Shenzhen – October 02, 2026 -- An artificial intelligence model jointly developed by XtalPi Holdings Limited (2228.HK) and Fangda Carbon New Material Co., Ltd. (600516.SH) has passed project acceptance testing and entered full operational use within Fangda Carbon's production workflow.
The system ranks raw material formulation candidates before physical trials, meeting the companies' established targets for predictive accuracy, cost optimization, and operational efficiency.
Model cuts reliance on costly physical trials for graphite electrode production
Graphite electrodes conduct the high electrical currents required for electric arc furnace steelmaking, with conductivity and mechanical strength depending on complex interactions among raw materials, blend ratios, and manufacturing conditions. Because input materials vary by supplier and batch, formulations must be reassessed whenever inputs change, and exhaustively testing every combination through physical trials is prohibitively slow and expensive.
XtalPi structured six decades of Fangda Carbon production data into predictive algorithms
The partners combined Fangda Carbon's six decades of proprietary manufacturing data with XtalPi's data engineering and algorithmic capabilities. XtalPi structured dispersed production records, engineered predictive features, and deployed a hybrid system integrating performance forecasting, optimization algorithms, and embedded expert rules.
System identifies cost-cutting blend ratios and flags substitute materials during supply shifts
For any fixed set of raw materials, the model identifies blend proportions that reduce costs while meeting quality requirements. When supply availability or pricing shifts, the system evaluates alternative inputs and recommends formulation adjustments, expanding Fangda Carbon's procurement flexibility.
Validation confirms model meets accuracy and speed requirements for production use
Validation against independent test datasets and production trials confirmed the model achieved required predictive accuracy across multiple key performance indicators, with computational speed matching the pace of industrial production. The AI narrows the field for human experts, who retain final authority over formulations through experimental or production validation.
Deployment follows 2025 strategic agreement, sets stage for graphene and carbon nanotube applications
The model is the first core module delivered under a graphite electrode formulation optimization project stemming from a strategic agreement signed in 2025 between the two companies. The partners plan to expand the system into comprehensive formulation design and process optimization, with XtalPi targeting adaptation of the infrastructure to additional advanced carbon materials including graphene and carbon nanotubes.