System and method of determining freight/load distribution for multiple vehicles

• US 9,235,817 B2
• Filed: 01/12/2012
• Issued: 01/12/2016
• Est. Priority Date: 01/12/2011
• Status: Active Grant

• Current Assignee: Cummins Intellectual Property, Inc. (Cummins Incorporated)
• Original Assignee: Cummins Intellectual Property, Inc. (Cummins Incorporated)
• Inventors: Sujan, Vivek A., Vajapeyazula, Phani, Follen, Kenneth, Wu, An, Moffett, Barty L.
• Primary Examiner(s): Chen, George
• Application Number:
  US13/349,557
• Publication Number:
  US 20120239588A1
• Time in Patent Office: 1,461 Days
• Field of Search: 705/338
• US Class Current: 1/1
• CPC Class Codes: F02D 2200/0625 Fuel consumption, e.g. measured in fuel liters per 100 kms or miles per gallon
• CPC Class Codes: F02D 41/1406 with use of a optimisation method, e.g. iteration
• CPC Class Codes: G06Q 10/047 Optimisation of routes or paths, e.g. travelling salesman problem
• CPC Class Codes: G06Q 10/083 Shipping
• CPC Class Codes: G06Q 10/08355 Routing methods
• CPC Class Codes: G06Q 50/40 Business processes related to the transportation industry shipping G06Q10/083

Abstract

Systems and methods of vehicle freight/load distribution are provided to assist in determining optimal freight distribution. Although it is standard practice to fill each vehicle to its maximum limit, due to the non-linear nature of engine fueling maps (that is, fueling maps vary non-linearly as a function of torque and speed), the optimal distribution may not be obtained by the standard practice. Finding a solution for the optimal freight distribution may also need to account for the cost of fueling and operator costs, particularly if the situation involves multiple vehicles not filled to capacity. The benefit is increased freight efficiency in transporting cargo from source to destination amongst a fleet of vehicles.

Claims

1. A method for determining freight distribution amongst multiple vehicles, comprising:

accessing, by a processor, operator input data indicative of at least one operator input, said at least one operator input including at least one of route beginning and ending points to define a route, number of vehicles, maximum vehicle freight weight for each of a plurality of vehicles, and total freight weight;

accessing, by the processor, predetermined vehicle parameter data for each of said plurality of vehicles indicative of at least one predetermined vehicle parameter, said at least one predetermined vehicle parameter including at least one of vehicle mass, vehicle drag, vehicle rolling resistance, tire circumference, front area of vehicle, powertrain torque loss, vehicle tank capacity, and engine efficiency;

accessing, by the processor, road terrain data indicative of at least one road terrain element, said at least one road terrain element including at least one of speed limit changes, road grade, air density, position, elevation, and traffic patterns;

determining, by the processor, a first plurality of freight distribution solutions based on said maximum vehicle freight weight for each of said vehicles and said total freight weight, each solution comprising a random distribution of freight amongst the plurality of vehicles, the first plurality of freight distribution solutions comprising an initial generation;

evaluating, by the processor, each of said first plurality of freight distribution solutions of the initial generation for fitness based on fuel economy and ranking each solution based on the evaluation;

determining, by the processor, a second plurality of freight distribution solutions based on said maximum vehicle freight weight for each of said vehicles and said total freight weight, each solution comprising a distribution of freight amongst the plurality of vehicles according to a genetic algorithm structured to selectively execute a reproduction process first, a crossover process second, and a mutation process third, wherein the second plurality of freight distribution solutions includes a next generation;

generating, by the processor, a first random number associated with the crossover process and a second random number associated with the mutation process;

executing, by the processor, the crossover process responsive to the first random number being less than a crossover probability;

executing, by the processor, the mutation process responsive to the second random number being less than a mutation probability;

evaluating, by the processor, each of said freight distribution solutions of the next generation for fitness based on fuel economy, ranking each solution based on the evaluation, and assessing solution convergence based on the ranking;

determining, by the processor, an optimal freight distribution based on said convergence assessment; and

generating, by the processor, an electronic recommendation signal corresponding to said determined optimal freight distribution and communicating said recommendation signal to a receiver.

2. The method of claim 1, wherein said receiver comprises a display adapted to indicate the optimal freight distribution to an operator and indicate a measure of freight efficiency improvement.

3. The method of claim 2, wherein said display comprises a touch-screen interface.

4. The method of claim 1, wherein said solution convergence is assessed based on a fitness threshold.

5. The method of claim 1, wherein each of said crossover probability and mutation probability comprises a plurality of probabilities associated with each of the crossover and mutation processes respectively, said probabilities being adjusted based on said convergence assessment.

6. The method of claim 1, wherein said at least one operator input further includes a number representing a predetermined quantity of freight distribution solutions.

7. The method of claim 1, further comprising determining route grade and speed profile vectors for the defined route based on said operator input data, said predetermined vehicle parameter data, and said road terrain data.

8. The method of claim 1, wherein evaluating for fitness is based on an evaluation of engine speed and torque values yielding said engine fueling demand for a given vehicle power, said power being based on route grade and speed profile vectors for the defined route, said route grade and speed vectors being based on said operator input data, said predetermined vehicle parameter data, and said road terrain data.

