Optimizing wastewater treatment through real-time control (RTC) of plant operation can aid in reducing the volume of untreated combined sewer overflow discharged into receiving waters. Therefore, adequate lead-time prediction of wastewater inflow is critical to successful RTC of different unit processes in a WWTP. This study presents an innovative forecasting tool that utilizes the capabilities of artificial neural networks (ANNs) for real-time predictions of wastewater inflow. A number of ANN models were developed to predict Gold Bar WWTP inflow 1-hr in advance, using inputs from rain gauges, flow monitors, and temperature probes. The validation results of this case study have shown that the ANN modeling tool is VDWLVIDFWRU\ WKH RQVHW RI § DQG RI WKH UDLQ DQG VQRZ PHOWLQJ HYHQWV ZHUH FRUUectly captured by the models, respectively). The developed early warning system (EWS) was designed to accommodate equipment malfunctions by implementing 33 modelling algorithms within the EWS.