CN112508235A - Bus departure time interval optimization method based on ant colony algorithm - Google Patents

Abstract

The invention discloses a bus departure time interval optimization method based on an ant colony algorithm. The bus departure time interval optimization is based on statistical analysis of bus-mounted data, bus route stop data and passenger IC card swiping data, the average speed of bus adjacent stops at each moment is obtained through analysis of the bus-mounted data, and the time and speed distribution rule of bus driving is obtained to represent road condition information of a road. The passenger IC card swiping data analysis is carried out to obtain the stop passenger flow volume distribution rule and the passenger travel rule of the stop, the passenger distribution rule and the passenger travel rule are used for representing the distribution information of passengers, the time of the bus reaching each stop and the number of passengers getting on and off are simulated through the road condition information and the passenger distribution information, and the ant colony algorithm is utilized to search and solve the bus departure time interval. According to the invention, by finding a group of departure time interval sequence, the bus arrival time interval and the bus full load rate can achieve the optimal effect.

Description

Bus departure time interval optimization method based on ant colony algorithm

Technical Field

The invention relates to a bus departure time interval optimization method, in particular to a bus departure time interval optimization method based on an ant colony algorithm.

Background

With the rapid development of economy, the living standard of people is continuously improved, the total population of cities is increased rapidly, the number of private vehicles is increased rapidly, and the problem of urban traffic jam is increasingly prominent. In order to advocate people's public transit trip, alleviate the city problem of blocking up, it is the key of solving the problem to improve public transit quality of service and passenger satisfaction. In the bus operation, the long waiting time of the passengers is the main reason of low satisfaction, so that how to dispatch the vehicles according to the passenger flow conditions in the bus operation process by the bus company to improve the satisfaction of the passengers is an important decision problem.

In conclusion, the invention designs the bus departure time interval optimization method based on the ant colony algorithm.

Disclosure of Invention

Aiming at the defects in the prior art, the invention aims to provide an ant colony algorithm-based bus departure time interval optimization method, which enables the bus arrival time interval and the bus full load rate to achieve the optimal effect by finding a group of departure time interval sequences.

In order to achieve the purpose, the invention is realized by the following technical scheme: an ant colony algorithm-based bus departure time interval optimization method comprises the following steps:

1. firstly, carrying out statistical analysis on bus-mounted data, bus route stop data and passenger IC card swiping data, and identifying the starting time and the ending time of each shift by using the principle that the vehicle course angle and the shift direction are consistent according to the GPS, the vehicle speed and the driving time of the bus-mounted data and the stop data of the bus route; calculating the distance between two adjacent stations, acquiring GPS track points of the two adjacent stations by means of a Gaode map OPENAPI technology, obtaining the distance between the stations by calculating the integral of the spherical distance of the two track points, and calculating the running time of the vehicle between the stations by analyzing the time of the vehicle reaching each station, wherein the running time is used for calculating the speed distribution of the vehicle between the stations at each moment; counting and analyzing the card swiping and boarding time of passengers at each station to obtain the passenger flow distribution condition of each station in each time period;

2. simulating each shift by utilizing a multithreading technology simulation on the basis of a given departure time interval according to the identified shift, the inter-site information result and the passenger distribution in different time periods among the sites, and counting the arrival time of the vehicle and the full load rate index of the vehicle;

3. by adjusting the departure time interval sequence, the times of the arrival time interval in a reasonable interval are more, and the average full load rate of the vehicle is larger, namely, the reasonable multi-objective optimization model with the maximum arrival time interval times and the larger average full load rate of the vehicle is solved.

4. And finally, searching and solving the bus departure time interval through an ant colony algorithm, outputting a departure interval sequence and constructing a departure timetable.

The solving step of the step 4 is as follows:

(1) dividing one day into 5 time periods of a flat peak period, an early peak period, a flat peak period, a late peak period and a flat peak period according to a passenger flow travel rule, wherein the departure time intervals in each time period are the same, for example, a departure time interval sequence [10,4,7,4,8,8] represents that the 1 st flat peak period is issued one shift every 10 minutes and serves as the initial position of ants;

(2) initializing the positions of M ants according to the rule to form a plurality of initial paths;

(3) initializing pheromones on each path;

(4) ants skip between each time interval, and add the selected value into the path of the solution;

(5) sequentially calculating the objective function value of each path, and selecting the optimal path to enter the next iteration;

(6) and when the iteration times reach the maximum value or the objective function value is not changed basically, terminating the iteration and outputting the departure interval sequence.

