Predicting taxi utilization information

• US 9,424,515 B2
• Filed: 12/05/2012
• Issued: 08/23/2016
• Est. Priority Date: 12/05/2011
• Status: Active Grant

• Current Assignee: Fasterfare, LLC
• Original Assignee: Fasterfare, LLC
• Inventors: Atlas, Stephen Andrew
• Primary Examiner(s): Hill, Stanley K
• Assistant Examiner(s): FINK, THOMAS ANDREW
• Application Number:
  US13/706,331
• Publication Number:
  US 20130144831A1
• Time in Patent Office: 1,357 Days
• Field of Search: 706 5- 6, 706/341, 706/925
• US Class Current: 1/1
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 50/40 Business processes related to the transportation industry shipping G06Q10/083

Abstract

Techniques are described for automatically analyzing contingencies in information predicting taxi demand. This is to generate representative information regarding current or future taxi demand, and for using such generated representative taxi demand. Contingent demand information may be generated for a variety of types of useful measures of taxi demand rates, such as for projecting expected likelihood of finding a passenger at each of several road locations. Generated representative contingent taxi demand information may be used in various ways to assist taxi and livery service drivers plan optimal routes and schedules. The historical and/or recent contingent demand data may be used to generate the representative traffic flow information. This may include data readings from mobile data sensors in the one or more vehicles, data sensors in or near the roads, or aggregate data sources collected from one or more sensors or through publicly available data sets.

Claims

1. A method of predicting taxi utilization information, comprising:

storing in a database at least one of historical taxi utilization information and recent taxi utilization information;

receiving a request for current or future taxi utilization information;

identifying taxi aggregation categories based on the requested current or future taxi information;

applying the taxi aggregation categories using a computer processor, to at least one of the historical taxi utilization information and the recent taxi utilization information to generate summary information about data relevant to the aggregation categories; and

predicting the requested current or future taxi information based on generated summary information;

wherein the step of predicting comprises solving;

2. The method of claim 1, wherein the current or future taxi utilization information includes at least one of current or future taxi demand information and supply information.

3. The method of claim 1, further comprising providing taxi relevant conditions.

4. The method of claim 3, wherein the identifying step comprises identifying taxi aggregation categories relative to the taxi relevant conditions.

5. The method of claim 4, wherein the generated summary information includes information about the taxi aggregation categories relative to the taxi relevant conditions.

6. The method of claim 3, wherein the taxi relevant conditions comprise one or more of weather and event data.

7. The method of claim 1, wherein the step of predicting current or future taxi utilization information factors at least one of a) the expected number of pickups at a location and time, b) expected number of pickups at various locations regardless of time, c) expected number of pickups at various times regardless of location, and d) expected unmet demand based on predicted supply and predicted demand.

8. The method of claim 1, wherein the weighting function W comprises a weight placed on each type of historical or recent taxi utilization information based on a calculated relative historical accuracy value for each type of taxi utilization information for each aggregation category relative to all types of taxi utilization information included in the summary information.

9. The method of claim 1, wherein the average historical or recent taxi utilization information for an aggregation category is standardized so that multiple measures of historical or recent taxi utilization information are comparable.

10. A taxi utilization apparatus comprising:

a memory storing at least one of historical taxi utilization information and recent taxi utilization information;

a computer processor configured to receive a request for current or future taxi utilization information, generate aggregation categories based on the request, and apply the taxi aggregation categories to at least one of the historical taxi utilization information and the recent taxi utilization information to generate summary information about data relevant to the aggregation categories, and to predict the requested current or future taxi utilization information, wherein the prediction is made by solving;

11. The taxi utilization apparatus of claim 10, wherein the transmitter transmits the taxi demand information to a graphical display.

12. The taxi utilization apparatus of claim 11, wherein the graphical display is disposed in a taxi for viewing by a taxi driver.

13. The taxi utilization apparatus of claim 12, wherein the graphical display is a map illustrating locations for probable fares.

14. The taxi utilization apparatus of claim 11, wherein the graphical display is provided to a dispatcher to dispatch taxis to locations for probable fares.

15. The taxi utilization apparatus of claim 10, wherein the memory further stores taxi relevant conditions.

16. The taxi utilization apparatus of claim 15, wherein the taxi relevant conditions comprise at least one of weather and event data.

17. The taxi utilization apparatus of claim 10, wherein the weighting function W comprises a weight placed on each type of historical or recent taxi demand information based on a calculated relative historical accuracy value for each type of taxi utilization information for each aggregation category relative to all types of taxi utilization information included in the summary information.

