|About this Abstract
||7th World Congress on Integrated Computational Materials Engineering (ICME 2023)
||Data-Driven Modeling for Service Lifetime Prediction of Acrylic Polymers
||Hein Htet Aung, Jayvic Cristian Jimenez, Leean Jo, Roger H. French, Laura S. Bruckman
|On-Site Speaker (Planned)
||Hein Htet Aung
Polymethyl methacrylate (PMMA) is a thermoplastic polymer known for its optical clarity and hardness, among other properties, making it a commodity polymer. However, PMMA is susceptible to degradation arising from environmental factors such as UV, heat, and humidity. These stressors lead to performance loss and reduced lifetime service. In our work, six different formulations of PMMA containing various UV absorbers and stabilizers are exposed under three accelerated conditions, as per ASTM G154 and G155, in a longitudinal stepwise manner. We then quantified their performance metrics using colorimetric methods such as the Yellowness Index and obtaining UV-vis, fluorescence and Fourier-transform infrared (FTIR) spectra. The data collected are then modeled with netSEM (Network Structural Equation Modeling), which provides inferential and predictive data-driven models. These data-driven models are selected based on adjusted R2 values and stepwise regression is performed using either Akaike (AIC) or Bayesian (BIC) Information Criteria to avoid complexity and overfitting.
||Planned: Other (describe below)