University calendar

Data Science MS Thesis Defense by Raksha Mohan

Thursday, April 23, 2026 at 12:00pm to 2:00pm

Title: "Predicting Microbiologically Influenced Corrosion Severity from Electrochemical Impedance Spectroscopy Using Interpretable Machine Learning"

by Raksha Mohan

Thesis Advisor: Maricris Mayes, Associate Professor, Chemistry & Biochemistry

Thesis Committee:
Firas Khatib, Associate Professor, Computer & Information Science
Donghui Yan, Associate Professor, Mathematics

Abstract:

Microbiologically influenced corrosion (MIC) accounts for an estimated 20-30% of global corrosion losses, yet reliable quantitative prediction of MIC severity remains unsolved. Electrochemical measurements combined with interpretable machine learning provide a promising framework for addressing this challenge. MIC depends on coupled microbial, biofilm, and interfacial electrochemical processes, and charge-transfer resistance (Rct) is a useful proxy for corrosion severity, as expressed in log鈧佲個(Rct) from electrochemical impedance spectroscopy (EIS). However, physically interpretable predictive models for MIC remain limited. Here, we show that machine learning regressors can predict log鈧佲個(Rct) in MIC systems involving Pseudomonas and Vibrio across varying environmental conditions.

A dataset of 116 EIS and 83 potentiodynamic polarization observations was compiled from ten peer-reviewed sources. Random forest and Gradient Boosting Machine (GBM) regressors were compared using stratified five-fold cross-validation, and Shapley additive explanations (SHAP) were used to interpret model behavior in physically meaningful terms. GBM outperformed Random Forest, achieving a higher cross-validated R². SHAP analysis identified double-layer CPE admittance as the dominant predictor of corrosion severity, consistent with its role as a reporter of biofilm-induced interfacial disorder, while genomic species descriptors contributed modest but interpretable signals consistent with known difference in metabolic versatility between the two organisms.

This work establishes a reproducible and interpretable baseline for quantitative MIC prediction and, to our knowledge, provides the first application of SHAP analysis to EIS-derived features in MIC while demonstrating the value of integrating genomic descriptors with electrochemical features for corrosion modeling.

Zoom Meeting ID: 922 3504 5299
Passcode: 941562

SENG 311 :
Maricris Mayes
maricris.mayes@umassd.edu

Add to my calendar