CtrlK
SUPCON Technology Co., Ltd. (Chinese: 中控技术; pinyin: Zhōngkòng Jìshù), formally Zhejiang Supcon Technology Co., Ltd., is a Chinese industrial automation and industrial artificial intelligence company headquartered in Hangzhou, Zhejiang Province. Founded in 1993 as a spin-out from Zhejiang University, the company has grown into the dominant supplier of Distributed Control Systems (DCS) in China and one of the country's most prominent vendors of industrial AI software for the process industries. SUPCON is publicly listed on the Science and Technology Innovation Board (STAR Market) of the Shanghai Stock Exchange under stock code 688777.
Although originally a hardware-focused control systems vendor, SUPCON has repositioned itself in the 2020s as an "industrial AI platform-based" enterprise. Its product strategy now centers on three layers: a plant operating system called supOS, a cloud-native Universal Control System called Nyx, and a family of industrial large language and time-series models branded TPT (Time-series Pre-trained Transformer). These products target process industries such as petrochemicals, refining, chemicals, electric power, pulp and paper, mining, and food and beverage. The company has also entered embodied AI through the Zhejiang Humanoid Robot Innovation Center, where it launched a humanoid robot called Navigator alpha aimed at high-risk industrial tasks.
With more than 35,000 customers across over 50 countries and a 2024 share of roughly 40 percent of China's DCS market (and over 60 percent in chemicals), SUPCON occupies a position in industrial AI and process automation that is sometimes compared to that of Siemens, Honeywell, Emerson, ABB, and Yokogawa, while also reflecting the broader rise of China AI infrastructure.
SUPCON Technology operates as the listed flagship of the broader SUPCON Group. The group's roots trace to the State Key Laboratory of Industrial Control Technology at Zhejiang University, and the listed entity focuses specifically on industrial software, control systems, instrumentation, and AI for process manufacturing. The company describes its current architecture as "1 Factory Operating System + Mechanism and AI Dual-Engine Drive + N Business Scenario Intelligence," which is shorthand for combining first-principles process models with data-driven AI on a single industrial software stack.
SUPCON has consistently been ranked the largest DCS vendor in China for more than a decade. Its 2024 share of the domestic DCS market was approximately 40.4 percent, with even higher concentration in chemicals (about 63 percent) and petrochemicals (about 56 percent). It now positions itself less as a control systems vendor and more as a provider of "autonomous operating plants," a vision in which large fractions of routine plant operation are handled by AI agents under human supervision.
The company is led by founder and chairman Chu Jian (褚健), a former vice president of Zhejiang University and director of the State Key Laboratory of Industrial Control Technology. Chu earned his PhD in automation from Zhejiang University in 1989 through a China and Japan joint training program and is widely regarded as one of the architects of China's domestic process control industry.
SUPCON's history begins in 1993, when Chu Jian and a team of researchers from the State Key Laboratory of Industrial Control Technology at Zhejiang University founded the original Supcon entity to commercialize laboratory research on distributed control. At the time, China's process industries were almost entirely dependent on imported DCS hardware from foreign vendors such as Honeywell, Foxboro, Yokogawa, and Siemens. The early team set out to design a domestically engineered DCS that could compete on reliability while undercutting imports on price and service responsiveness.
The original company operated initially as part of Zhejiang Highne Science and Technology Company, a vehicle owned by Zhejiang University to commercialize academic research. In 1999, after Zhejiang Highne was listed on a Chinese exchange, Supcon was identified as one of three core sectors and spun out to operate independently. This spin-out is generally treated as the founding moment of SUPCON Technology Co., Ltd. as a standalone enterprise.
Through the 2000s and early 2010s, SUPCON grew rapidly by displacing foreign DCS vendors in mid-tier Chinese process plants, then progressively moving into larger and more complex installations. Its product line expanded from the original ECS series of DCS controllers into safety instrumented systems, advanced process control software, real-time databases, instrumentation, and engineering services. During this period, SUPCON also began building a software portfolio above the control layer, including manufacturing execution systems, plant-wide data historians, and energy management products. These offerings would later become the foundation for the supOS platform.
On November 24, 2020, Zhejiang Supcon Technology Co., Ltd. listed on the Science and Technology Innovation Board of the Shanghai Stock Exchange under stock code 688777. The IPO raised approximately USD 268 million. At the time of listing, founder Chu Jian held a personal stake reported to be worth approximately USD 1.6 billion. The listing gave SUPCON capital to accelerate investment in cloud-native control architectures and industrial AI.
