A Hybrid Response Surface Methodology and Machine Learning Framework for Quantifying Effects of Physicochemical
Harsh V Patel1, Jazmin Green2, Hyoshin Park3
1Department of Civil, Architectural, and Environmental Engineering, North Carolina A&T State University, Greensboro, North Carolina 27411, United States.
ACS ES&T Water
|May 14, 2026
Summary
A new hybrid model predicts per- and polyfluoroalkyl substances (PFAS) adsorption by integrating Response Surface Modeling with machine learning. This approach quanties the complex interplay of adsorbent properties, PFAS chemistry, and operational conditions for improved environmental risk assessment.
Area of Science:
- Environmental Chemistry
- Adsorption Science
- Predictive Modeling
Background:
- Per- and polyfluoroalkyl substances (PFAS) adsorption prediction is complex due to multiple influencing factors.
- Existing models struggle to account for adsorbent properties, PFAS characteristics, and operational conditions simultaneously.
- Accurate PFAS adsorption prediction is crucial for environmental remediation and risk assessment.
Purpose of the Study:
- To develop and evaluate a unified hybrid modeling framework integrating Response Surface Model (RSM) with machine learning algorithms.
- To quantify the impact of six key variables (surface area, Log Kow, pHpzc, pKa, log-dose, log-initial concentration) on PFAS distribution coefficients (Log Kd).
- To provide a transparent method for predicting PFAS adsorption and understanding the interplay of influencing factors.
Main Methods:
- Compiled and preprocessed a dataset of over 1000 PFAS adsorption observations.
- Integrated Response Surface Model (RSM) with machine learning algorithms (Gradient Boosting, Random Forest).
- Evaluated model performance using 80/20 split and Leave-One-PFAS-Out (LOPO) cross-validation.
Main Results:
- Gradient Boosting achieved high performance (R2 = 0.93) in the 80/20 split scenario.
- Random Forest showed better generalization under LOPO validation (R2 = 0.30), indicating compound-specific prediction challenges.
- Feature importance analysis identified log-dose and log-initial concentration as primary predictors, followed by surface area and pHpzc.
- Nonlinear relationships and interaction effects among adsorption drivers were revealed.
Conclusions:
- The hybrid RSM-machine learning framework offers a transparent approach for predicting PFAS adsorption.
- The model effectively disentangles the coupled roles of adsorbent properties, PFAS chemistry, and operational conditions.
- Findings highlight the importance of considering multiple factors and their interactions for accurate PFAS adsorption prediction.
Keywords:
adsorptiondistribution coefficientmachine learningper- and polyfluoroalkyl substance (PFAS)response surface methodologyMore Related Videos
07:06Investigating Long-Distance Transport of Perfluoroalkyl Acids in Wheat via a Split-Root Exposure Technique
Published on: September 28, 2022
06:00Optimization of the Epimedii Folium Mutton-Oil Processing Technology and Testing Its Effect on Zebrafish Embryonic Development
Published on: March 17, 2023
Related Concept Videos
01:16Response Surface Methodology
Response Surface Methodology (RSM) is a collection of statistical and mathematical techniques used to develop, improve, and optimize processes. It is particularly valuable when many input variables or factors potentially influence a response variable.
The process of RSM involves several key steps:
The process of RSM involves several key steps:
01:28Methods of Medium Optimization
Optimizing growth media enhances microbial proliferation and maximizes product yield. Statistical experimental design methodologies provide structured and reproducible approaches, offering progressively higher levels of robustness and efficiency.The One-Factor-at-a-Time (OFAT) MethodThe One-Factor-at-a-Time (OFAT) method involves adjusting a single variable while keeping all others constant. However, it cannot detect interactions between variables, often leading to suboptimal outcomes when...
01:29Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
01:06Model Approaches for Pharmacokinetic Data: Distributed Parameter Models
Pharmacokinetic models are mathematical constructs that represent and predict the time course of drug concentrations in the body, providing meaningful pharmacokinetic parameters. These models are categorized into compartment, physiological, and distributed parameter models.
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
The distributed parameter models are specifically designed to account for variations and differences in some drug classes. This model is particularly useful for assessing regional concentrations of anticancer or...
01:27Pharmacodynamic Models: Overview
Pharmacodynamic (PD) responses describe the interaction between a drug and its biological target, culminating in a physiological effect. These responses can be classified into different types: continuous variables, such as blood glucose levels; categorical outcomes, like survival rates; and time-to-event metrics, such as disease progression. Understanding and modeling PD responses are critical for optimizing drug efficacy and safety.PD models describe the relationship between drug concentration...
01:14Pharmacokinetic–Pharmacodynamic Relationship: Model Components
Pharmacokinetic-pharmacodynamic (PK–PD) modeling is essential in drug development and clinical pharmacology. It provides a quantitative framework to predict drug behavior and response over time. This approach integrates pharmacokinetics (PK), which describes the drug's absorption, distribution, metabolism, and excretion, with pharmacodynamics (PD), which characterizes the drug’s biological effects and mechanisms of action.The disposition kinetics of a drug determine its plasma...