In order to meet the growing water demand, two sustainable options are available—desalination and wastewater recycle and reuse. The wastewater recycling and reuse concept, which is already prevalent in most of Europe and America, is currently in high demand in India with Swachh Bharat Mission and smart cities program of Government of India. The necessity to treat wastewater for recycling and reuse is slowly gaining grounds in industries, municipalities, and residential segments. The wastewater can be treated with different means—physical process like separation of suspended solids, chemical process like pH neutralization, and biological process to decompose organic substances. In pH neutralization of wastewater, the incoming untreated wastewater has a wide range of pH, and it is difficult to treat wastewater with such a high variability of pH. Neutralization is the process used for adjusting pH to optimize treatment efficiency. Neutralization may take place in a holding, rapid mix, or an equalization tank. In this work, a novel approach using NARX (Nonlinear Auto-Regressive eXogenous) model and enhanced moth flame optimization (EMFO) is proposed for pH neutralization of the wastewater and its performance is compared with ARX (Auto-Regressive eXogenous), ARMAX (Autoregressive Moving Average eXogenous), and BJ’s (Box-Jenkins) model performance in terms of ISE, IAE, MSE, settling time, and peak overshoot for evaluating controller performance. The outcomes are then evaluated for both proposed and existing approaches to prove the effectiveness of the proposed scheme for the case under study. © 2021, Springer Nature Singapore Pte Ltd.