While Automatic Identification System (AIS) data provides raw vessel positioning, it lacks the granular cargo tracking details necessary to identify the specific product and quantity being transported. This work presents a two-stage sequential optimization pipeline, triggered by real-time events, designed to reconstruct commodity flows by integrating vessel positioning, drafts, and port constraints. The first stage employs a Mixed-Integer Linear Program (MILP) to accurately determine the overall quantities exchanged during port visits. Subsequently, the second stage leverages Constraint Programming (CP) to enumerate all feasible product combinations, which are then aggregated to establish the final, specific product assignment for each vessel operation.