Method and system for predicting demand for vehicles

1. A method for predicting demand of one or more vehicles in a transportation network, said method comprising:

determining, by one or more microprocessors, one or more demand events at each of one or more locations for one or more time intervals based on a historical demand data associated with each of said one or more locations, wherein said one or more demand events correspond to a demand of said one or more vehicles at said one or more locations during a plurality of time intervals throughout a day based on a count of passengers at a location for entry onto the one or more vehicles, wherein the count of passengers of the location for entry onto the one or more vehicles is determined by one or more proximity sensors installed at an entry or exit point of said location for entry;

creating, by said one or more microprocessors, a graph comprising one or more nodes, and one or more edges connecting said one or more nodes, each node being representative of a demand event from said one or more demand events, wherein an edge is representative of a dependency between two demand events from said one or more demand events;

predicting, by said one or more microprocessors, said demand of said one or more vehicles at a location from said one or more locations during a predetermined time interval based on said created graph and a real time demand of said one or more vehicles associated with at least one other demand event;

instructing, by said one or more processors, the one or more vehicles to leave the transportation network based on the demand prediction;

displaying, by a display device, said demand prediction on a computing device at said one or more locations of said transportation network; and

physically removing one or more vehicles from the transportation network in response to the instructing of the one or more vehicles.

2. The method of claim 1 further comprising determining, by said one or more microprocessors, said dependency between said two demand events using at least a linear regression model.

3. The method of claim 2 , wherein said linear regression model comprises at least a Poisson distribution, lasso regression model, or penalized lasso regression.

4. The method of claim 1 , wherein said historical demand data comprises at least data pertaining to a sale of one or more tickets for said one or more vehicles at said one or more locations for said one or more time intervals.

5. The method of claim 1 , wherein said graph corresponds to at least a direct acyclic graph.

6. The method of claim 1 further comprising assigning, by said one or more microprocessors, weights to said one or more edges of said graph, wherein said weights on said one or more edges are indicative of a strength of said dependency.

7. The method of claim 6 , wherein said demand is predicted based on said weights assigned to each of said one or more edges in said graph.

8. A system for predicting demand of one or more vehicles in a transportation network, said system comprising:

one or more microprocessors configured to:

determine one or more demand events at each of one or more locations for one or more time intervals based on a historical demand data associated with each of said one or more locations, wherein said one or more demand events correspond to a demand of said one or more vehicles at said one or more locations during a plurality of time intervals throughout a day based on a count of passengers at a location for entry onto the one or more vehicles, wherein the count of passengers of the location for entry onto the one or more vehicles is determined by one or more proximity sensors installed at an entry or exit point of said location for entry;

create a graph comprising one or more nodes, and one or more edges connecting said one or more nodes, each node being representative of a demand event from said one or more demand events, wherein an edge is representative of a dependency between two demand events from said one or more demand events;

predict said demand of said one or more vehicles at a location from said one or more locations during a predetermined time interval based on said created graph and a real time demand of said one or more vehicles associated with at least one demand event;

instruct the one or more vehicles to leave the transportation network based on the demand prediction; and

a display device configured to display said demand prediction on a computing device at said one or more locations of said transportation network; and

one or more vehicles operated to be removed from the transportation network in response to the instruction of said one or more vehicles based on the demand prediction.

9. The system of claim 8 , wherein said one or more microprocessors are further configured to determine said dependency between said two demand events using at least a linear regression model.

10. The system of claim 9 , wherein said linear regression model comprises at least a Poisson distribution, lasso regression model, or penalized lasso regression.

11. The system of claim 8 , wherein said historical demand data comprises at least data pertaining to a sale of one or more tickets for said one or more vehicles at said one or more locations for said one or more time intervals.

12. The system of claim 8 , wherein said graph corresponds to at least a direct acyclic graph.

13. The system of claim 8 , wherein said one or more microprocessors are further configured to assign weights to said one or more edges of said graph, wherein said weights on said one or more edges are indicative of a strength of said dependency.

14. The system of claim 13 , wherein said demand is predicted based on said weights assigned to each of said one or more edges in said graph.