News

Industrial AI is no longer a speculative investment for Chinese automation suppliers—it has become a measurable revenue driver. SUPCON Technology, one of China's leading distributed control system providers, reported in late September 2026 that its industrial AI business generated 385 million yuan in the first half of the year, representing year-over-year growth of approximately 230 percent. The figure accounts for 10.6 percent of the company's total revenue, demonstrating that artificial intelligence has moved from laboratory demonstration to commercial deployment in process industries.

The revenue growth comes despite broader headwinds facing the automation sector. SUPCON's total revenue for the first half of 2026 reached 3.635 billion yuan, a 5.1 percent decline compared to the same period last year. Net profit fell 56.2 percent to 155 million yuan. The company attributes this pressure to strategic investment in AI capabilities and reduced government software tax rebates, suggesting that management is deliberately sacrificing short-term profitability to build long-term competitive position in industrial intelligence.

The core of SUPCON's AI strategy centers on its Time-Series Pre-trained Transformer, known internally as TPT. This large language model, trained specifically on industrial process data, has been deployed across 13 different industries including petrochemicals, chemicals, pharmaceuticals, and building materials. The model analyzes time-series data from distributed control systems, identifying patterns, predicting equipment failures, and recommending operational adjustments that human operators might miss.

What distinguishes SUPCON's approach from generic AI applications is the integration with its existing control system platform. The company's UCS, or Universal Control System, serves as both the data source and the execution layer for AI recommendations. Rather than operating as a separate analytics platform that provides insights for human decision-making, the TPT model can directly influence control parameters through the UCS architecture, enabling what the company calls autonomous operating plant, or AOP, functionality.

This integration matters because it addresses a fundamental challenge in industrial AI adoption: the gap between insight and action. Many AI platforms can analyze data and generate recommendations, but implementing those recommendations requires human operators to interpret results, validate safety constraints, and manually adjust setpoints. By embedding AI directly into the control layer, SUPCON reduces this friction, enabling faster response to changing conditions and more consistent application of optimization strategies.

Case studies from early deployments illustrate the potential impact. At Wanhua Chemical, one of China's largest chemical manufacturers, SUPCON's AI system has reportedly delivered annual cost savings exceeding 10 million yuan while improving labor productivity by 50 percent. The system monitors critical process parameters, predicts equipment degradation, and adjusts operating conditions to maintain optimal performance. At Hubei Xingrui, another chemical producer, the AI platform reduced alarm frequency by 99.8 percent and improved operator efficiency by 69 percent, according to company reports.

The alarm reduction figure deserves attention. In traditional control rooms, operators face hundreds or even thousands of alarms per shift, many of them nuisance alarms or cascading alerts triggered by a single underlying issue. This alarm flood makes it difficult for operators to identify genuine problems requiring intervention. AI systems that can distinguish between critical and non-critical conditions, suppress nuisance alerts, and highlight root causes directly improve operator effectiveness and reduce the risk of missed warnings.

SUPCON's market position provides a structural advantage for AI deployment. The company holds a 45.1 percent share of China's DCS market, making it the dominant supplier for process control in the country. This installed base represents millions of control loops generating real-time data across thousands of facilities. Each installation becomes a potential AI deployment site, with data already flowing through SUPCON's infrastructure. The company does not need to convince customers to adopt new hardware or migrate to different platforms—it can layer AI capabilities onto existing systems.

The business model for industrial AI is also evolving. SUPCON has shifted from one-time software licenses to subscription-based offerings, including Model-as-a-Service and Robotics-as-a-Service arrangements. This approach reduces upfront costs for customers while creating recurring revenue streams for the supplier. It also enables continuous model updates and improvements, as the AI system can be retrained on new data without requiring customer intervention.

The expansion into robotics represents another dimension of SUPCON's AI strategy. The company's Plantbot solutions combine mobile robots with AI-driven decision-making, enabling autonomous inspection, sampling, and basic maintenance tasks in hazardous or repetitive environments. These robots integrate with the same control systems that manage the broader production process, creating a unified automation architecture from field devices to mobile platforms.

International expansion is underway. SUPCON has established presence in Southeast Asia, the Middle East, and other regions where Chinese process industries have invested heavily. The company's participation in OGA 2026 in Malaysia and Gastech 2026 in Thailand demonstrates its intent to compete globally, not just domestically. However, international success will depend on more than technology capability—it requires local support, regulatory compliance, and integration with existing automation ecosystems that may include competing DCS platforms.

The competitive response from other automation suppliers is inevitable. Emerson, Honeywell, Yokogawa, and ABB all offer AI-enabled control systems and analytics platforms. What differentiates SUPCON in the Chinese market is its scale, its integration depth, and its willingness to accept near-term profit pressure for long-term positioning. Whether this strategy succeeds globally remains to be seen, but the domestic results suggest that industrial AI has crossed the threshold from experiment to essential capability.

For control system engineers and plant managers, SUPCON's experience offers practical lessons. First, AI deployment requires robust data infrastructure—reliable sensors, consistent data quality, and sufficient historical records for model training. Second, integration with existing control systems matters more than standalone analytics platforms. Third, the business case must be clear: AI should deliver measurable improvements in safety, efficiency, or cost that justify the investment. SUPCON's reported results suggest these conditions can be met, but success requires careful planning and realistic expectations.

The broader implication for the automation industry is that AI is no longer optional. Companies that cannot demonstrate AI capabilities risk losing competitive position as customers increasingly expect intelligent features as standard functionality. The question is not whether to adopt AI but how to do it effectively, balancing innovation with the reliability and safety requirements that define industrial operations.

Written by: Maxwell, an industrial automation specialist with over 15 years of experience in process control systems and digital transformation.