Systems and methods for estimating time of arrival of vehicle systems

A computer-implemented method, comprising using one or more processors disposed onboard a vehicle system and coupled to on-board sensors to perform operations comprising:

• obtaining a transportation event and a transportation event time of a vehicle system at a current location of the vehicle system on a route from an origin to a destination, wherein obtaining the transportation event comprises receiving, from a GPS receiver and one or more speed or remote sensors provided on the vehicle system, real-time location and speed information of the vehicle system and receiving, via electronic network communications, operational transportation trip-event data associated with the vehicle system;
• determining transportation event conditions based on historical transportation data;
• predicting transportation routes based on one or more of historical transportation routes, contractual routes, contractual junctions, or station master data;
• generating, from historical transportation data associated with the historical transportation routes, a plurality of historical transportation context vectors corresponding to historical transportation route legs;
• clustering transportation event data clusters from the historical transportation data using a machine learning classification method;
• generating, by the one or more processors and based at least in part on the historical transportation data and the transportation event, a multidimensional transportation context vector stored in a computer-readable data structure and corresponding to a current transportation route leg of the vehicle system, the multidimensional transportation context vector comprising numerical feature values derived from traffic flows, speed, distance, and location attributes for the current transportation route leg;
• matching the transportation event data clusters to historical transportation clusters at the current location by comparing the multidimensional transportation context vector to the plurality of historical transportation context vectors for transportation route legs that correspond to the current transportation route leg;
• predicting an estimated time of arrival (ETA) of the vehicle system to the destination based at least in part on the matched transportation event data clusters using a machine learning model executed by one of the one or more processors;
• and
• transmitting the predicted ETA to a central dispatch or enterprise resource planning system configured to update routing instructions for the vehicle system based on the predicted ETA;
• wherein the transportation event comprises one or more transportation event attributes including one or more of traffic of other vehicle systems on the route, traffic of other vehicle systems at one or more of the origin and the destination, a distance of the vehicle system from one or more of the origin or the destination, the current location of the vehicle system, a speed of the vehicle system, or weather conditions.