Northeast: The demand for automation equipment for optical modules and AI servers is expected to increase.

Northeast released a research report stating that domestic equipment manufacturers have achieved large-scale breakthroughs in mid-to-late-stage packaging, coupling, and functional testing, gradually replacing imported equipment, forming advantages in cost and service response. With the continuous expansion of AI server shipments and the promotion of manufacturing process standardization, automated assembly and testing equipment will become the core basic equipment for hardware manufacturing, supporting the simultaneous improvement of industry capacity and quality. The main points of Northeast are as follows: Optical modules: Manufacturing and testing equipment support technology iteration and capacity release Under the background of the continuous expansion of AI computing power networks and the acceleration of high-speed optical module iteration, optical module manufacturing and testing equipment have become the core constraint links for industrial chain capacity release, quality control, and technological upgrading. The localization and automation level of equipment directly determine the industry's supply capacity and cost competitiveness. From the production process perspective, optical module manufacturing involves multiple precise processes such as wafer testing, TO packaging, optical coupling, high-speed electrical performance tuning, reliability aging, and final testing, corresponding to the formation of four major core equipment categories: packaging equipment, coupling alignment equipment, high-speed testing equipment, aging and reliability equipment. With the mass production of 800G and the gradual introduction of 1.6T, the requirements for high-speed signal integrity have significantly increased, and traditional semi-automatic production lines are unable to meet consistency and yield targets, driving the industry towards intelligent transformation of fully automated packaging coupling, high-speed automated testing, and full-process data tracing. The core bottleneck of equipment lies in high-precision optical alignment platforms, high-speed optoelectronic testing systems, automatic loading and unloading, and flexible production line scheduling. The alignment accuracy of high-precision coupling equipment needs to reach sub-micron level, and the high-speed electrical testing equipment needs to support 112G/224G PAM4 signal generation and analysis, posing higher requirements on equipment manufacturers' optical, RF, and automation integration capabilities. The bank believes that domestic equipment manufacturers have achieved large-scale breakthroughs in mid-to-late-stage packaging, coupling, and functional testing, gradually replacing imported equipment, forming advantages in cost and service response. With the continuous increase in the penetration rate of high-end optical modules, the space for domestic replacement of front-end and high-end testing equipment for optical modules has further opened up, becoming a key foundation for supporting the expansion of the optical module industry scale and technological iteration. AI servers: Automation assembly and testing equipment strengthen the foundation of intelligent manufacturing AI servers feature multiple GPU clusters, high-speed interconnection, and high-density computing power as core characteristics, driving the industry to build an intelligent manufacturing system with full-process automation assembly and full-dimensional automated testing. The automation assembly process focuses on precision mechanical docking, modular integration of boards and modules, collaborative assembly of computing units, and other core processes, relying on multi-axis precision assembly from Siasun Robot & Automation, flexible positioning fixtures, and intelligent material scheduling systems to achieve high-precision assembly of key components such as GPU modules, high-speed interfaces, and heat dissipation components, effectively avoiding human operational deviations, and adapting to the large-scale and standardized production capacity expansion demands of AI servers. The automation testing process covers component-level, board-level, and whole-machine-level quality verification, completing automated verification in dimensions such as electrical performance, interface connectivity, computing unit operation conditions, and structural reliability through automated testing platforms, online operational monitoring systems, and whole-machine pressure and temperature cycle testing equipment. It achieves real-time data collection, automatic anomaly detection, and full-process quality traceability, providing stable quality control support for the large-scale delivery of AI servers. The report believes that with the continuous expansion of AI server shipments and the advancement of manufacturing process standardization, automated assembly and testing equipment will become the core basic equipment for computing hardware manufacturing, supporting the simultaneous improvement of industry capacity and quality. Related companies: Assembly: Bozhon Precision Industry Technology, Guangdong Anda Automation Solutions; Testing: Far East Smarter Energy, Lionhead Technology Development, Shenzhen Colibri Technologies, Shenzhen Everbest Machinery Industry, Quest Intelligence (to be listed). Risk warning: Policy progress falls short of expectations; downstream demand falls short of expectations, etc.