This study aimed to determine the volume of institutional wastewater generated on a university campus for better wastewater management and reuse purposes. The study also involved the development of a predictive model to forecast the volumes of wastewater to be generated at future dates using the Artificial Neural Network (ANN). Data on the volume of wastewater was collected over 81 days by measuring the institution’s wastewater at the final exit point. Levenberg Marquardt and Bayesian Regularization algorithms were used to train the dataset, using a 9-15-1 structure for both algorithms. The dataset from 50 days was used to train the algorithms, while the dataset from 20 days was used for model validation. The remaining dataset from the last 11 days was used to perform an external test. The Bayesian Regularization algorithm performed better at predicting wastewater volumes with an accuracy of 95%, outperforming Levenberg Marquardt’s algorithm with 91% accuracy. Additionally, the study proposed a three-phase systematic approach for planning a wastewater reuse project. The phases comprise the preliminary, planning, and execution phases. Planners can use the findings from this research to manage wastewater treatment plants that receive more wastewater volumes than their design capacity.