18. The taxi utilization apparatus of claim 10, wherein the average historical or recent taxi utilization information for an aggregation category is standardized so that multiple measures of historical or recent taxi utilization information are comparable.

19. A method of predicting taxi utilization information, comprising:

storing in a database at least one of historical taxi utilization information and recent taxi utilization information;

receiving a request for current or future taxi utilization information from a requestor regarding probability of finding a fare in a certain location;

providing taxi relevant conditions, wherein the taxi relevant conditions include at least a schedule of taxi-relevant events;

selecting taxi aggregation categories based on (i) groupings of locations in a transportation grid and (ii) periods in time, and the taxi relevant conditions;

applying the selected taxi aggregation categories using a computer processor, to at least one of the historical taxi utilization information and the recent taxi utilization information to generate summary information about data relevant to the aggregation categories; and

predicting a current or future taxi information along a given route based on the generated summary information and informing the requestor'"'"'s choice of what route to take to maximize chance of finding a fare, wherein the predicting is a function of the types of historical or recent taxi utilization information included in the summary information for each of the aggregation categories, and wherein each type of historical or recent taxi utilization information included in the summary information is weighted based on the accuracy of each information type for each aggregation level across historical taxi utilization information.

20. The method of claim 19, wherein the transportation grid includes location described by latitude and longitude or by positions relative to roads.

21. The method of claim 19, wherein the periods in time include at least one of similar hours of the day, days of the week, and months of the year.

22. The method of claim 19, wherein the summary information includes an indication of correspondence of at least one of a taxi supply and a taxi demand at at least one of a taxi pickup and a taxi dropoff in the transportation grid.

23. A method of predicting taxi utilization information, comprising:

storing in a database at least one of historical taxi utilization information and recent taxi utilization information;

receiving a request for current or future taxi utilization information from a requestor including probability of finding a fare in a certain location;

providing taxi relevant conditions, wherein the taxi relevant conditions include at least a schedule of taxi-relevant events;

identifying taxi aggregation categories based on the requested current or future taxi information and the taxi relevant conditions;

applying the taxi aggregation categories using a computer processor, to at least one of the historical taxi utilization information and the recent taxi utilization information to generate summary information about data relevant to the aggregation categories; and

predicting the requested current or future taxi information based on the generated summary information and informing the requestor'"'"'s choice of what route to take to maximize chance of finding a fare, wherein the predicting is a function of the types of historical or recent taxi utilization information included in the summary information for each of the identified aggregation categories, and wherein each type of historical or recent taxi utilization information included in the summary information is weighted based on the accuracy of each information type for each aggregation level across historical taxi utilization information.

24. The method of claim 23, wherein each type of historical or recent taxi utilization information included in the summary information is weighted based on a calculated relative historical accuracy value for each type of taxi utilization information for each aggregation category relative to all types of taxi utilization information included in the summary information.

25. The method of claim 23, wherein the current or future taxi utilization information includes at least one of current or future taxi demand information and supply information.

26. The method of claim 23, wherein the taxi relevant conditions comprise one or more of weather and event data.

27. The method of claim 23, wherein the step of predicting current or future taxi utilization information factors at least one of a) the expected number of pickups at a location and time, b) expected number of pickups at various locations regardless of time, c) expected number of pickups at various times regardless of location, and d) expected unmet demand based on predicted supply and predicted demand.

28. A taxi utilization apparatus comprising:

a memory storing at least one of historical taxi utilization information and recent taxi utilization information;

a computer processor configured to receive a request for current or future taxi utilization information from a requestor including probability of finding a fare in a certain location, receive taxi relevant conditions wherein the taxi relevant conditions include at least a schedule of taxi-relevant events, generate aggregation categories based on the request and the taxi relevant conditions, and apply the taxi aggregation categories to at least one of the historical taxi utilization information and the recent taxi utilization information to generate summary information about data relevant to the aggregation categories, and to predict the requested current or future taxi utilization information and inform the requestor'"'"'s choice of what route to take to maximize chance of finding a fare, wherein the prediction is based on generated summary information, wherein the predicting is a function of the types of historical or recent taxi utilization information included in the summary information for each of the identified aggregation categories, and wherein each type of historical or recent taxi utilization information included in the summary information is weighted based on a calculated relative historical accuracy value for each type of taxi utilization information for each aggregation category relative to all types of taxi utilization information included in the summary information; and

a transmitter for conveying to a user the requested taxi utilization information.