JULY 2026 NEWSLETTER

Applications of AI in Reliability (Fall 2026 Graduate Course)

This University of Maryland course brings together the disciplines of machine learning, reliability, and prognostics. It is taught by Prof. Michael Pecht, Dr. Michael Azarian, and Prof. Jay Lee. You can register now for this course to take it online or in person.

This is an interdisciplinary course, and students in many areas, including aerospace, civil, electrical, and mechanical engineering, as well as engineering management, are welcome. Students will have the opportunity to learn the basic scientific foundations of PHM and to work on its implementation in real-life applications through projects. In addition to the faculty, experts from industry, government, and academia will give guest lectures in this course.

Webinar - Chip Packaging Process-induced Stresses: Can We Go Beyond Prediction and Design It?

This Webinar presents a novel predictive scheme for simulating the entire compression molding process, including demolding and post-molding. The model incorporates all critical properties required for prediction, including (1) chemical and diffusion-controlled cure kinetics, (2) cure-induced shrinkage after the gel point, and (3) cure-extent-dependent viscoelastic Young's and bulk viscoelastic properties. Predictions of warpage under various molding parameters are presented, and their applications to package reliability are discussed. …

Presented by: Dr. Sukrut Phansalkar

Tuesday, July 14, 2026
11:00 am US Eastern

Battery Strain Sensing Project Featured in ULRI Stories of Impact

This CALCE project is highlighted in the UL Research Institutes’ Stories of Impact Series. The research focuses on using strain-sensing technology to catch internal gas buildup before it triggers thermal runaway, paving the way for safer consumer electronics and electric vehicles. A demonstration of strain-sensing technology for battery safety was given by Dr. Michael Azarian and Ph.D. candidate Sahithi Maddipatla, who is responsible for the study's design, execution, and analysis, and it was supported by CALCE Lab Director, Dr. Robert Utter.

IEEE Std 1413-2026: IEEE Standard for Framework for Reliability Prediction of Hardware is Published

Dr. Diganta Das has led the update and release of this standard, which covers the framework for reliability prediction of electronic hardware. The required elements for an understandable and credible reliability prediction, along with information to evaluate the prediction results and assess their usability for the intended application, are identified. A reliability prediction generated according to this standard would include sufficient information on inputs, assumptions, data sources, methodologies, and uncertainties to assess the risk associated with using the prediction results.

The original standard was conceptualized and developed under the leadership of CALCE Founding Director, Prof. Michael Pecht.

CALCE/SMTA Counterfeit Parts and Materials Symposium Proceedings Now Available

The proceedings from this year’s CALCE/SMTA Counterfeit Parts and Materials Symposium are now available. This year’s technical program featured eight thematic sessions covering policy and data, trusted sources, technology tools for avoidance, standards implementation, detection cases, and laboratory practices. Presentations included trusted micro-electronics, design for sourcing, AI-enhanced physical inspection, inkjet-printed silicon nanoparticle ink for anti-counterfeiting, and crucial SAE AS6171 updates. CALCE Consortium members and attendees can now access these proceedings. Others can contact SMTA for information on accessing the proceedings.

Prof. Han Wins IEEE EPS Outstanding Sustained Technical Contribution Award

Bongtae Han, Keystone Professor in the University of Maryland’s (UMD) Department of Mechanical Engineering, has been honored by the Electronic Packaging Society (EPS) of the Institute of Electrical and Electronics Engineers (IEEE) with its Outstanding Sustained Technical Contribution Award.

Prof. Han received the recognition in honor of his contributions to electronic packaging design for reliability and workforce development, and particularly for his work relating to measurement of advanced package behavior and material properties. The award was presented on May 28 at the 2026 Electronic Components and Technologies Conference in Orlando, FL. …

Professor Jay Lee to Chair New ASTM International Committee on AI in Manufacturing Systems

ASTM International’s board of directors has approved the creation of a new technical committee, F50, on artificial intelligence (AI) in manufacturing systems, chaired by Professor Jay Lee, Clark Distinguished Professor and director of the Industrial AI Center at the University of Maryland, College Park. The committee will focus on frameworks, best practices, and standards for the use of AI technologies in manufacturing.

An Attentional Deep Learning Model Based on Improved Crested Porcupine Optimizer for State of Health Prediction of Lithium-Ion Batteries

This paper aims to accurately predict and effectively manage the state of health (SOH) of lithium-ion batteries to maximize their utilization. Traditional SOH prediction models based on deep learning frequently encounter issues like inadequate representation capabilities of health features (HFs) and improper hyper-parameter settings. …

Investigation of the Interface Reliability of Multi-level Stacked Copper Microvias: Modeling and Simulation

Microvias serve as electrical interconnections between conducting layers of laminates in high-density interconnect printed circuit boards. However, stacked microvias can interfacially separate or crack due to fatigue if not properly designed and manufactured. …

Cross-domain Integrated Digital Twin for Heterogeneous Integrated CPUs

This cross-domain study presents an end-to-end digital twin system for heterogeneous integrated(HI) processors. This simulation framework is built to bridge the gap between architectural-level modeling and package-level physical design for HI systems. The framework combines architectural performance, power consumption, thermal profiles, and mechanical stress analysis into a unified digital twin environment that allows the system to be examined by various cross-domain metrics. …

Warpage Prediction After Compression Molding Using Cure-dependent Viscoelastic Properties

Unlike the transfer molding process, the mold temperature, pressure and time as well as the post-mold curing temperature and time can be process-parameters for the compression molding process simply because the transfer of an epoxy-based molding compound (EMC) through tiny channels does not have to occur. Accurate prediction capability is demanded to identify an optimum process condition. …