ACS – Joint Meeting with the Cleveland Section

Data and AI-Driven Material Development
in the Tire & Rubber Industry

Schedule:
5:30pm: Social time
6:00pm: Dinner
7:00pm: Talk

Menu: TBD

Registration: Buy tickets at the links below. $25 for members and guests; $10 for retirees and unemployed. Undergrad and grad students and award winners are free but please RSVP at the link below.

Abstract
      There are over 300 million new tires produced each per year in the United States.  Tires are a complex blend of materials, engineering, and design.  The typical tire is made up of approximately 80-85% rubber compound by weight.  These compounds require constant improvement to meet the changing demands of customers (e.g., longer lifetimes, better grip in snow/ice) and to keep up with broader changes in society (e.g., environmental sustainability, transition to battery electric vehicles). For the past century, the tire industry has accumulated vast amounts of data and expertise on rubber compounds, their compositions/properties, and how these relate to tire performance.
In the last decade, the chemical industry, including tire and rubber, have started employing data driven approaches such as machine learning (ML) and artificial intelligence (AI) for material
development.   A typical problem involves developing ML/AI models to predict compound properties using information about the raw materials in the compound formula, process parameters, and other relevant factors.  The benefits of using ML/AI tools include accelerating the pace of product development by reducing the volume of experimental testing needed and using the tools in an inverse fashion to suggest compound formulas likely to achieve certain targets. However, the relationships between the rubber compound formulae and performance are complex. Nonlinearities, missing/incomplete data, noisy data, sparseness of the data, and high-dimensional feature spaces pose technical challenges requiring expertise at the intersection of material science and data science.
This presentation will discuss some of the unique opportunities and challenges faced by scientists and engineers in the tire and rubber industry responsible for data driven material
development.

BIOS

Ron Shaffer is Senior Manager, West Material Digitization (DX) and Approvals at the Bridgestone Americas Technology Center in Akron, Ohio. The West Material DX team develops AI-driven digital tools to accelerate material innovation and optimize rubber compound formulations. Before joining Bridgestone in 2018, Dr. Shaffer held scientific and leadership roles at the Naval Research Laboratory, General Electric Global Research Center, and Centers for Disease Control and Prevention. He earned his Ph.D. in Chemistry from Ohio University in 1996 and is the author or co-author of over 90 peer-reviewed publications (H-index = 49, 8000+ citations) and 7 US patents.

Rocco Panella is a Lead Data Tools Developer in the West Materials Digitization Team at the Bridgestone Americas Technology Center in Akron, Ohio. He leads development on virtual materials discovery as well as other data science efforts in materials for Bridgestone. His prior experience includes deploying manufacturing-focused machine learning systems at Intel Corp and Nestle. Rocco also developed data visualization and analytics software for NAVAIR as a principal engineer at BGI, and was the principal investigator on several Small Business Innovative Research (SBIR) grants through the Department of Defense. He received his PhD in Chemical Engineering from Carnegie Mellon University in 2013.