Israeli author Yuval Noah Harari wrote in his world‑renowned book *Sapiens*: The Industrial Revolution is a permanent revolution.
Indeed, industry matters immensely.
Today, China’s technology sector is rising. Emerging information technologies keep innovating, traditional industries are reshaped by digital tools, and post‑pandemic industrial policies are growing in influence. Closely tied to people’s livelihoods, employment, corporate production and national strength, industry has entered a new era — the era of software‑defined new‑industry.
What is the “new‑industry”? It represents an industrial form less subject to individual human interference, an industrial model where technologies deliver maximum automation value, and a new phase where factories become increasingly data‑driven.
In this phase, technology service providers such as Sinovation Tech, Jiyuan Tech, Yitu Technology and Black Lake Manufacturing have emerged. They boost the operational efficiency of machinery inside factories and push industrial manufacturing forward.
New Technologies Empower Factories
Software has long been the soul of new‑generation information technologies. Amid the integration of cloud computing, big data, artificial intelligence, IoT and digital twins, software is defining everything that can be defined.
Within modern industry, many enterprises deploy partial software products to empower equipment, yet operations still largely rely on manual labor and human cognition. Consequently, factory operational efficiency is heavily constrained by skilled workers, and extensive management remains commonplace.
There exists no compression algorithm for real‑world operational experience, and human‑driven models inevitably face limitations.
Today, industrial software is the product of software‑enabled industrial technologies and represents the pinnacle of industrialization. Why? It encapsulates industrial expertise, establishes rules for automatic data flow, and builds the “brain and nervous system” for machinery. As a result, machines grow smarter and their functions can be adjusted on demand.
Industrial software describes, integrates, simulates, accelerates, amplifies, optimizes and innovates traditional manufacturing workflows, giving rise to new forms of industrial intelligence. In 1964, General Motors and IBM jointly developed the DAC‑I system for linear design of automotive windshields, regarded as the world’s first genuine industrial software. Starting in the 1970s, CAD technology entered most engineering and product‑design domains. By the end of the 20th century, widespread computer adoption, together with industrial software, chips and the internet, rapidly advanced manufacturing mechanization, electrification, speed‑up, precision and automation. A modern high‑tech, highly‑informative industrial system took shape, with industrial software becoming a core source of industrial competitiveness.
The cloud revolution led by cloud vendors decouples industrial computing, networking and storage from heavy‑duty hardware, shifting toward lightweight cloud‑native architectures and enabling smarter mechanical operations. Artificial intelligence has achieved distinctive progress in visual inspection, video analytics and natural language processing. The Internet of Things performs impressively in collecting industrial data, interconnecting diverse terminals, effectively accumulating production data and unlocking its value, opening new possibilities for the Industrial Internet.
China’s Industrial Internet has entered a period of rapid growth, registering an annual growth rate of 11.28 %, ranking first among major industrial economies worldwide. China accounts for 20.19 % of the total Industrial‑Internet market across the world’s top‑ten industrial nations, exceeding the combined share of Japan and Germany. The Industrial Internet delivers strong driving forces for secondary‑industry development. It will comprehensively reshape China’s manufacturing DNA and realize smart manufacturing featuring both high quality and low costs.
Once the tide of digitalization arrives, everything changes.
Software Is Defining Everything That Can Be Defined
China’s manufacturing sector boasts the world’s most complete range of industrial categories, covering more than 220 types.
Chen Long, Chief Marketing Officer of Jiyuan Tech, a service provider focusing on energy‑power and industrial automation, notes that China has an enormous manufacturing base, yet vertical‑sector scenarios and requirements vary widely. Enterprises operate at disparate maturity levels, and many business owners lack urgent digital‑transformation demands. Hence, no single solution can satisfy all corporate requirements.
Industrial‑Internet enterprises represented by Jiyuan Tech keep exploring how to identify valuable common requirements amid complexity, facilitate manufacturing digital transformation to improve quality and efficiency, and jointly cultivate the market.
Shen Xiaoyong, Co‑founder and CEO of Sinovation Tech, an industrial‑AI‑focused service provider, observes that AI deployment still faces major pain points: difficulties in quantitative standardization, frequent scenario changes, divergent requirements, plus barriers where enterprises cannot afford, dare not use or simply do not know how to adopt AI technologies. These pain points, nevertheless, are regarded as core opportunities and driving forces for the future industrial‑AI and smart‑manufacturing markets.
Sinovation Tech’s ongoing efforts to build an industrial‑AI hardware‑software ecosystem, industrial‑AI operating system and integrated hardware‑software products serve as an ideal entry point.
Today, “technology implementation” has become a guiding principle for tech companies of all sizes. Whether it refers to downward‑penetration strategies rolled out by established internet giants, case studies from AI vendors, or numerous vertical‑sector niche champions focusing on specialized tracks, the integration of technology and real‑world scenarios is on a steep upward trajectory.
