CN108734337B - Customized bus station generation method based on cluster center correction - Google Patents

Customized bus station generation method based on cluster center correction Download PDF

Info

Publication number
CN108734337B
CN108734337B CN201810348137.4A CN201810348137A CN108734337B CN 108734337 B CN108734337 B CN 108734337B CN 201810348137 A CN201810348137 A CN 201810348137A CN 108734337 B CN108734337 B CN 108734337B
Authority
CN
China
Prior art keywords
clustering
station
ride
cluster
demand
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810348137.4A
Other languages
Chinese (zh)
Other versions
CN108734337A (en
Inventor
闫学东
李云伟
邵雯
刘凤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Jiaotong University
Original Assignee
Beijing Jiaotong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Jiaotong University filed Critical Beijing Jiaotong University
Priority to CN201810348137.4A priority Critical patent/CN108734337B/en
Publication of CN108734337A publication Critical patent/CN108734337A/en
Application granted granted Critical
Publication of CN108734337B publication Critical patent/CN108734337B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/231—Hierarchical techniques, i.e. dividing or merging pattern sets so as to obtain a dendrogram
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • Operations Research (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Game Theory and Decision Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Traffic Control Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The embodiment of the invention provides a customized bus carpooling station generation method based on cluster center correction. The method comprises the following steps: determining the number of the best stations of the customized bus, constructing an initial clustering carpooling station set, correcting operation of a clustering center and judging the generation result of the clustering carpooling stations. The method and the system can reasonably arrange the custom bus station based on the passenger reservation data, provide corresponding reference for site selection of the custom bus station at the current stage, and provide method support for opening and implementing the custom bus. The method for generating the customized bus station for co-taking is suitable for network topological structures of any real road network, overcomes the defect that the traditional bus station planning needs artificial subjective adjustment according to experience, overcomes the defect that the current research on the layout of the customized bus station is not based on the actual road network environment, and ensures the scientificity, accuracy, rationality and effectiveness of the layout of the station for co-taking.

Description

Customized bus station generation method based on cluster center correction
Technical Field
The invention relates to the technical field of intelligent public transportation, in particular to a customized bus station generation method based on cluster center correction.
Background
With the acceleration of the urbanization process in China and the limitation of urban public transportation systems, customized public transportation gradually enters the visual field of people. To date, customized public transportation routes such as Beijing, Qingdao, Jinan and the like are established in many cities in China, and the customized public transportation is the first choice for resident travel of some cities due to the high-quality service characteristic of the customized public transportation different from the traditional public transportation. The development of the customized public transport in China is still in a preliminary exploration stage, and a unified methodology for station layout, line design, vehicle scheduling and the like of the customized public transport is not formed yet. Therefore, how to reasonably arrange the customized bus station according to the passenger reservation data based on the actual road network environment is an important issue worthy of the current traffic researchers.
At present, with the opening and the rise of the customized bus, a small number of scholars in China begin to pay attention to the theoretical research of the layout of the customized bus stops, wherein the more prominent scholars comprise Hulier grids, Majiui and the like, but the research time is short, and the results are relatively few. Most of the methods applied in the current research are based on methods such as K-means clustering and hierarchical clustering, most of the achievements adopt one clustering method to research the problem, the limitation of the single method is not considered, meanwhile, a site layout method is not matched with an actual road network, even if the actual road network is considered, the site generation position is mostly adjusted artificially and subjectively according to the actual situation, and the scientificity and the effectiveness are lacked. The density of foreign population is low, the positioning of demand response type public transportation is mainly used for 'door-to-door' transportation in a low-density area for serving transportation travel demands, most of related researches do not consider the problem of setting of ride-sharing stations, each passenger demand point is used as a station for research, and the demand response type public transportation is not suitable for the situation of China.
Therefore, it is necessary to design a method for generating a customized bus stop pool, which is used for reasonably laying the customized bus stop pool and providing a corresponding reference for site selection of the customized bus stop pool at the current stage.
Disclosure of Invention
The embodiment of the invention provides a customized bus ride sharing station generation method based on cluster center correction, which aims to solve the problems in the background technology.