From 2021 onward, SUPCON has progressively repositioned itself as an industrial AI company. The pivot included three highly visible product launches: the supOS plant operating system, the Nyx Universal Control System, and the TPT industrial large model family. The company also moved aggressively into international markets, exhibiting at Hannover Messe and ADIPEC in Abu Dhabi and opening or expanding subsidiaries in Singapore, Saudi Arabia, the Netherlands, Japan, India, Malaysia, and Indonesia. Alongside this software pivot, SUPCON entered embodied AI in 2024 by anchoring the Zhejiang Humanoid Robot Innovation Center, which subsequently launched the Navigator alpha humanoid robot.
SUPCON's product portfolio spans four broad layers: instrumentation and field devices, control systems (DCS, SIS, and the Nyx UCS), industrial software (including supOS), and industrial AI models and agents (including TPT). The table below summarizes the most prominent AI-relevant products.
| Product | Category | Description |
|---|---|---|
| Nyx | Universal Control System (UCS) | Cloud-native, software-defined replacement for traditional DCS, with AI-driven control loops, IEC 61131-3 plus C++ and Python programming, and Ethernet-APL field networking. |
| NyxOS | Real-time operating system | Cloud-native real-time OS underpinning Nyx, supporting millisecond-level deterministic scheduling and "control-as-a-service" deployment. |
| supOS | Plant operating system | Industrial OS that unifies plant data, runs industrial apps, and supports SaaS, on-premise, and edge deployments. supOS-CE is an open-source community edition. |
| TPT | Industrial time-series large model | Pre-trained transformer for industrial time-series data, used for forecasting, simulation, and optimization across process plants. |
| TPT2 | Second-generation industrial AI model | Released August 2025, uses a Mixture of Experts (MoE) architecture and was pre-trained on more than 10 trillion industrial data points. |
| AI-PID | Adaptive control loop | Reinforcement learning based PID controller tuning for industrial loops, integrated into Nyx. |
| AI Code Assistant | Engineering productivity tool | Generates Structured Text (ST) programs from natural language descriptions, embedded in Nyx engineering tooling. |
| Industrial intelligent agents | Application layer | Domain-specific AI agents for safety, quality, energy, and equipment maintenance scenarios, built on TPT and supOS. |
| Navigator alpha | Humanoid robot | 1.5 meter, 50 kg humanoid platform with 15-finger-joint hand, designed for high-risk industrial tasks and intended for integration with large AI models. |
Launched in 2024, Nyx is positioned by SUPCON as the world's first Universal Control System (UCS). It is described as a software-defined, fully digitalized, cloud-native architecture intended to replace traditional DCS structures that have remained largely unchanged for roughly 50 years. Instead of relying on dedicated controllers and physical I/O modules in cabinets, Nyx centralizes control logic in a control data center and connects to field devices over an Ethernet-APL based all-digital network, with each instrument addressable by IP.
Key design points include a cloud-native real-time operating system (NyxOS) supporting deterministic scheduling at millisecond resolution; multi-language programming, with IEC 61131-3 control languages alongside C++ and Python; AI-PID controllers that use reinforcement learning to self-tune parameters online; an AI code assistant that generates Structured Text programs from natural language prompts; and tight integration with GPU-equipped control engines so that AI inference workloads can run close to the control loop.
SUPCON has reported that Nyx deployments can reduce on-site footprint by around 90 percent, copper cabling by around 80 percent, and project lead time by around 50 percent compared with conventional DCS architectures. International coverage of Nyx in Control Engineering and Drives and Controls has framed it as a serious challenger to the long-dominant DCS paradigm pioneered by vendors like Honeywell and Emerson.
supOS is SUPCON's industrial operating system. It serves as a unified data and application layer for industrial plants, integrating data from DCS, SCADA, IIoT devices, MES, and ERP systems, then exposing that data to industrial applications. SUPCON describes supOS as the first industrial operating system in China to be built on proprietary domestic technology.
supOS provides a configuration-driven low-code development environment for building plant-specific apps; a library of industrial applications spanning intelligent inspection, process dashboards, production management, predictive maintenance, energy optimization, and quality analytics; flexible deployment options across local, edge, and cloud environments; and an open community edition, supOS-CE, distributed free of charge under an open-source license via GitHub.
The supOS layer is also the runtime in which SUPCON's industrial AI agents are deployed. Agents that use TPT models for forecasting or optimization are typically delivered as supOS apps that read data from supOS and send setpoints back to the underlying control system, whether legacy DCS or Nyx.
