Modeling of an activated sludge process for effluent prediction—a comparative study using ANFIS and GLM regression Dauda Olurotimi Araromi & Olukayode Titus Majekodunmi & Jamiu Adetayo Adeniran & Taofeeq Olalekan Salawudeen Received: 29 January 2018 /Accepted: 25 July 2018 /Published online: 1 August 2018 # Springer Nature Switzerland AG 2018 Abstract In this paper, nonlinear system identification of the activated sludge process in an industrial waste- water treatment plant was completed using adaptive neuro-fuzzy inference system (ANFIS) and generalized linear model (GLM) regression. Predictive models of the effluent chemical and 5-day biochemical oxygen demands were developed from measured past inputs and outputs. From a set of candidates, least absolute shrinkage and selection operator (LASSO), and a fuzzy brute-force search were utilized in selecting the best combination of regressors for the GLMs and ANFIS models respectively. Root mean square error (RMSE) and Pearson’ s correlation coefficient (R-value) served as metrics in assessing the predicting performance of the models. Contrasted with the GLM predictions, the ob- tained modeling results show that the ANFIS models provide better predictions of the studied effluent vari- ables. The results of the empirical search for the domi- nant regressors indicate the models have an enormous potential in the estimation of the time lag before a desired effluent quality can be realized, and preempting process disturbances. Hence, the models can be used in developing a software tool that will facilitate the effec- tive management of the treatment operation. Keywords Wastewater treatment process modeling . Predictive models . ANFIS . Fuzzy exhaustive search . GLM regression . LASSO regularization Introduction Tightening environmental constraints have made it in- creasingly important for chemical process plants to be operated efficiently and in an environmentally friendly manner. Although several strategies have been put in place to combat the menace of water pollution; waste- water treatment still remains a major global challenge for process industries, as the performance of any waste- water treatment plant (WWTP) is affected by a distinct combination of some physical, chemical, and biological factors (Belanche et al. 2000; Mjalli et al. 2007; Singh et al. 2010). Important water quality parameters frequently used to assess the performance of WWTPs include 5-day biochemical oxygen demand (BOD), chemical oxygen demand (COD), and total suspended solids (TSS). Fre- quent monitoring of these parameters helps plant oper- ators preempt process disturbances and, hence, maintain environmental balance. This task often involves labori- ous and expensive laboratory analyses (Dupuit et al. 2007). Therefore, the determination of wastewater Environ Monit Assess (2018) 190: 495 https://doi.org/10.1007/s10661-018-6878-x D. O. Araromi : O. T. Majekodunmi (*) : J. A. Adeniran : T. O. Salawudeen Department of Chemical Engineering, Ladoke Akintola University of Technology, Ogbomoso P.M.B. 4000, Nigeria e-mail: olukayodemajekodunmi@iyte.edu.tr O. T. Majekodunmi Department of Chemical Engineering, Izmir Institute of Technology, 35430 Izmir, Turkey J. A. Adeniran Department of Chemical Engineering, University of Ilorin, Ilorin P.M.B. 1515, Nigeria