Methods, systems and processor-readable media for optimizing intelligent transportation system strategies utilizing systematic genetic algorithms

• US 9,183,742 B2
• Filed: 10/26/2012
• Issued: 11/10/2015
• Est. Priority Date: 10/26/2012
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Zou, Guangyu, Kulkarni, Rakesh, Yang, Ming
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): Zidanic, Michael
• Application Number:
  US13/661,856
• Publication Number:
  US 20140122032A1
• Time in Patent Office: 1,110 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming
• CPC Class Codes: G06N 5/00 Computing arrangements using knowledge-based models
• CPC Class Codes: G08G 1/0112 from the vehicle, e.g. floating car data [FCD]
• CPC Class Codes: G08G 1/0141 for traffic information dissemination
• CPC Class Codes: G08G 1/096811 where the route is computed offboard
• CPC Class Codes: G08G 1/096844 where the complete route is dynamically recomputed based on new data

Abstract

Methods, systems and processor-readable media for modeling and optimizing multiple ITS (Intelligent Transportation System) strategies utilizing a systematic genetic algorithm. A traffic simulation model can be configured in conjunction with a genetic algorithm based optimization engine for optimizing the transportation models. An origin-destination matrix that minimizes discrepancies between a simulated and an observed link traffic count can be estimated by considering a road network and a traffic count with respect to a region. A driver behavior can then be determined utilizing the origin-destination matrix via calibration so that the simulation model can replicate a freeway traffic flow in the region. An optimal parameter with respect to the ITS strategies can be determined to optimize a set goal with respect to a given constraint. Such an approach meets a level of service (LOS) metric as well as a revenue target under the applied ITS strategies.

Claims

1. A computer-implemented method for optimizing a multiple intelligent transportation strategies system, said method comprising:

associating a traffic simulation model with a genetic algorithm based optimization engine by processing with a computing device said traffic simulation model and said genetic algorithm based optimizing engine in order to optimize a plurality of intelligent transportation strategies utilizing said traffic simulation model;

estimating an origin-destination matrix that minimizes discrepancies between a simulated and an observed link traffic count by considering via a computing device, a road network and a traffic count with respect to a region;

determining a driver behavior parameter via a computing device utilizing said origin-destination matrix via calibration so that said traffic simulation model replicates a freeway traffic flow in said region; and

obtaining an optimal set of parameters comprising a pricing algorithm parameter, a ramp meter mechanism and a speed limit with respect to said plurality of intelligent transportation strategies to optimize a set goal with respect to a given constraint and to meet a level of service metric as well as a revenue target under said plurality of intelligent transportation strategies.

2. The method of claim 1 wherein said driver behavior parameter comprises at least the following parameters:

a mean target headway;

a mean reaction time; and

a time step.

3. The method of claim 2 further comprising:

combining a genetic algorithm with a simplex approach by designating an objective function and encoding a set of configuration parameters with respect to a gene;

computing a fitness for each gene and varying said gene utilizing a genetic operator;

selecting genes with higher fitness as a next generation in terms of distance between an observed metric and a simulated metric; and

terminating an iterative process if a maximum number of iteration times is attained and/or said gene with a fitness greater than a predefined threshold emerges.

4. The method of claim 3 further comprising initially setting said driver behavior parameter to be optimized for a particular stage at a certain value and batch-run said traffic simulation model to output a simulated metric.

5. The method of claim 3 further comprising:

comparing a simulated metric with a desired metric to determine a fitness of said gene and feedback a new parameter value obtained from said optimization engine into said traffic simulation model; and

repeating said process to obtain an optimized parameter for a particular stage wherein an output of said particular stage is employed as an input for a next stage.

6. The method of claim 3 further comprising maximizing revenue and throughput obtained from a toll charged for a high occupancy toll lane based on a pricing algorithm and a ramp metering mechanism without loss of generality.

