China’s Key Technology Challenges II – Industrial Software: Bottlenecks and Solutions
By Zhang Ting, Qiu Ping, China Science and Technology Investment Magazine
Introduction: The world is undergoing unprecedented changes, with major power strategic competition further focusing on manufacturing. The United States’ “Advanced Manufacturing Leadership Strategy,” Germany’s “National Industrial Strategy 2030,” and Japan’s “Society 5.0” all center on revitalizing manufacturing, aiming to seize the high ground in the new round of global manufacturing competition. China’s manufacturing value added has maintained the world’s largest share for many years, but it has long been at the mid-to-low end of the value chain. In the transition from a manufacturing giant to a manufacturing power, the role of industrial software is crucial. The EDA/MATLAB supply cut incident has further accelerated the development of domestic industrial software.
Due to its late start, industrial software in China has long been monopolized by foreign countries in high-end core technologies. China’s manufacturing industry accounts for nearly 30% of the global market, but the market share of domestic industrial software is less than 6%. Industrial software is becoming a major bottleneck in China’s transition from a manufacturing giant to a manufacturing power. This article discusses the bottlenecks and solutions of industrial software.
1.Bottlenecks in China’s Industrial Software
Industrial software is the result of the softwareization of industrial knowledge, integrating knowledge from various disciplines such as mathematics, physics, chemistry, electronics, and mechanics, making industrial software an intelligent tool that defines industrial products, controls production equipment, optimizes business processes, and improves operational efficiency. Its core value lies in helping industrial enterprises improve quality, reduce costs, and increase efficiency, thereby enhancing their competitiveness in high-end manufacturing.
In 2021, China’s industrial software products achieved revenues of 241.4 billion yuan, a year-on-year increase of 24.8%. Over the past five years, the global industrial software market has maintained an average growth rate of 6%. As China enters the later stage of industrialization, there is a significant demand gap for industrial software. Industrial software grows alongside advanced manufacturing; Germany, the United States, and the United Kingdom have completed their industrialization processes and, with their deep industrial accumulation, have birthed giant companies worth billions of euros. Overseas industrial software giants emerged during the transformation of industrial powers, continuously enhancing the practical application performance of their products through integration with industries.
Currently, foreign software giants occupy more than 80% of the domestic industrial software market, controlling core technologies in simulation design, analysis tools, enterprise management, and advanced control. There are over 150 commonly used industrial software applications in various fields, covering R&D design, production control, testing, and validation, almost all provided by foreign companies, and the software is closed-source and not open. Based on product forms, uses, and characteristics, industrial software can be divided into four categories: R&D design, production control, information management, and embedded software. The bottlenecks or key issues in each category of industrial software are briefly described as follows:
R&D Design Software: Currently, the biggest shortcoming of China’s industrial software, almost monopolized by foreign vendors. It is mainly used to improve the R&D capabilities and efficiency of enterprises, including CAD/CAE/PLM/EDA, etc. R&D design industrial software is a shortcoming of China’s industrial software, with a low domestic production rate. From the perspective of the number of leading enterprises, among the top ten suppliers in various subfields of R&D design industrial software (CAD/CAE/CAM, etc.), the proportion of domestic enterprises is relatively small. The revenue and R&D investment of domestic enterprises still have a significant gap compared to foreign enterprises. In terms of actual R&D investment, the R&D investment of six leading domestic enterprises is less than 1/9 of that of SAP (the global enterprise software supplier). General-purpose R&D design software is mainly dominated by foreign manufacturers, making it difficult for domestic manufacturers to catch up. R&D design software has interdisciplinary and complex knowledge system engineering characteristics, resulting in significant commercialization challenges and difficulties in ecosystem construction. At the same time, general-purpose software manufacturers are rapidly developing in a platform-oriented manner, making it difficult for domestic manufacturers to catch up. In fields such as CAD, EDA, and BIM, domestic manufacturers have achieved breakthroughs in specific vertical areas.
Production Control Software: The domestic production rate reaches50%, concentrated in mid-to-low-end software. It is mainly used to improve the control level of the manufacturing process and enhance the efficiency and utilization of production equipment, including PLC/EMS/MES, etc. Representative companies include Siemens, Honeywell, Baosight Software, and Zhongkong Technology. The industry has distinct vertical characteristics, with domestic manufacturers being closer to customer needs, achieving a domestic production rate of 50%. Foreign manufacturers have a high market share in high-end discrete industries, while domestic manufacturers mainly focus on mid-to-low-end niche markets, and their scale is relatively small. Domestic manufacturers have initially completed domestic substitution in process industries with monopolistic and mature production techniques. In fields such as DCS, MES, and SCADA, domestic manufacturers have a certain degree of domestic foundation, and the modularity and productization of solutions from various manufacturers are improving, leading to a virtuous cycle.
Information Management Software: Domestic manufacturers dominate the mid-to-low-end market, while foreign manufacturers account for over61% of the high-end market. It is mainly used to improve enterprise management levels and resource utilization efficiency, including ERP/PM/CRM, etc. Representative companies include SAP, Oracle, Yonyou Network, and Kingdee. The industry has entered a stable development period, with the mid-to-low-end market basically achieving domestic substitution, and cloud performance accelerating. Domestic information management industrial software has already seen representative manufacturers emerge, holding a high market share in the mid-to-low-end market, such as Kingdee and Yonyou Network, which have significant scale and strength. Domestic ERP companies have shown strong cloud transformation momentum in terms of financial data and strategic execution capabilities. Domestic manufacturers have launched core cloud-native products with “PaaS + multi-module SaaS” for large, medium, and small enterprises, providing flexible deployment solutions such as localized, hybrid cloud, and public cloud for medium and large enterprises, leading overseas manufacturers in both strategy and implementation.
