlights the sector's strong focus on innovation and technological advancement[36][37] --- Factor Backtesting Results - **R&D Intensity Factor**: - 2022: 4.31% - 2023: 8.64% - 2024: 8.71% - 2025 (Q3): 9.22%[36][37] --- Additional Observations - The index exhibits a strong small-cap bias, with an average market capitalization of 276.24 billion RMB, and over 70% of its components having a market cap below 200 billion RMB[45][46] - The index is highly concentrated in the computer industry (62%), with significant weights in sub-sectors like software (38.63%), cloud services (15.69%), and electrical equipment (15.51%)[39][42] - The index includes 34 "specialized, refined, distinctive, and innovative" (SRDI) companies, accounting for 68% of its components, significantly higher than other mainstream indices[49][53] - Projected growth: - 2025E Net Profit Growth: 59.22% - 2026E Net Profit Growth: 30.12% - 2025E EPS Growth: 59.93% - 2026E EPS Growth: 59.32%[47][51]

Core Viewpoint - Aerospace Software (688562.SH) reported a total operating revenue of 1.216 billion yuan for the fiscal year 2025, representing a year-on-year increase of 4.60% [1] - The company recorded a net profit attributable to shareholders of the parent company at -11.3126 million yuan, and a net profit excluding non-recurring gains and losses at -70.4733 million yuan [1] Group 1 - During the reporting period, the company focused on market expansion, adjusted its market layout, enhanced sales capabilities, and improved project delivery quality, leading to revenue growth compared to the previous year [1] - The company adjusted its business structure to increase the proportion of revenue from software and information technology services, which have relatively higher gross margins [1] - The company intensified reform efforts to control costs, optimize processes, and strictly manage expense expenditures [1] Group 2 - Despite the revenue growth, the industry remains highly competitive, and the investment in industrial software is substantial with long cycles [1] - To enhance core competitiveness, the company increased its investment in core product research and development while also expanding into new technology areas such as AI and software factories [1] - Research and sales expenses increased year-on-year as the company continued to invest in R&D and sales efforts [1]

Core Viewpoint - Aerospace Software (688562.SH) reported a total operating revenue of 1.216 billion yuan for the fiscal year 2025, representing a year-on-year increase of 4.60% [1] Financial Performance - The net profit attributable to the parent company was -11.3126 million yuan [1] - The net profit attributable to the parent company, excluding non-recurring gains and losses, was -70.4733 million yuan [1] Business Strategy - The company focused on market expansion, adjusting its market layout, enhancing sales force, and improving project delivery quality, leading to revenue growth compared to the previous year [1] - The company adjusted its business structure to increase the proportion of revenue from higher-margin software and IT service businesses [1] - The company intensified reform efforts to control costs, optimize processes, and strictly manage expense expenditures [1] R&D and Investment - Due to intense market competition and the long investment cycle of industrial software, the company increased R&D investment in core products to enhance core competitiveness [1] - The company is also expanding into new technology areas such as AI and software factories, with continued increases in R&D and sales expenses year-on-year [1]