# PDF Solutions Published articles for PDF Solutions. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## How Much Can Semiconductor Models Hide? DevFeed: [How Much Can Semiconductor Models Hide?]() Original publisher: [Read original article]() Author: Gregory Haley Published: 2026-10-28T07:01:04Z Content type: article Language: en Sources: [Semiconductor Engineering]() Topics: [Chip design](), [Data Quality]() Tags: [3d](), [3d-ic](), [about](), [account](), [added](), [advanced](), [advanced-packaging](), [ai](), [amkor](), [ansys](), [ase](), [brewer-science](), [but](), [can](), [chiplets](), [company](), [cross](), [digital-twins](), [eda](), [engineering](), [first](), [heterogeneous-integration](), [ip](), [make](), [manufacturing-packaging-and-materials](), [modeling](), [models](), [osats](), [packaging](), [pdf-solutions](), [prediction](), [processes](), [reduced-order-models](), [requires](), [silvaco](), [simulation](), [substrates](), [synopsys](), [thermal](), [top-stories]() ### AI overview Advanced packaging models must expose enough process behavior for other companies to predict how components work together while keeping recipes and formulations private. The article describes encrypted HFSS models as a way to use simulation models without revealing geometry or materials, and reduced-order or surrogate models as ways to make analysis practical; neither alone establishes that a model is valid for a given design. Surrogates are more reliable within their training region than outside it, while material properties can shift with temperature, curing, moisture, time, and interfaces. Useful handoffs therefore need to preserve relevant behavior and carry context such as data origin, calibration, assumptions, validity range, and intended decisions. The source emphasizes that model builders may still need detailed process recipes to create predictive models, even when users see only selected parameters. ### Source excerpt Advanced packaging requires models to cross company boundaries without exposing proprietary processes, but hiding too much can make the prediction break down. The post How Much Can Semiconductor Models Hide? appeared first on Semiconductor Engineering. ## Shift Left Complicates Fab Data Management DevFeed: [Shift Left Complicates Fab Data Management]() Original publisher: [Read original article]() Author: Anne Meixner and Laura Peters Published: 2026-10-08T07:02:25Z Content type: article Language: en Sources: [Semiconductor Engineering]() Topics: [data-engineering](), [data-architecture](), [Feature Engineering](), [AI Chat]() Tags: [advanced-process-control](), [ai-ml-inference](), [apc](), [data](), [data-management](), [data-veracity](), [data-volume](), [digital-twins](), [engineering](), [equipment-sensors](), [intel-foundry](), [lam-research](), [onto-innovation](), [pdf-solutions](), [semiconductor](), [shift-left](), [siemens-eda](), [streaming](), [synopsys](), [test-measurement-and-analytics](), [top-stories](), [virtual-twins](), [yield]() ### AI overview Semiconductor fabs are using high-resolution equipment sensor data and AI/ML inference to detect yield problems earlier and improve in-situ process control. The resulting growth in data volume and speed creates infrastructure and data-management challenges. Fabs must preserve context and traceability across process modules so that siloed local optimization does not undermine cross-fab analytics and model quality. ### Source excerpt Streaming sensor data creates new challenges for advanced process control. The post Shift Left Complicates Fab Data Management appeared first on Semiconductor Engineering. ## From Process Learning To Production Control: Trends In Advanced Packaging And Implications To Characterization And Test DevFeed: [From Process Learning To Production Control: Trends In Advanced Packaging And Implications To Characterization And Test]() Original publisher: [Read original article]() Author: PDF Solutions Published: 2026-10-08T07:02:08Z Content type: article Language: en Sources: [Semiconductor Engineering]() Topics: [Chip design](), [tsmc]() Tags: [advanced-packaging](), [charaterization](), [heterogeneous-integration](), [hybrid-bonding](), [interposers](), [metrology](), [pdf-solutions](), [process-control](), [semiconductor](), [semiconductor-test](), [stacking](), [test-measurement-and-analytics](), [whitepapers](), [yield]() ### AI overview The presentation describes an integrated methodology for accelerating process development and maintaining production control in advanced packaging. It highlights distinct yield sensitivities and defect mechanisms in heterogeneous integration technologies, including interposers, hybrid bonding, and 3D stacking, and the need for tailored metrology, characterization, and analytics. ### Source excerpt An integrated methodology for accelerating process development and sustaining production control for advanced packaging technologies. The post From Process Learning To Production Control: Trends In Advanced Packaging And Implications To Characterization And Test appeared first on Semiconductor Engineering. ## How Robotics And Intelligent Equipment Control Drive Next-Gen Fab Productivity DevFeed: [How Robotics And Intelligent Equipment Control Drive Next-Gen Fab Productivity]() Original publisher: [Read original article]() Author: Laura Peters Published: 2026-09-21T07:01:36Z Content type: article Language: en Sources: [Semiconductor Engineering]() Topics: [Robotics](), [Automation](), [Artificial Intelligence](), [Computing](), [autonomous vehicles]() Tags: [acoustic-inspection](), [ai](), [automation](), [autonomous-robot](), [boosting](), [cobot](), [computing](), [equipment-control](), [intel-foundry](), [lam-research](), [manufacturing-packaging-and-materials](), [microtronic](), [pdf-solutions](), [preventive-maintenance](), [robotics](), [thermal-inspection](), [tool-fingerprinting](), [tool-to-tool-matching](), [top-stories](), [umc](), [uptime](), [visual-inspection](), [yield]() ### AI overview The article describes how robotics, automation, AI, and equipment-control systems improve semiconductor-fab productivity by increasing repeatability, uptime, and yield. It also explains how connected material, equipment, and wafer data helps engineers detect variation, enforce operating limits, and prevent processing problems. ### Source excerpt Behind the scenes, advanced equipment control and robotics are enabling higher levels of repeatability, tool uptime, and yield. The post How Robotics And Intelligent Equipment Control Drive Next-Gen Fab Productivity appeared first on Semiconductor Engineering.