Neural network computing systems for predicting vehicle requests
Embodiments are described for minimizing a wait time for a rider after sending a ride request for a vehicle. An example computer-implemented method includes receiving a ride request, the request being for travel from a starting location to a zone in a geographic region during a specified timeslot. The method further includes predicting travel demand based on a number of ride requests in the zone during the specified timeslot. The method further includes requesting transport of one or more vehicles to the zone in response to the predicted number of ride requests when the travel demand is predicted to exceed a number of vehicles in the zone during the specified timeslot.
1. A system comprising:
a memory; and
a processor operably coupled to the memory, the processor configured to:
receive a ride request, the request being for travel from a starting location to a zone in a geographic region during a specified timeslot;
predict travel demand based on a number of ride requests to the zone during the specified timeslot, wherein the travel demand is predicted based on a combination of a first prediction from a 3D convolutional neural network (CNN) generated using a first input, and a second prediction from a recurrent neural network (RNN) generated using a second input, the combination of predictions being an average of the first prediction and the second prediction; and
request transport of one or more vehicles to the zone in response to the predicted number of ride requests when the travel demand is predicted to exceed a number of vehicles in the zone during the specified timeslot; wherein:
the 3D CNN is trained using travel analysis zone (TAZ) timeslot-cubes, wherein a TAZ timeslot cube includes a plurality of matrices, each matrix associated with a predetermined factor, and wherein a first matrix associated with a first predetermined factor includes values of the first predetermined factor for each of the timeslots at each zone of the geographic region; and
the recurrent neural network (RNN) is trained using travel analysis zone (TAZ) factor-cubes, wherein a TAZ factor-cube includes a plurality of matrices, and wherein a matrix includes values of a corresponding traffic parameter at each zone during a specific timeslot.
2. The system of claim 1 , wherein the processor trains the 3D convolutional neural network (CNN), and the recurrent neural network (RNN).
3. A computer program product for minimizing wait times of riders after sending a ride request for a vehicle, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:
receive a ride request, the request being for travel from a starting location to a zone in a geographic region during a specified timeslot;
predict travel demand based on a number of ride requests to the zone during the specified timeslot, wherein the travel demand is predicted based on a combination of a first prediction from a convolutional neural network (CNN) generated using a first input, and a second prediction from a recurrent neural network (RNN) generated using a second input, the combination of predictions being an average of the first prediction and the second prediction; and
request transport of one or more vehicles to the zone in response to the predicted number of ride requests when the travel demand is predicted to exceed a number of vehicles in the zone during the specified timeslot; wherein:
the 3D CNN is trained using travel analysis zone (TAZ) timeslot-cubes, wherein a TAZ timeslot cube includes a plurality of matrices, each matrix associated with a predetermined factor, and wherein a first matrix associated with a first predetermined factor includes values of the first predetermined factor for each of the timeslots at each zone of the geographic region; and
the recurrent neural network (RNN) is trained using travel analysis zone (TAZ) factor-cubes, wherein a TAZ factor-cube includes a plurality of matrices, and wherein a matrix includes values of a corresponding traffic parameter at each zone during a specific timeslot.