Embedded Software: A competitive landscape between domestic software and foreign giants. This includes operating systems and development tools embedded in hardware, such as PLC/SCADA/DNC, with representative companies including WNDRVR, Siemens, Haier, and ZTE. It has been widely applied in intelligent transformation, with a high domestic production rate. Embedded software has been extensively applied in the digital transformation of industries. The revenue of the embedded software sector is positively correlated with the proportion of industrial added value and industrial growth rate. A 1% increase in the proportion of embedded software revenue to industrial added value leads to an average increase of 2.3% in per capita industrial added value growth rate, playing a key role in the development of intelligent manufacturing.
2. Solutions for the Development of China’s Industrial Software
China currently faces the challenge of upgrading its manufacturing industry. To upgrade, there must be matching industrial software as a foundation. As a carrier of process accumulation and inheritance, industrial software involves all aspects of the production process and has become the core foundation for forging intelligent manufacturing and operational systems. Industrial software is key to achieving intelligent manufacturing in the era of Industry 4.0. To avoid being “choked” by foreign advanced industrial software and to catch up and surpass it, experts suggest the following approaches:
First, at the national strategic level, high-end industrial software should be placed at the key position of the transition from “Made in China” to “Intelligent Manufacturing in China.” The R&D of industrial software is challenging, with complex system design, high technical barriers, significant hardware costs, a shortage of interdisciplinary R&D talent, and high reliability requirements, leading to long R&D cycles and slow iteration speeds. Generally, the R&D cycle for large industrial software takes 3-5 years, and it takes about 10 years to be recognized by the market, requiring long-term accumulation to achieve goals. Encourage collaboration among software companies, industrial enterprises, universities, and research institutions to avoid technological hollowing, root in the industrial field, focus on industrial data accumulation, and jointly develop core industrial software for the entire product lifecycle and manufacturing process, gradually breaking the situation of being constrained by others in industrial software.
Second, strengthen talent support and cultivate R&D talent. R&D in industrial software requires interdisciplinary talent with both industry knowledge and software development backgrounds. It is essential to fully tap into the research potential of universities and research institutions, benchmark against international leading vocational and technical talent training models, facilitate school-enterprise connections, and train students according to industry needs, while enterprises emphasize the re-education and re-training of talent.
Third, increase funding investment. From the 15th to the 12th Five-Year Plan, the state invested less than 200 million yuan in core industrial software such as CAD/CAE. In contrast, the world’s largest CAE simulation software company, Ansys, invested $298 million in R&D in 2019, approximately 2.1 billion yuan, ten times the investment made by China in 15 years. Industrial software development should be placed on par with industries such as aviation, aerospace, military, and shipbuilding, leveraging the advantages of the national system to increase R&D funding.
Fourth, meet specific needs. Support the development and growth of domestic industrial software companies, helping to expand the application scenarios of innovative industrial software products. Strengthen deep cooperation with internationally leading companies in terms of technology R&D standards, jointly develop and promote high-end products with Chinese industrial characteristics, focus on industrial software network security, and join international joint organizations.
Appendix 1: The EDA/MATLAB Supply Cut Incident
On May 15, 2019, the U.S. Department of Commerce officially issued a ban on Huawei and its subsidiaries in China. The next day, the U.S. EDA software company ANSYS immediately prohibited all business dealings, services, and training with Huawei and its subsidiaries.
ANSYS is a major player in the global engineering simulation software market, playing a crucial role in the creation of many products across various fields, including aerospace, construction, technology, and biomedicine. Through ANSYS’s engineering simulation software, various product designs are simulated and validated to achieve rapid market entry and low-cost objectives. Currently, for Taiwan, a global technology hub, including TSMC, MediaTek, and other component manufacturers, even system manufacturers are customers of ANSYS. Recently, ANSYS also received certification for TSMC’s 5nm FinFET process technology and advanced 3D chip stacking technology for new system integration, making it an indispensable tool in the development of modern technology products.
EDA stands for “Electronic Design Automation,” known as the “mother of chips,” and is the foundational software for all chip designs. If chips are banned, Huawei still has HiSilicon as a substitute, but without EDA software, HiSilicon has no tools to design chips. Globally, 70% of the EDA industry’s revenue is concentrated in three U.S. companies: Cadence, Synopsys, and Mentor Graphics. Although Huawei and its IC subsidiary HiSilicon have remedial plans, there is still a significant gap in support for advanced technologies and processes.
On May 22, 2020, the U.S. Department of Commerce added 33 Chinese entities to the “Entity List,” including Harbin Institute of Technology and Harbin Engineering University, which faced a ban on MATLAB.
MATLAB, short for “Matrix Laboratory,” is known as the “artifact for engineering.” For many engineering students, it is software that must be learned during their academic careers. It is called an “artifact” because of its wide application in engineering fields, covering a vast range of disciplines including mathematical computation, modeling and simulation, electronic communication, mechanical engineering, automotive aviation, power energy, economic finance, and biomedicine.
MATLAB itself has become part of the “qualification certification” process, and many submitted papers must include MATLAB program validation. If MATLAB is not allowed, many researchers will face significant challenges, which highlights the power of standards.
Whether it is MATLAB or EDA, both point to the issue of the lack of core industrial software in China.