In order to achieve the purpose, the invention adopts the following technical scheme:
the embodiment of the invention provides a customized bus stop-by-stop generation method based on cluster center correction, which is characterized by comprising the following steps:
step 1: classifying the passenger reservation demand points by utilizing a hierarchical clustering method, and determining the number of the best bus station;
step 2: constructing a passenger demand set and a sample characteristic vector set according to the number of the optimal stations and the space coordinates of the demand points, and acquiring an initial clustering carpooling station set;
and step 3: correcting and clustering the initial clustering centers which are not overlapped with the alternative ride-sharing sites in the initial clustering ride-sharing site set to enable the clustering centers to fall on a road network to obtain clustering ride-sharing sites;
and 4, step 4: and judging the generation result of the clustering ride-sharing site through the updated value function value.
Preferably, the classifying the passenger reservation demand points by using a hierarchical clustering method to determine the number of the best bus station for customization comprises the following steps:
step 1.1: collecting spatial position coordinate data of a passenger reservation demand point, wherein the passenger reservation demand comprises: an entering point and a leaving point;
step 1.2: calculating Euclidean distances among all the demand points;
step 1.3: performing hierarchical clustering on all demand points according to the Euclidean distance and a maximum distance measurement criterion to generate a clustering tree;
step 1.4: and determining the value of the maximum distance, and determining the optimal number c of stops of the customized bus by taking the value of the maximum distance as a classification standard according to the clustering tree.
Preferably, the determining the value of the maximum distance, and determining the optimal number c of stops of the customized bus by using the value of the maximum distance as a classification standard according to the clustering tree, includes:
the values for the maximum distance are as follows: determining a station service radius R, wherein the station service radius R is the maximum traveling distance of passengers, and the value is as follows: 500 ~ 1000m, the coverage of website is for using this website as the centre of a circle, with R as the circular range of radius external radiation, the maximum distance value between two websites is: 4R;
and determining the optimal station number c of the customized bus by taking the maximum distance value 4R as a classification standard according to the clustering tree.
Preferably, the constructing a passenger demand set and a sample feature vector set according to the optimal number of stations and the spatial coordinates of demand points, and obtaining an initial clustering carpooling station set includes:
step 2.1: the passenger demand set X is constructed as:
X＝{x1,x2,...,xn}， (1)
wherein n is the required number;
each demand x in the set of passenger demandsiThe sample feature vector of (a) is:
(xi1,xi2,...,xim)T， (2)
wherein m is the index number of classification research, aggregation is carried out on a demand point space, and m is 2;
step 2.2: and acquiring an initial clustering carpooling station point set by using a fuzzy c-means clustering method.
Preferably, the obtaining an initial clustering multiplying station set by using a fuzzy c-means clustering method includes:
step 2.2.1: defining fuzzy c space to obtain c × n matrix U, and initializing membership matrix U by using random numbers with values between [0,1] so as to satisfy the following formula:
Figure BDA0001632545200000031
step 2.2.2: constructing a clustering co-multiplication site coordinate set Y:
Y＝{(x,y)|(x,y)∈S}， (4)
the method comprises the following steps that S represents a set of alternative station points for co-taking in the road network, and is the sum of road network nodes and intermediate points which can serve as stations in road sections;
initialize Y, order
Figure BDA0001632545200000041
Step 2.2.3: substituting the matrix U into the following equation:
Figure BDA0001632545200000042
wherein p is a weighted index, and p ∈ (1, ∞);
obtaining initial clustering centers of the c ride-sharing sites, and simultaneously constructing temporary clustering ride-sharing sites as follows:
CL＝{ci|i＝1,...,c}，Card(CL)＝c， (6)
wherein, CLI.e. the initial cluster pool site set.
