The model is formulated based on the Gaussian maximum likelihood method. The two key random variables considered in the model are flows and flow increments. By assuming that these two random variables are normally distributed, we generate an estimate by maximizing the likelihood of the flow level at the next time interval, expressed as a product of the two probability density functions. The model departs from other existing models in that the parameters of the model are obtained based little on data fitting. The model, therefore, is very robust in terms of calibration of its parameters.