Abstract : We present a constraint programming formulation for the elevator trip origin-destination matrix estimation problem, and propose different approaches to solve the problem. An elevator trip consists of successive stops in one direction of travel with passengers inside the elevator. It can be defined as a directed network, where the nodes correspond to the stops on the trip, and the arcs to the possible origins and destinations of the passengers boarding and alighting at the stops. The goal is to estimate the count of passengers for the origin-destination pairs of every elevator trip occurring in a building. These counts are used to make passenger traffic forecasts which are needed in elevator dispatching to make robust dispatching decisions in constantly changing traffic conditions. The proposed approach can be used to estimate elevator trip origin-destination matrices within a real time limit, and provides a flexible method to solve a challenging real life problem.