One of the most promising tools in that arsenal is the electro-Fenton process, an electrochemical advanced oxidation technology that generates hydroxyl radicals, among the most reactive oxidizing species known, in situ within the wastewater itself. In a typical electro-Fenton configuration, oxygen is reduced at a cathode to produce hydrogen peroxide, while ferrous iron added as a catalyst reacts with that peroxide in the classic Fenton reaction to yield hydroxyl radicals capable of shredding persistent organic molecules into smaller, less harmful fragments. Because the process relies on electricity rather than continuous chemical dosing of hydrogen peroxide, it is comparatively safe, controllable, and compatible with renewable power. Studies have demonstrated strong performance in degrading fluoroquinolones and other recalcitrant pharmaceuticals, and reviews of the technique highlight its versatility for both decontamination and nutrient removal without problematic byproduct formation.

Yet a persistent problem has limited real-world deployment: knowing exactly how to run the process. Electro-Fenton performance depends on a delicate interplay of variables, including the concentration of ferrous catalyst, the applied current density, the initial pollutant load, pH, electrode material, and treatment time. Push any one of these too far and the economics collapse. Excess iron generates sludge that must be disposed of; excessive current density wastes electricity in side reactions and energy losses; overdosing catalysts drives up chemical costs. Previous optimization efforts, often built on response surface methodology paired with desirability functions or on standalone metaheuristic algorithms, have tended to identify a single static operating point that maximizes degradation but says little about what that performance costs. For treatment plant operators, that blind spot is critical, because they must reconcile the goal of destroying as much antibiotic as possible with the pragmatic requirement of keeping the price per milligram of pollutant removed within a defensible budget.

A new study published in Clean Technologies and Environmental Policy tackles that trade-off head-on. Alijaeh Joshua A. Go and Angelo Earvin Sy Choi of De La Salle University in Manila developed a fuzzy multi-objective optimization framework designed specifically for electro-Fenton treatment of norfloxacin-contaminated wastewater. Rather than hunting for one best point on the assumption that only degradation matters, the framework treats degradation velocity and operating cost as competing objectives whose relative importance can be expressed through membership functions, the mathematical backbone of fuzzy logic. These functions quantify, in graded rather than binary terms, how satisfied a decision-maker is with a given outcome, allowing the optimization to seek the compromise that best reflects real priorities rather than an abstract mathematical extreme.

To ground their framework in solid experimental data, the researchers worked with the Box-Behnken design data set generated by Larralde-Pina and colleagues, whose 2023 study optimized an electro-Fenton pretreatment for degrading a mixture of ofloxacin, norfloxacin, and ciprofloxacin. The Box-Behnken design, a classic three-level experimental design introduced by Box and Behnken in 1960, allows researchers to model curved response surfaces efficiently with relatively few experimental runs, making it a popular foundation for regression-based process models. Go and Choi layered a parametric analysis on top of this model and then generated a Pareto frontier using the epsilon-constraint method, a technique that systematically converts a multi-objective problem into a sequence of constrained single-objective problems. The Pareto frontier maps out the full range of non-dominated solutions, those where no improvement in degradation can be achieved without increasing cost, and vice versa, giving engineers a complete picture of the available trade-offs rather than a single recommendation.

The fuzzy layer then does something that neither response surface desirability functions nor standalone metaheuristics can do as transparently: it lets decision-maker preferences enter the calculation directly. Membership functions encode how fully each objective is satisfied at any candidate operating point, and the solution that maximizes the overall degree of satisfaction is selected as the optimal compromise. The result is not merely a numerical answer but a defensible, interpretable one, which matters enormously when wastewater characteristics shift from day to day and when treatment objectives conflict across stakeholders such as regulators, utility managers, and the public.

Applied to the norfloxacin degradation data, the framework converged on a set of operating parameters that tells a striking economic story. The optimal compromise called for a ferrous ion concentration of 0.50 millimolar, a current density of 107.47 milliamperes per square centimeter, and an initial fluoroquinolone concentration of 90.00 milligrams per liter. At these settings the model predicted a norfloxacin degradation velocity of 0.0940 per minute at a total operating cost of 0.1870 US dollars per milligram of fluoroquinolone degraded. Compared with the previously reported single-objective optimum, this compromise delivered a 1.84 percent improvement in degradation performance while cutting the operating cost by 51.03 percent. In other words, by accepting a marginal, statistically modest gain in speed of antibiotic destruction, operators can halve the running cost of the process, a trade-off that single-objective optimization was structurally incapable of revealing.

The authors argue that the implications extend well beyond norfloxacin. Because the framework is built around the general structure of Box-Behnken response models and standard electro-Fenton economics, it can be generalized to other advanced oxidation processes, other pollutants, and other experimental data sets without redesigning the underlying machinery. The fuzzy approach also aligns naturally with a broader trend in environmental engineering, in which artificial intelligence and machine learning tools are being used to model nonlinear process behavior, optimize full-scale treatment plants, and support decision-making in increasingly complex sustainable infrastructure projects. Recent reviews have chronicled rapid progress in applying such computational methods to Fenton-based chemistry, heterogeneous catalysts, and pharmaceutical wastewater treatment, and the fuzzy multi-objective framework fits squarely within that movement while offering something distinct: an explicit, auditable way to encode human priorities.

For the water sector, the timing is significant. Regulators worldwide are beginning to scrutinize pharmaceutical residues in effluents, and utilities face rising energy and chemical costs that make any halving of operating expenses consequential. A technology that can reliably destroy antibiotics before they reach the environment, at a cost operators can justify, addresses both the technical and the economic barriers that have kept advanced oxidation processes largely confined to pilot studies. The Manila team’s demonstration that fuzzy optimization can convert an efficient but expensive lab-scale process into a considerably cheaper one suggests a practical pathway from bench to treatment basin.

There remain, of course, the familiar challenges of scale-up. Real wastewater carries suspended solids, competing organic matter, and variable salinity that can interfere with radical chemistry and iron cycling, and the study’s cost model reflects laboratory-scale assumptions. The authors acknowledge that enquiries about data availability should be directed to the authors, and they frame their contribution as a generalizable design framework rather than a turnkey plant specification. Even so, the central finding stands: when the objectives of clean water and affordable treatment are allowed to negotiate through fuzzy logic rather than compete in isolation, both sides win. As antimicrobial resistance tightens its grip on global health, tools that make sophisticated oxidation chemistry economically viable may prove to be among the most quietly transformative technologies of the coming decade in environmental engineering.

Subject of Research: Fuzzy multi-objective optimization of the electro-Fenton process for cost-effective norfloxacin antibiotic degradation in wastewater treatment

Article Title: Fuzzy optimization of electro-Fenton process for norfloxacin degradation in wastewater treatment

Article References: Go, A. J. A., & Choi, A. E. S. (2026). Fuzzy optimization of electro-Fenton process for norfloxacin degradation in wastewater treatment. Clean Technologies and Environmental Policy, 28(10), Article 257. https://doi.org/10.1007/s10098-026-03611-8

Image Credits: AI Generated

DOI: 10.1007/s10098-026-03611-8

Keywords: electro-Fenton process, norfloxacin degradation, fuzzy optimization, wastewater treatment, advanced oxidation processes, antimicrobial resistance, fluoroquinolone antibiotics, multi-objective optimization, Pareto frontier, Box-Behnken design, operating cost reduction, pharmaceutical micropollutants