The bus departure time interval optimization is based on statistical analysis of bus-mounted data, bus route stop data and passenger IC card swiping data, the average speed of bus adjacent stops at each moment is obtained through analysis of the bus-mounted data, and the time and speed distribution rule of bus driving is obtained to represent road condition information of a road. The passenger IC card swiping data analysis is carried out to obtain the stop passenger flow volume distribution rule and the passenger travel rule of the stop, the passenger distribution rule and the passenger travel rule are used for representing the distribution information of passengers, the time of the bus reaching each stop and the number of passengers getting on and off are simulated through the road condition information and the passenger distribution information, and the ant colony algorithm is utilized to search and solve the bus departure time interval.

The invention has the beneficial effects that: according to the method, firstly, the distribution rule of station passengers is systematically analyzed, the average speed of the bus in each time interval is established, a multi-objective optimization model is established according to two indexes of the bus arrival time interval and the full load rate of the bus, and the multi-objective problem is solved by utilizing an ant colony algorithm. The essence is that the bus arrival time interval and the bus full load rate can achieve the optimal effect by finding a group of departure time interval sequences.

Drawings

The invention is described in detail below with reference to the drawings and the detailed description;

FIG. 1 is a flow chart of an optimization method of the present invention;

FIG. 2 is a flow chart of simulation of the present invention;

fig. 3 is a flowchart of the ant colony algorithm solving of the present invention.

Detailed Description

In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further described with the specific embodiments.

Referring to fig. 1, the following technical solutions are adopted in the present embodiment: an ant colony algorithm-based bus departure time interval optimization method comprises the following steps:

1. the bus-mounted data, the bus line stop data and the passenger IC card swiping data are subjected to data processing, and the data processing method specifically comprises the following steps:

1.1 class identification

1.11 according to the GPS of the bus data, the speed and the running time, and the station data of the bus line, the starting time and the ending time of each shift are identified by using the principle that the heading angle and the shift direction of the bus are consistent. The identification of the shift is based on the uplink and downlink conditions of the same bus on the same day of data analysis operation of the same bus, and the actual shift identification rule is as follows:

1.12 the current time is the first uplink and downlink state, and the current time is the start time of the shift;

1.13, if the time difference between the current data time and the previous data time is more than 1 hour, the current time is the start time of the shift;

1.14, if the current uplink and downlink state is different from the previous time point, the current time is the start time of the shift;

1.15, the current time is the end time of the shift when the current time is the last uplink and downlink state;

1.16, if the time difference between the current data time and the next data time is more than 1 hour, the current time is the end time of the shift;

1.17, if the current uplink and downlink state is different from the next time point, the current time is the end time of the shift;

the shift data structure is as follows:

[0001]name (R)	[0002]Name of field	[0003]Remarks for note
			[0004]Line	[0005] line_name	[0006]
[0007]Uplink and downlink	[0008] up_down	[0009]
			[0010]Vehicle ID	[0011] carno	[0012]
[0013]Number of shifts	[0014] schedule	[0015] yyyy-MM-dd hh:mm
			[0016]Departure time of starting station	[0017] start_time	[0018]
[0019]Time of receiving at terminal	[0020] end_time	[0021]
			[0022]Starting site name	[0023] start_station_name	[0024]
[0025]Terminating site name	[0026] end_station_name	[0027]
			[0028]Initial site number	[0029] start_station_seq	[0030]
[0031]Terminating site sequence number	[0032] end_station_seq	[0033]

1.2 inter-site information calculation

The inter-station information calculation refers to calculating the distance between two adjacent stations, acquiring GPS track points of the two adjacent stations by means of a Gauss map OPENAPI technology, obtaining the distance between the stations by calculating the integral of the spherical distance of the two track points, and calculating the running time of the vehicle between the stations by analyzing the time of the vehicle reaching each station, so as to calculate the speed distribution of the vehicle between the stations at each moment.