Preferably, the modifying and clustering the initial clustering centers which are not overlapped with the alternative ride-sharing sites in the initial clustering ride-sharing site set to make the clustering centers fall on the road network to obtain the clustering ride-sharing sites includes:
step 3.1: repeating the judging operation until
Figure BDA0001632545200000043
Step 3.2: the judging operation is as follows:
if c isiIf the new cluster belongs to Y, the initial cluster carpooling station set is obtained again, and if the new cluster carpooling station set does not belong to Y, the cluster center c is judgediTo generate a corresponding candidate clustered site set Ci；
Step 3.3: let CL＝CL\{ciAnd for the alternative clustering site set CiMake a judgment if CiIf the set is empty, storing the ci in the Y, and returning to the step 3.1 to perform the repeated judgment operation; if it is
Figure BDA0001632545200000044
Returning to the step 2.2, obtaining the initial clustering co-multiplication station by using a fuzzy c-means clustering method; otherwise, C is addediThe elements in the cluster are sequentially taken out and recorded as a, and then the current cluster and ride station set Y' is:
Y'＝{Y,a,CL}， (7)
the formula for calculating the value of the cost function is as follows:
Figure BDA0001632545200000051
Figure BDA0001632545200000052
wherein d isijRepresenting the Euclidean distance between the clustering center i and the demand point j;
if it is
Figure BDA0001632545200000053
Selecting the CiThe point with the minimum value function value is stored as a clustering site in Y, otherwise, C is madei'＝Ci-Cin.Y, selecting CiThe point with the minimum value function value is used as a clustering site and stored into Y, and the corresponding value function value is recorded as F;
step 3.4: updating the rows i +1 to c of the U matrix according to the formulas (10) and (11), and returning to the step 3.1;
Figure BDA0001632545200000054
Figure BDA0001632545200000055
preferably, the generation of the corresponding candidate clustered site set CiThe method comprises the following steps:
setting UNIT as the minimum closed polygon which can be formed by the road network nodes and is the minimum road network UNIT;
step 3.2.1: if the cluster center c is generatediLocated in the road network UNIT UNIT, then CiA set is formed by points in all S covered by each side of the unit, and the points in all S comprise vertexes;
step 3.2.2: if the generated clustering center ci is located on the road network, the distance c between the two sides of the road section is respectively selectediThe nearest point in S is taken as ciAlternative clustering sites for points, this time CiForming a set for the selected two sites；
Step 3.2.3: if the generated clustering center ci is coincident with the point in S covered by each side of UNIT of road network, the clustering center ci is determined to be coincident with the point in S covered by each side of UNIT
Figure BDA0001632545200000056
The points in S include vertices.
Preferably, the determining the generated result of the clustering and factoring site through the updated cost function value includes:
calculating the change quantity of the finally updated value function value F relative to the last value function value;
if the change amount is larger than or equal to the preset threshold epsilon, the initialized U matrix is made to be the U matrix finally updated by the passenger clustering center correction operation method in the step 3, the step 2.2 is returned, the initial clustering carpooling station set is obtained again, and iteration is continued;
and if the change amount is smaller than a preset threshold epsilon, stopping the calculation, wherein Y is the set of the customized public transportation cluster and carpooling stations.
According to the technical scheme provided by the embodiment of the invention, the embodiment of the invention provides the customized bus ride-sharing stop generation method based on cluster center correction, an initial cluster ride-sharing stop set is constructed by determining the optimal stop number of the customized buses, the passenger cluster center is corrected, and the cluster ride-sharing stop generation result is judged, so that the customized bus ride-sharing stops are reasonably arranged. The invention is suitable for the network topology structure of any real road network, can overcome the defect that the current customized bus stop layout research is not based on the actual road network environment, can ensure the scientificity, accuracy, rationality and effectiveness of the layout of the ride-sharing stops, provides corresponding reference for the site selection of the customized bus stop in the current stage, and provides method support for the opening and implementation of the customized bus.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive labor.
Fig. 1 is a processing flow chart of a customized bus stop generation method based on cluster center correction according to an embodiment of the present invention;
fig. 2 is a flow chart of a method for determining the number of best stations of a customized bus based on a method for generating a customized bus stop for a passenger pool based on cluster center correction according to an embodiment of the present invention;
fig. 3 is a flowchart of an initial cluster pool station set construction method of a customized bus pool station generation method based on cluster center correction according to an embodiment of the present invention;
FIG. 4 is a flowchart of a passenger cluster center correction operation method of a customized bus pool station generation method based on cluster center correction according to an embodiment of the present invention;
fig. 5 is a schematic view of passenger cluster center correction operation of a customized bus pool station generation method based on cluster center correction according to an embodiment of the present invention;
fig. 6 is a diagram of an initial cluster ride-sharing stop set generation result of the customized bus ride-sharing stop generation method based on cluster center correction according to the embodiment of the present invention;
fig. 7 is a diagram of a result generated by a customized bus stop after modification of the customized bus stop generation method based on cluster center modification according to the embodiment of the present invention;
fig. 8 is a diagram illustrating analysis of a classification effect of a demand sample of a customized bus ride-sharing station generation method based on cluster center correction according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or coupled. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
For the convenience of understanding the embodiments of the present invention, the following description will be further explained by taking several specific embodiments as examples in conjunction with the drawings, and the embodiments are not to be construed as limiting the embodiments of the present invention.