TPT, short for Time-series Pre-trained Transformer, is SUPCON's family of industrial time-series foundation models. Unlike general-purpose large language models, TPT is trained primarily on industrial sensor data, including pressure, temperature, flow, composition, and equipment vibration signals, with the goal of capturing the dynamic behavior of process units.
TPT is designed to forecast key process variables across short and medium time horizons, simulate the response of process units to operator actions, power optimization agents that recommend or execute setpoint changes, and detect anomalies in equipment behavior to support predictive maintenance.
The second generation, TPT2, was released globally on August 28, 2025 at the SUPCON Global New Product Launch and Industrial AI Innovation Development Conference in Hangzhou. According to SUPCON, TPT2 was pre-trained on more than 10 trillion industrial data points and uses a Mixture of Experts (MoE) architecture so that different expert sub-models specialize in different process regimes. SUPCON describes TPT2 as part of a broader shift away from "N models for N applications" toward "a single TPT model as the foundation supporting multiple application scenarios." IDC's 2025 China Industrial Enterprise Survey, cited in SUPCON's launch materials, reported that the share of Chinese industrial enterprises using large models and agents jumped from 9.6 percent in 2024 to 47.5 percent in 2025, providing a tailwind for products such as TPT2.
On top of TPT and supOS, SUPCON delivers a portfolio of "industrial intelligent agents" that target specific operational scenarios, including safety agents that monitor process variables for early warning signs of upsets, quality agents that adjust setpoints to keep product specifications inside target windows, energy and emissions agents that minimize utility consumption subject to production targets, and equipment health agents that predict failures from vibration analysis, thermal trends, and historical maintenance records. SUPCON has stated that revenue from its industrial AI business surpassed RMB 100 million in 2024 and continues to grow rapidly as TPT2 and the agent portfolio mature.
In 2024, SUPCON anchored the establishment of the Zhejiang Humanoid Robot Innovation Center, an entity designed to bridge academic robotics research with industrial demand. Through this center, SUPCON launched its first humanoid robot platform, Navigator alpha. Reported specifications include a height of approximately 1.5 meters and weight of around 50 kg, lightweight mechanical arms with multiple degrees of freedom, and a dexterous hand with 15 finger joints, six active degrees of freedom, fingertip force of about 10 N, total hand mass around 600 grams, and joint speeds up to 150 degrees per second. The platform is intended to integrate with large AI models for natural language interaction and high-level task planning.
The innovation center secured around USD 48 million in early funding from Beijing Yizhuang Investment Holding. SUPCON has framed Navigator alpha as a platform for high-risk industrial tasks such as inspection of hazardous areas in chemical plants and field intervention work that complements its core process automation business. This places SUPCON alongside companies like UBTech, Unitree, AgiBot, Foxconn, and Tesla in the broader humanoid robot race, but with a distinctly industrial process focus.
SUPCON's customer base is concentrated in the process industries, with major segments spanning oil refining and petrochemicals, specialty and bulk chemicals, electric power generation, mining and metals processing, pulp and paper, pharmaceuticals, food and beverage manufacturing, building materials, and water and wastewater treatment.
For large engineering, procurement, and construction (EPC) firms, SUPCON acts as a one-stop integrated automation supplier, combining DCS or Nyx control systems, safety instrumented systems, instrumentation, MES software, supOS, and AI services. The company serves both brownfield retrofits, where it competes head-to-head with Honeywell, Emerson, ABB, Siemens, and Yokogawa, and greenfield projects, particularly in China, the Middle East, and Southeast Asia.
SUPCON has built out a global footprint with subsidiaries in Singapore, Saudi Arabia, the Netherlands, Japan, India, Malaysia, and Indonesia. Its products and services have been deployed in more than 50 countries. International revenue has been growing notably faster than domestic revenue, supported by exhibitions at Hannover Messe in Germany and ADIPEC in Abu Dhabi. The company has reported that overseas orders grew more than 35 percent year over year in 2024 and expects faster growth in 2025 from expansion in the Middle East, Central Asia, Europe, South America, and Southeast Asia.
In fiscal year 2024, SUPCON reported operating revenue of approximately RMB 9.139 billion, an increase of around 6 percent year over year, with net profit attributable to shareholders of approximately RMB 1.117 billion, up 1.4 percent. For the first three quarters of 2025, the company reported operating revenue of about RMB 5.654 billion (down approximately 10.8 percent year over year) and net profit attributable to shareholders of about RMB 432 million (down approximately 39.8 percent), a slowdown that reflected weaker capital spending across parts of the Chinese chemicals sector and pricing pressure on traditional DCS hardware. Industrial AI revenue and overseas revenue continued to grow strongly within this overall mix.
