On May 27, 2026, the the 10th Jiwei Semiconductor Summit 2026 themed “AI Reshapes the Future, Ecosystem Collaboration Drives Long-term Growth” grandly held at the Zhangjiang Science Hall in Shanghai. As an influential annual summit in the semiconductor industry, the conference brought together nearly 200 industry leaders and more than 70 listed companies to chart new pathways for industrial innovation.
At the Advanced Packaging & Testing Technology Innovation Summit, Zhang Hao, Head of the Solution Center at Glorysoft, delivered a keynote speech themed “Heterogeneous Integration and Digital Intelligence Empowerment”. He showcased Glorysoft’s in-depth layout and cutting-edge practices in the field of advanced packaging intelligence.
Zhang Hao, Head of the Solution Center at Glorysoft, delivered a keynote speech at the 10th Jiwei Semiconductor Summit 2026
#01
Advanced Packaging: From “Back-end Process” to “Core Procedure”
The semiconductor industry is undergoing a profound paradigm shift. As Moore’s Law approaches its physical limits, advanced packaging is no longer merely a back-end process as traditionally perceived. Instead, it has become a core procedure determining the final performance, power consumption and cost of chips.
#02
Industry Challenges: Rapid Technology Iteration amid Backward System Infrastructure
Technological iteration in the industry far outpaces the progress of factory intelligent transformation. Most manufacturers are forced to deploy multiple sets of CIM systems requiring constant patching and upgrades. The lack of holistic planning in initial architecture leads to poor interoperability, making capacity expansion and system updates extremely difficult.
#03
Glorysoft’s Solution: One Unified Ecosystem Connecting the Entire Industrial Chain
Drawing on 19 years of experience developing CIM systems for the packaging and testing industry, Glorysoft has built a unified platform covering all advanced packaging businesses, including 8/12-inch fully automatic wafer lines, Bumping/TSV/RDL, 2.5D/3D WLP/PLP, DP, CP, conventional packaging & testing, and module assembly.
Built upon CIM, AI and big data technologies, the platform establishes an industrial large model equipped with personnel knowledge base. Combined with the LOFA and a unified factory portal, it ultimately achieves autonomous decision-making in lights factories.
#04
Comprehensive Penetration and Application of AI
This integrated ecosystem optimizes not only individual production segments but also connects end-to-end data pipelines spanning production, quality inspection and R&D, achieving dual breakthroughs in cost reduction, efficiency improvement and quality enhancement.
Key AI Application Scenarios on Advanced Packaging Lines
LOFA+IGI / LOFA+AI
Digital workforce, engineering assistants and management aids covering WAT/CP/Saw/AOI workstations, replacing large volumes of repetitive technical operations and supporting decision-making.
→ Up to over 75% savings on technical manpower
Big Data+AI
Leverage wafer/die selection algorithms, yield prediction and analysis, deep excavation of failure data, and intelligent root cause analysis of abnormalities to shift from passive troubleshooting to active prediction.
→ Additional predictive capabilities bring substantial yield gains
PMS+AI
Generate fault diagnosis solutions, support predictive equipment maintenance and accurate spare parts demand forecasting, shifting maintenance strategy from breakdown repair to proactive prevention.
→ Over 20% cost savings verified in practical projects
Automatic instrument reading, image defect recognition, automated traceability for quality checks, SOP behavior monitoring and patrol inspection to build an end-to-end quality control barrier.
→ Further reduce quality losses when coordinated with other systems
Empowered by GloryIGI
Automated code generation and testing, auto-documentation and low-code development support, significantly lowering barriers for system development and maintenance.
→ More than 75% savings on development and maintenance costs
#05
Future Factories: From “Observing” to “Predicting”
In terms of form: Digital twin evolve from passive replication to active simulation and deduction.
In terms of operation: The operation mode shifts from passive response to proactive prediction.
In terms of technology: Industrial large models deliver comprehensive empowerment, enabling factories to move from observing current status to foreseeing future trends.
Centered on the vision of “one ecosystem connecting the whole chain”, Glorysoft transforms the complex process of advanced packaging into digital assets that AI can understand, intervene in and optimize. Manpower reduction, yield enhancement and lower operation costs all demonstrate that with heterogeneous integration enabled by digital intelligence, smart factories are no longer a distant vision but an ongoing efficiency revolution.Taking CIM+AI as its technical foundation, Glorysoft will continue to collaborate with industrial partners to advance toward the future of advanced packaging featuring “self-perception, self-decision-making and self-optimization”.