The inter-site information data structure is as follows:

[0034]name (R)	[0035]Name of field	[0036]Remarks for note
			[0037]Line	[0038] line_name	[0039]
[0040]Uplink and downlink	[0041] up_down	[0042]
			[0043]Vehicle ID	[0044] carno	[0045]
[0046]Number of shifts	[0047] schedule	[0048]
			[0049]The station name	[0050] from_station_name	[0051]
[0052]Starting time of the station	[0053] from_station_time	[0054]
			[0055]The departure date of the station	[0056] from_station_day	[0057]Such as 2018-09-01
[0058]The starting time of the station	[0059] from_station_time2	[0060]E.g. 16:48
			[0061]Name of station	[0062] to_station_name	[0063]
[0064]Arrival time of the station	[0065] to_station_time	[0066]
			[0067]Arrival date of arrival at the station	[0068] to_station_day	[0069]Such as 2018-09-01
[0070]Arrival time of the station	[0071] to_station_time2	[0072]E.g. 16:48
			[0073]Distance between two stations	[0074] distance	[0075]Unit: rice and its production process
[0076]When used in two stations	[0077] use_time	[0078]Unit: second of

1.3 passenger distribution at a stop

The passenger distribution among the stations is to statistically analyze the card swiping and boarding time of the passengers of each station to obtain the passenger flow distribution condition of each station in each time period.

The passenger distribution data structure between the stations is as follows

[0079]Name (R)	[0080]Name of field	[0081]Remarks for note
			[0082]Line	[0083] line_name	[0084]
[0085]Uplink and downlink	[0086] up_down	[0087]
			[0088]Vehicle ID	[0089] carno	[0090]
[0091]Number of shifts	[0092] schedule	[0093]
			[0094]Site name	[0095] station_name	[0096]
[0097]Site number	[0098] station_seq	[0099]
			[00100]Time of arrival	[00101] time_get_in	[00102]Such as 2018-09-0116: 48
[00103]Date of arrival	[00104] day_get_in	[00105]Such as 2018-09-01
			[00106]Time of arrival	[00107] time_get_in2	[00108]E.g. 16:48
[00109]Time of departure	[00110] time_get_out2	[00111]E.g. 17:02
			[00112]The number of persons getting on the bus	[00113] num_get_in	[00114]
[00115]The number of people getting off	[00116] num_get_off	[00117]
			[00118]Longitude (G)	[00119] jingdu	[00120]Local station longitude
[00121]Latitude	[00122] weidu	[00123]Latitude of the station

2. Simulation of

The simulation is to simulate each shift by utilizing a multithread technology simulation on the basis of a given departure time interval according to the identified shift, information results among stations and passenger distribution among stations in different time periods, and to count the arrival time of the vehicle and the full load rate index of the vehicle.

The simulation flow chart is shown in fig. 2.

3. Establishing a multi-objective optimization model

From the perspective of a public transportation enterprise, the larger the vehicle full load rate is, the better the vehicle full load rate is, the smaller the waiting time of passengers from the perspective of the passengers is, the better the passenger waiting time can be converted into the arrival time interval of the analysis vehicle, the more uniform the arrival time interval is, the smaller the average waiting time of the passengers is, therefore, the optimization of the departure time interval in the shift is essentially through adjusting the departure time interval sequence, so that the times of the arrival time interval in a reasonable interval are more and the average full load rate of the vehicle is larger, namely, through solving the multi-objective optimization model of the maximum times of the reasonable arrival time interval and the larger average full load rate of the vehicle. According to experience, the bus arrival time interval is reasonable in 4-10 minutes, the peak time is 4-6 minutes, the peak time is 6-10 minutes, and the arrival time interval distribution in the peak time and the peak time is calculated through simulation.

3.1 departure time interval optimization model

Representing the time when the ith vehicle reaches the kth station for the jth shift;

the number of passengers getting on the ith bus to the kth station in the jth shift;

the number of people getting off the ith vehicle from the jth station in the jth shift is shown;

m represents the total number of shifts;

c represents the maximum number of passengers of the vehicle;

k represents the station number of the line;

n represents the number of vehicles;

1) arrival time interval:

wherein the function

2) Average loading rate:

constructing a Multi-objective function

Where Q is the arrival time interval, S is the average loading rate, and α, β are coefficients, respectively.

4. Ant colony algorithm solution

The ant colony optimization algorithm is an intelligent optimization algorithm, solves a complex problem through ant colony optimization, and is a probability algorithm for searching an optimized path in a graph. Ants leave a kind of things called pheromone in the moving process, and with the moving distance, the spread pheromone is less and less, so that the concentration of the pheromone is strongest at home or around food, the ants can select the direction according to the pheromone, of course, the thicker the pheromone is, the greater the probability of selection is, and the pheromone has a certain volatilization function.