SUPCON's industrial AI strategy combines three building blocks. First, mechanism models, which are first-principles representations of physical and chemical processes derived from decades of process engineering experience. Second, AI models, especially the TPT family of time-series transformers and adjacent agents. Third, the supOS plant operating system as the runtime that brings these models, control systems, and operator workflows together. The company refers to this combination as a "mechanism and AI dual-engine" approach, which it argues is better suited to safety-critical process plants than purely data-driven AI.
SUPCON maintains close ties with Zhejiang University and the State Key Laboratory of Industrial Control Technology, where founder Chu Jian served as director. The company also collaborates with major Chinese refiners, chemical producers, and power generators on joint engineering projects that frequently double as proving grounds for new control architectures and AI agents. Through the Zhejiang Humanoid Robot Innovation Center, SUPCON has extended these partnerships into robotics and embodied AI.
In 2025, SUPCON announced that it had become the first industrial automation company to receive ISO/IEC 42001:2023 certification, the international Artificial Intelligence Management System (AIMS) standard. The certification covers SUPCON's processes for the research, development, deployment, and ongoing safety management of AI products. For an industrial AI vendor whose models are intended to influence operations in plants where failure modes can include fires, explosions, and toxic releases, this kind of governance certification is more than a marketing point: it provides a structured framework for demonstrating responsible AI practices to safety regulators and risk-averse customers.
| Year | Milestone |
|---|---|
| 1993 | Supcon founded by Chu Jian as a spin-out from the State Key Laboratory of Industrial Control Technology at Zhejiang University. |
| 1999 | Spun out from Zhejiang Highne to operate as an independent automation company. |
| 2000s | Rapid expansion of domestic DCS deployments across Chinese refineries, chemical plants, and power stations. |
| 2010s | Build-out of MES, historian, and engineering software portfolio above the DCS layer. |
| 2020 | Listing on the STAR Market of the Shanghai Stock Exchange (688777), raising approximately USD 268 million. |
| 2021 to 2023 | Launch and expansion of supOS plant operating system, including the open-source supOS-CE community edition. |
| 2024 | Launch of Nyx Universal Control System; anchoring of Zhejiang Humanoid Robot Innovation Center; Navigator alpha humanoid robot unveiled; DCS market share in China reaches around 40 percent. |
| 2025 | Release of TPT2 industrial large model on August 28; ISO/IEC 42001:2023 AI Management System certification; major appearances at Hannover Messe 2025 and ADIPEC 2025. |
In the Chinese DCS market, SUPCON's primary competitors are foreign incumbents such as Honeywell, Emerson, ABB, Siemens, and Yokogawa, plus a handful of domestic peers like HollySys. SUPCON's combination of cost-competitive hardware, deep local engineering presence, and broad spare parts and service coverage has allowed it to take share from foreign vendors, particularly in chemicals and petrochemicals.
In the global industrial AI space, SUPCON competes more directly with industrial software platforms from Aspen Technology, AVEVA, Siemens (MindSphere and related products), GE Digital, and Schneider Electric, as well as with hyperscale cloud providers offering industrial AI services. Its bet on a tightly integrated stack from the field instrument up to the foundation model is intentionally distinct from horizontal cloud AI offerings.
In humanoid robotics, SUPCON's Navigator alpha enters a crowded field that includes Foxconn industrial humanoid programs, Tesla's Optimus, Figure AI, 1X, Apptronik, UBTech, Unitree, AgiBot, and others. SUPCON's distinctive angle is its deep relationship with process plant operators, which gives it a clearer near-term industrial deployment story than many consumer-leaning humanoid programs.
Independent industrial automation analysts, including ARC Advisory Group and Control Engineering, have written favorably about SUPCON's Nyx UCS architecture, framing it as one of the more credible attempts in decades to fundamentally rethink the DCS paradigm. Coverage in Reuters, the South China Morning Post, Caixin, the Global Times, and trade publications such as Drives and Controls and The Robot Report has highlighted the company's role in China's industrial automation localization, its growing international footprint, and its recent moves into industrial large models and humanoid robotics.
At the same time, observers have noted execution risks. Cyclical weakness in Chinese chemicals capex weighed on 2025 results. Replacing entrenched DCS systems is a slow process, and many international customers are conservative about adopting cloud-native control architectures from any vendor. SUPCON's industrial AI revenue, while growing quickly, remains a small share of total revenue.