Compared with retail, education, healthcare, finance and other sectors, industry features the highest complexity, most fine‑grained sub‑sectors and greatest significance to the national economy. How to achieve efficient, sustainable and green development for China’s manufacturing industry stands as a shared topic for discussion today.
Classic solutions frequently emerge within industry, such as MES (Manufacturing Execution System), offered by many vendors. It manages manpower, machinery, materials, methods, environment, measurement and control throughout production workflows, enabling error‑proofing before manufacturing, in‑process control, and full‑lifecycle product‑quality tracking and traceability.
Nevertheless, such “old‑industry” approaches can no longer satisfy today’s industrial‑development demands. Managing industrial clouds, optimizing process workflows, analyzing industrial big data, enabling intelligent inspection and decision‑making, and realizing smart supply‑chain management represent the true technical essence of the “new‑industry”.
The popularization of smartphones has driven wireless‑charging adoption, with wireless‑charging coils serving as key components. Generally speaking, each component of a wireless‑charging coil is tiny and intricate, featuring highly diverse defect patterns. Manual inspection is prone to misjudgment and missed defects.
During interviews, we learned that Sinovation Tech built exactly such an automated AI visual‑inspection device for wireless‑charging coils for a global top‑tier mobile‑phone brand. Powered by Sinovation’s self‑developed integrated industrial‑AI platform SMore ViMo, deep‑learning‑based classification and segmentation algorithm models are trained to judge OK/NG status of materials. It detects 14 types of minor defects and sorts materials into six categories, comprehensively boosting production‑line inspection efficiency (reaching 1300‑1400 pieces per hour, far exceeding manual inspection’s 800 pieces, lifting inspection speed by 50 %).
This serves as a highly representative case.
In another practical reference case, Jiyuan Tech deployed its technologies within cement‑production scenarios. By collecting and analyzing data from raw‑material mills, coal mills, cement mills and rotary kilns for a cement‑plant customer, scenario‑oriented software systems were built to aggregate and analyze key production‑process data. Combining AI algorithms with production experts’ domain knowledge, optimal scheduling‑and‑production‑planning solutions were derived, and start‑stop control strategies for multiple production devices were output to guide on‑site operations. Without compromising output, the cement plant achieved daily electricity‑cost savings of more than RMB 5,000.
Such perspectives and performance improvements would be inconceivable under traditional industrial‑process optimization approaches.
“Data‑Centric” Becomes Industry Consensus
Driven by smart‑manufacturing trends, factory operations place growing emphasis on data value. Analysis and mining of digital‑workshop production data equip industrial enterprises with enhanced capabilities for quality improvement, cost reduction and efficiency gains. Software systems combining AI algorithms to realize closed‑loop automatic backward‑control optimization greatly reduce factories’ dependence on specialized job roles. Highly customized software systems can redefine and reconstruct manufacturing workflows, substantially lifting production efficiency and product‑quality stability.
Chen Long, CMO of Jiyuan Tech, holds that software will redefine traditional industry and empower industry in the new era. All of this is built upon data foundations.
“We are exploring how AI‑algorithm application architectures can achieve great outcomes with limited datasets.” Heated industry debates once circled around “why AI delivers unsatisfactory performance”. Many attributed it to insufficient datasets and low neural‑network efficiency.
When applied to industrial‑AI scenarios, Shen Xiaoyong from Sinovation Tech points out that data scarcity remains a major challenge for AI adoption in manufacturing. Many manufacturers and factories suffer from insufficient data volume. Does limited raw‑material‑like data rule out AI industrialization? Not necessarily.
Sinovation Tech has proposed a solution: store all intermediate‑result data to generate large‑scale datasets for continuous AI learning. Factories can then feed in small‑volume data to generate well‑adapted algorithms. More input data enables the model to better learn problem‑solving patterns. Later, machines only need key feature values to deliver rapid judgments.
Though the “data‑centric” slogan has circulated for years, few legacy factories genuinely break down data silos and unlock value from flowing data. This is not caused by technical barriers. The root cause lies in enterprise managers’ lack of strategic awareness for building unified information platforms, plus lingering doubts over whether legacy systems can be technically interconnected.
This requires gradual digital‑mindset education, as well as hardware‑software solutions to build technological trust.
Better Support for China’s Industrial Development
Consensus has been reached and technologies keep advancing. China’s manufacturing industry is gradually overcoming traditional pain points concerning quality, efficiency, cost and inventory. Meanwhile, the dual‑carbon goals and stricter dual‑control over energy consumption bring new tests for manufacturing’s technical supply capacity.
Given diverse business portfolios, highly variable scenarios, scarce common management logic and lack of unified standards, building collaborative ecosystems is especially critical within industrial smart‑manufacturing.
In the past, ZhenFund has supported many outstanding tech enterprises. With years‑long insights and investments in artificial intelligence and industrial manufacturing, it is ready to explore future smart‑manufacturing possibilities alongside technology companies.
Driven by emerging software‑related opportunities, manufacturing DNA is being reshaped, and demand for digitally‑driven new‑industrial‑infrastructure keeps rising.
Industry is being redefined.