The ant colony algorithm is used for solving the TSP problem initially, and in a bus departure time interval optimization model, by combining the travel rule of the public traffic, the system is divided into a peak balancing period, a peak early period, a peak balancing period and a peak balancing period, wherein the early peak time is 7:00-9:00, the late peak time is 17:30-19:30, the departure time of the first bus is 6:00, the departure time of the last bus is 23:00, according to the above time-interval division, 5 time intervals are divided into 'cities' assumed by ants, ants randomly take one value as the destination of the city in each time interval division, so that initially, ants are randomly placed in each time interval, in each period, the ants randomly select a departure time interval value as the initial position of the ant, and the ant colony searches for food through collective intelligence of the ant colony, so that the ant colony can select a path with the optimal objective function at a high probability.

The ant colony algorithm solving step of the present embodiment:

1. dividing one day into 5 time periods of a flat peak period, an early peak period, a flat peak period, a late peak period and a flat peak period according to a passenger flow travel rule, wherein the departure time intervals in each time period are the same, for example, a departure time interval sequence [10,4,7,4,8,8] represents that the 1 st flat peak period is issued one shift every 10 minutes and serves as the initial position of ants;

2. initializing the positions of M ants according to the rule to form a plurality of initial paths;

3. initializing pheromones on each path;

4. ants skip between each time interval, and add the selected value into the path of the solution;

5. sequentially calculating the objective function value of each path, and selecting the optimal path to enter the next iteration;

6. when the iteration times reach the maximum value or the objective function value is not changed basically, terminating the iteration and outputting a departure interval sequence;

the ant colony algorithm solving flow chart is shown in fig. 3.

The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (2)

1. A bus departure time interval optimization method based on an ant colony algorithm is characterized by comprising the following steps:

(1) firstly, carrying out statistical analysis on bus-mounted data, bus route stop data and passenger IC card swiping data, and identifying the starting time and the ending time of each shift by using the principle that the vehicle course angle is consistent with the shift direction according to the GPS, the vehicle speed and the driving time of the bus-mounted data and the stop data of the bus route; calculating the distance between two adjacent stations, acquiring GPS track points of the two adjacent stations by means of a Gaode map OPENAPI technology, obtaining the distance between the stations by calculating the integral of the spherical distance of the two track points, and calculating the running time of the vehicle between the stations by analyzing the time of the vehicle reaching each station, wherein the running time is used for calculating the speed distribution of the vehicle between the stations at each moment; counting and analyzing the card swiping and boarding time of passengers at each station to obtain the passenger flow distribution condition of each station in each time period;

(2) simulating each shift by utilizing a multithreading technology simulation on the basis of a given departure time interval according to the identified shift, the inter-site information result and the passenger distribution in different time periods among the sites, and counting the arrival time of the vehicle and the full load rate index of the vehicle;

(3) the times of arrival time intervals in a reasonable interval are more and the average full load rate of the vehicles is larger by adjusting the departure time interval sequence, namely a reasonable multi-objective optimization model with the maximum times of arrival time intervals and the larger average full load rate of the vehicles is solved;

(4) and finally, searching and solving the bus departure time interval through an ant colony algorithm, outputting a departure interval sequence and constructing a departure timetable.

2. The ant colony algorithm-based bus departure time interval optimization method according to claim 1, wherein the solving step of the step 4 is as follows:

(1) dividing one day into 5 time periods of a flat peak period, an early peak period, a flat peak period, a late peak period and a flat peak period according to a passenger flow travel rule, wherein the departure time intervals in each time period are the same, for example, a departure time interval sequence [10,4,7,4,8,8] represents that the 1 st flat peak period is issued one shift every 10 minutes and serves as the initial position of ants;

(2) initializing the positions of M ants according to the rule to form a plurality of initial paths;

(3) initializing pheromones on each path;

(4) ants skip between each time interval, and add the selected value into the path of the solution;

(5) sequentially calculating the objective function value of each path, and selecting the optimal path to enter the next iteration;

(6) and when the iteration times reach the maximum value or the objective function value is not changed basically, terminating the iteration and outputting the departure interval sequence.

Images