9. The method of claim 8, wherein a plurality of said engine speed and torque values can collectively form an engine fuel map, said map being employed to establish an engine fueling table.

10. A system adapted to determine freight distribution amongst multiple vehicles, comprising:

an operator input module containing data indicative of at least one operator input, said at least one operator input including at least one of fuel cost, trip time, route beginning and ending points to define a route, and maximum vehicle speed;

a vehicle parameter module containing data indicative of at least one predetermined vehicle parameter, said at least one predetermined vehicle parameter including at least one of vehicle mass, vehicle drag, vehicle rolling resistance, tire circumference, front area of vehicle, powertrain torque loss, vehicle tank capacity, and engine efficiency;

a road terrain element module containing data indicative of at least one road terrain element, said at least one road terrain element including at least one of speed limit changes, road grade, air density, position, elevation, and traffic patterns; and

a vehicle load distribution module adapted to;

determine a first and second plurality of freight distribution solutions based on said maximum vehicle freight weight for each of said vehicles and said total freight weight, each solution from said first plurality comprising a random distribution of freight amongst the plurality of vehicles, the first plurality of freight distribution solutions comprising an initial generation, each solution from said second plurality comprising a distribution of freight amongst the plurality of vehicles according to a genetic algorithm structured to selectively execute a reproduction process first, a crossover process second, and a mutation process third, the second plurality of freight distribution solutions comprising a next generation;

generate a first random number associated with the crossover process and a second random number associated with the mutation process;

execute the crossover process responsive to the first random number being less than a crossover probability;

execute the mutation process responsive to the second random number being less than a mutation probability;

evaluate each of said first and second freight distribution solutions for fitness based on fuel economy and ranking each solution based on the evaluation, said fitness module being further adapted to assess solution convergence based on said ranking and determine an optimal freight distribution based on said convergence assessment; and

generate an electronic recommendation signal corresponding to said determined optimal freight distribution and communicate said recommendation signal to a receiver.

11. The system of claim 10, wherein said receiver comprises a display adapted to indicate the optimal freight distribution to an operator and indicate a measure of freight efficiency improvement.

12. The system of claim 11, wherein said display comprises a touch-screen interface.

13. The system of claim 10, wherein said fitness evaluation is based on an evaluation of engine speed and torque values yielding said engine fueling demand for a given vehicle power, said power being based on route grade and speed profile vectors for the defined route, said route grade and speed vectors being based on said operator input data, said predetermined vehicle parameter data, and said road terrain data.

14. The system of claim 13, wherein a plurality of said engine speed and torque values can collectively form an engine fuel map, said map being employed to establish an engine fueling table.

15. The system of claim 10, wherein said solution convergence is assessed based on a fitness threshold.

16. The system of claim 10, wherein each of said crossover probability and mutation probability comprises a plurality of probabilities associated with each of the crossover and mutation processes respectively, said probabilities being adjusted based on said convergence assessment.

17. The system of claim 10, wherein said vehicle load distribution module is further adapted to determine route grade and speed profile vectors for the defined route based on said operator input data, said predetermined vehicle parameter data, and said road terrain data.

18. A tangible, non-transitory computer program product comprising a computer useable medium having stored thereon computer-readable instructions for determining freight distribution amongst multiple vehicles, the computer-readable instructions being executable by a processor to perform operations comprising:

accessing data indicative of at least one operator input, said at least one operator input including at least one of route beginning and ending points to define a route, number of vehicles, maximum vehicle freight weight for each of a plurality of vehicles, and total freight weight;

accessing data indicative of at least one predetermined vehicle parameter, said at least one predetermined vehicle parameter including at least one of vehicle mass, vehicle drag, vehicle rolling resistance, tire circumference, front area of vehicle, powertrain torque loss, vehicle tank capacity, and engine efficiency;

accessing data indicative of at least one road terrain element, said at least one road terrain element including at least one of speed limit changes, road grade, air density, position, elevation, and traffic patterns;

determining a first plurality of freight distribution solutions based on said maximum vehicle freight weight for each of said vehicles and said total freight weight, each solution comprising a random distribution of freight amongst the plurality of vehicles, the first plurality of freight distribution solutions comprising an initial generation;

evaluating each of said first plurality of freight distribution solutions of the initial generation for fitness based on fuel economy and ranking each solution based on the evaluation;

determining a second plurality of freight distribution solutions based on said maximum vehicle freight weight for each of said vehicles and said total freight weight, each solution comprising a distribution of freight amongst the plurality of vehicles according to a genetic algorithm structured to selectively execute a reproduction process first, a crossover process second, and a mutation process third, the second plurality of freight distribution solutions comprising a next generation;

generating a first random number associated with the crossover process and a second random number associated with the mutation process;

executing the crossover process responsive to the first random number being less than a crossover probability;

executing the mutation process responsive to the second random number being less than a mutation probability;

evaluating each of said freight distribution solutions of the next generation for fitness based on fuel economy, ranking each solution based on the evaluation, and assessing solution convergence based on the ranking;

determining an optimal freight distribution based on said convergence assessment; and

communicating said optimal freight distribution to a display.