7. A system for optimizing a multiple intelligent transportation strategies system, said system comprising:

a processor;

a data bus coupled to said processor; and

a computer-usable medium embodying computer program code, said computer-usable medium being coupled to said data bus, said computer program code comprising instructions executable by said processor and configured for;

associating a traffic simulation model with a genetic algorithm based optimization engine for optimizing a plurality of intelligent transportation strategies;

estimating an origin-destination matrix that minimizes discrepancies between a simulated and an observed link traffic count by considering a road network and a traffic count with respect to a region;

determining a driver behavior parameter utilizing said origin-destination matrix via calibration so that said traffic simulation model replicates a freeway traffic flow in said region; and

obtaining an optimal set of parameters comprising a pricing algorithm parameter, a ramp meter mechanism and a speed limit with respect to said plurality of intelligent transportation strategies to optimize a set goal with respect to a given constraint and to meet a level of service metric as well as a revenue target under said plurality of intelligent transportation strategies.

8. The system of claim 7 wherein said driver behavior parameter comprises at least one of the following parameters:

a mean target headway;

a mean reaction time; and

a time step.

9. The system of claim 8 wherein said instructions are further configured for:

combining a genetic algorithm with a simplex approach by designating an objective function and encoding a set of configuration parameters with respect to a gene;

computing a fitness for each gene and varying said gene utilizing a genetic operator;

selecting genes with higher fitness as a next generation in terms of distance between an observed metric and a simulated metric; and

terminating an iterative process if a maximum number of iteration times is attained and/or said gene with a fitness greater than a predefined threshold emerges.

10. The system of claim 9 wherein said instructions are further configured for initially setting said parameter to be optimized for a particular stage at a certain value and batch-run said simulation model to output said simulated metric.

11. The system of claim 9 wherein said instructions are further configured for:

comparing said simulated metric with a desired metric to determine a fitness of said gene and feed back a new parameter value obtained from said optimization engine into said simulation model; and

repeating said process to obtain said optimized parameter for that stage wherein an output of one stage is employed as an input for a next stage.

12. The system of claim 9 wherein said instructions are further configured for maximizing revenue and throughput obtained from a toll charged for a high occupancy toll lane based on a pricing algorithm and a ramp metering mechanism without loss of generality.

13. A non-transitory processor-readable medium storing code representing instructions to cause a process for optimizing a multiple intelligent transportation strategies system, said code comprising code to:

associate a traffic simulation model with a genetic algorithm based optimization engine for optimizing a plurality of intelligent transportation strategies;

estimate an origin-destination matrix that minimizes discrepancies between a simulated and an observed link traffic count by considering a road network and a traffic count with respect to a region;

determine a driver behavior parameter utilizing said origin-destination matrix via calibration so that said traffic simulation model replicates a freeway traffic flow in said region; and

obtain an optimal set of parameters comprising a pricing algorithm parameter, a ramp meter mechanism and a speed limit with respect to said plurality of intelligent transportation strategies to optimize a set goal with respect to a given constraint and to meet a level of service metric as well as a revenue target under said plurality of intelligent transportation strategies.

14. The processor-readable medium of claim 13 wherein said driver behavior parameter comprises at least one of the following parameters:

a mean target headway;

a mean reaction time; and

a time step.

15. The processor-readable medium of claim 14 wherein said code further comprises code to:

combine a genetic algorithm with a simplex approach by designating an objective function and encoding a set of configuration parameters with respect to a gene;

compute a fitness for each gene and varying said gene utilizing a genetic operator;

select genes with higher fitness as a next generation in terms of distance between an observed metric and a simulated metric; and

terminate an iterative process if a maximum number of iteration times is attained and/or said gene with a fitness greater than a predefined threshold emerges.

16. The processor-readable medium of claim 15 wherein said code further comprises code to initially set said parameter to be optimized for a particular stage at a certain value and batch-run said simulation model to output said simulated metric.

17. The processor-readable medium of claim 15 wherein said code further comprises code to:

compare said simulated metric with a desired metric to determine a fitness of said gene and feedback a new parameter value obtained from said optimization engine into said simulation model; and

repeat said process to obtain said optimized parameter for that stage wherein an output of one stage is employed as an input for a next stage.