Machine Learning + Decades of Experience
PDF Solutions enables companies to reap the benefits of Industry 4.0 by combining a big data infrastructure and machine learning applications with decades of manufacturing and test experience to provide customers with solutions tailored to deliver achieve specific outcomes.
Proven in High-Volume Production
Our Advanced Insights for Manufacturing, or AIM, is a configurable, knowledge-based system that learns from on-going calculations and user inputs to rapidly make intelligent decisions in a high-volume production environment. Over the past decade, we have developed a family of production-proven AIM Solutions that leverage our big data and machine learning capabilities to deliver significant ROI to customers in the areas of manufacturing, test operations, and assembly & packaging.
AIM Solutions Overview:
Adaptive Signature Diagnostics (ASD)
ASD is an automated system for anomalous wafer yield signature detection, classification and root cause analysis. Incoming new wafers are analyzed for spatial signatures and classified into categories that are refined on-the-fly using machine learning based on user inputs, a technique we call Collaborative Learning. Drilldown analyses are automatically performed for each unique wafer category and reports are generated that confirm re-emergence of known root causes or highlight new potential sources of wafer yield loss.
ROI: Root causes of wafer yield loss are identified and contained 5x faster than conventional analytical techniques and expert knowledge is captured in machine learning models for continuous improvement.
Data Sources: Wafersort/Binsort, PCM, LEH/WEH, Metrology, Defect, Tool FDC
Capacity and Efficiency Improvement (CEI)
CEI leverages the Equipment Performance Tracking (EPT) capabilities within the Exensio Analytics Platform in a step-by-step methodology to optimize OEE (overall equipment effectiveness), fab capacity and wafer throughput by matching tools and chamber operations. Data collected from each recipe micro-step captures any performance mismatches between tools which are then eliminated through a detailed analysis of tool FDC data.
ROI: 10% improvement in bottleneck tool capacity, >20% improvement in efficiency/throughput, fast identification of recipe vs. setup and equipment hardware issues impacting deviation from manufacturing throughput model.
Data Sources: Tool sensor FDC data
Consumable Cost Reduction (CCR)
The CCR solution utilizes the eBOM dataset collected by the Process Control module (a core module of the Exensio Analytics Platform) which includes ERP, MES, EAM, and FAC data as well as consumables, maintenance parts and materials, and chemical and material composition reports, to systematically reduce material consumption, optimize usage, and catch material composition excursions. Poor performing tools, parts, and suppliers are identified through structured analytic workflows, guiding the user through the optimization process.
ROI: Reduce material consumption costs, reduce yield and reliability excursions and optimize parts and material usage
Data Sources: Consumable batch ID, Event data, Recipe ID, FDC data, PM information, Material Composition reports, MES data
Early Life Failure Detection (ELF)
The ELF solution optimizes cost-of-quality tradeoffs between yield and reliability failure in the field. Classical outlier algorithms, such as “Part Average Testing” (PAT) are often used to identify and screen parts with a risk for early life failure. Exensio’s ELF goes beyond PAT and provides a comprehensive die quality grading and risk classification solution by taking advantage of the Exensio Analytics Platform’s end-to-end database and infrastructure. Advanced indicators generated from multiple data sources are analyzed with a multi-variate machine learning approach that adapts as new information comes to light (e.g. 8D reports, root causes found in FA, additional incoming RMA’s, etc.).
ROI: Prevent quality and reliability escapes by detecting high risk die at Wafer Sort
Data Sources: Wafer Sort, Final Test, PCM, Burn-in, Returns, Defect, Metrology, LEH/WEH, FDC
Equipment Trouble Protection (ETP)
ETP is the next generation FDC solution for wafer fabs and assembly floors. Going beyond the standard approach of FDC data collection, feature selection and SPC alarm limits, ETP links FDC data with tool events and uses AI and ML to detect and classify abnormal equipment sensor traces into good vs. bad vs. unknown. The classification system adapts as users judge new signals and identify root causes, enabling fast issue detection and containment.
ROI: +1% DPW Yield, +5% Fab Output, +2% Line Yield, +2% Tool Availability, Engineering FTE savings
Data Sources: Tool sensors FDC data and tool events
Intelligent Material Disposition (IMD)
A “Material Review Board” (MRB) is a common technique used to improve the quality of shipped product. The IMD solution dramatically reduces the manual work and frequency of human-induced variability in the MRB process. Automated workflows are implemented that capture the specific quality criteria of each customer’s business and product line to provide lot and wafer quality grading in minutes, rather than hours or days. Comprehensive analytics and full automation ensure uniform results and high quality decision making.
ROI: Reduces engineering effort for lot disposition by >50%. Prevents escapes, improves consistency, and decision quality of wafer dispositioning.
Data Sources: PCM/WAT, Wafersort/Binsort, Final Test
Smart Testing
Manufacturing complexity, advanced packaging technology, and high density chip designs conspire to drive up the cost of wafersort and final test. Exensio’s Smart Testing solution uses Machine Learning to find subtle signals in the massive data set associated with each product die and applies Artificial Intelligence to modulate test flows and achieve higher product quality at lower cost. The AI / ML approach identifies the highest quality die as candidates to skip expensive test insertions such as Burn-in, optimizing test cost while still meeting DPPM requirements. PDF can provide the machine learning algorithm, or you can provide your own. The system is designed for production operation, installed on your OSAT’s test floor “at the edge” for efficient, low latency operation, high uptime and minimal data loss.
ROI: Reduces burn-in requirements by 30-60%, saving up to millions of dollar per year depending on volume and cost of test.
Data Sources: PCM/WAT, Wafersort, Final Test, (and Metrology, Defect, MES, and FDC data as available)
Yield Aware FDC
YA-FDC is a combination of technology and services that leverages Exensio’s “big data” platform to improve process variation, identify equipment conditions and sources of variability that influence functional and parametric yield, and set appropriate SPC limits with proprietary analysis and modeling techniques that identify critical parameters. Analysis is automated with reporting and dashboards to drive fast improvement yield, variation and excursions. AI / ML provides predictive models to for finer feedback and feed-forward control, predictive PM’s to optimize tool availability and Virtual Metrology for adaptive inline sampling.
ROI: +8% yield improvement, +40% excursion reduction, +7% faster NPI ramp learning rate
Data Sources: Tool sensor FDC data, Metrology, Defect, PCM/WAT, Wafersort, Test, Assembly
