Effective water system management depends upon knowledge of the current state of a water pipeline system network. For example, in many cases, partial blockages in a water pipeline system are a source of inefficiencies, and result in an increase of pumping costs. These anomalies must be detected and corrected as early as possible. In this study, an algorithm is developed for detecting blockages by means of pressure transient measurements and estimating the diameter distribution resulting from their formation. The algorithm is a stochastic successive linear estimator that provides statistically the best unbiased estimate of diameter distribution due to partial blockages and quantifies the uncertainty associated with these estimates. We first present the theoretical formulation of the algorithm and then test it with a numerical case study.