Grate and curb-opening inlets are part of pavement drainage structure function to discharge excess stormwater runoff on the street. One of the main challenges of these inlets is to effectively discharge the runoff, part of which if not properly controlled would lead to surface ponding on the street and eventually floods. Flood occurrences are mostly governed by hydrological and hydraulics parameters that are highly associated with uncertainties. Uncertainties are expected to contribute from the spreads of water moving on the street, which define the head computation over the inlets. Thus, this paper aims at carrying out a reliability assessment of several types of grate and curb-opening inlets using data obtained from a full-scale half roadway model built in the laboratory. Measurement of parameters was translated into random variables to be incorporated into three newly developed limit state function models. The Monte Carlo simulation method was used to simulate the models and the final outcomes were in the form of probability of failure, Pf. Results showed that Pf gradually increases with the increase of approaching flows on the street. Sensitivity analysis carried out revealed that water depth is the most significant parameter that governs the performance of the inlets. Results obtained are significant, especially for better design optimization purposes. © 2021 American Society of Civil Engineers.