Example one
The embodiment of the invention provides a customized bus stop-taking station generation method based on cluster center correction, which is used for reasonably arranging customized bus stop-taking stations based on passenger reservation data.
The processing flow chart of the customized bus stop-and-ride station generation method based on cluster center correction provided by the embodiment of the invention is shown in fig. 1, and specifically comprises the following steps:
s110: and classifying the passenger reservation demand points by using a hierarchical clustering method, and determining the number of the best bus station to be customized.
Step 1.1: collecting spatial position coordinate data of a passenger reservation demand point, wherein the passenger reservation demand comprises: an upper vehicle point and a lower vehicle point.
Step 1.2: and calculating Euclidean distances among all the demand points.
Step 1.3: and performing hierarchical clustering on all the demand points according to the Euclidean distance and the maximum distance measurement criterion to generate a clustering tree.
Step 1.4: and determining the value of the maximum distance, and determining the optimal number c of stops of the customized bus by taking the value of the maximum distance as a classification standard according to the clustering tree.
The values for the maximum distance are as follows: determining a station service radius R, wherein the station service radius R is the maximum traveling distance of passengers, and the value is as follows: 500 ~ 1000m, the coverage of website is for using this website as the centre of a circle, with R as the circular range of radius external radiation, the maximum distance value between two websites is: 4R.
And determining the optimal station number c of the customized bus by taking the maximum distance value 4R as a classification standard according to the clustering tree.
S120: and constructing a passenger demand set and a sample characteristic vector set according to the number of the optimal stations and the space coordinates of the demand points, and acquiring an initial clustering carpooling station set.
Step 2.1: the passenger demand set X is constructed as:
X＝{x1,x2,...,xn}， (1)
wherein n is the required number.
Each demand x in the set of passenger demandsiThe sample feature vector of (a) is:
(xi1,xi2,...,xim)T， (2)
and m is the number of indexes of classification research, and is collected in a demand point space, wherein m is 2.
Step 2.2: and acquiring an initial clustering carpooling station point set by using a fuzzy c-means clustering method.
Step 2.2.1: defining fuzzy c space to obtain c × n matrix U, and initializing membership matrix U by using random numbers with values between [0,1] so as to satisfy the following formula:
Figure BDA0001632545200000091
step 2.2.2: constructing a clustering co-multiplication site coordinate set Y:
Y＝{(x,y)|(x,y)∈S}， (4)
wherein, S represents a set of alternative station points in the road network, and S is the sum of the nodes and the intermediate points of the road network which can be used as the stations.
Initialize Y, order
Figure BDA0001632545200000092
Step 2.2.3: substituting the matrix U into the following equation:
Figure BDA0001632545200000101
where p is a weighted index and p ∈ (1, ∞).
Obtaining initial clustering centers of the c ride-sharing sites, and simultaneously constructing temporary clustering ride-sharing sites as follows:
CL＝{ci|i＝1,...,c}，Card(CL)＝c， (6)
wherein, CLI.e. the initial cluster pool site set.
S130: and correcting and clustering the initial clustering centers which are not overlapped with the alternative ride-sharing sites in the initial clustering ride-sharing site set to enable the clustering centers to fall on the road network to obtain the clustering ride-sharing sites.
Step 3.1: repetition judgmentCutting off the operation until
Figure BDA0001632545200000102
Step 3.2: the judgment operation is as follows: if c isiIf the new cluster belongs to Y, the initial cluster carpooling station set is obtained again, and if the new cluster carpooling station set does not belong to Y, the cluster center c is judgediTo generate a corresponding candidate clustered site set Ci。
And setting the UNIT as a minimum closed polygon which can be formed by the road network nodes and is a minimum road network UNIT.
Step 3.2.1: if the cluster center c is generatediLocated in the road network UNIT UNIT, then CiThe set of points in all S covered by each side of the cell, the points in all S including the vertices.
Step 3.2.2: if the cluster center c is generatediOn the road net, the distance c between two sides of the road section is selectediThe nearest point in S is taken as ciAlternative clustering sites for points, this time CiA set is formed for the two sites selected.
Step 3.2.3: if the generated clustering center ci is coincident with the point in S covered by each side of UNIT of road network, the clustering center ci is determined to be coincident with the point in S covered by each side of UNIT
Figure BDA0001632545200000103
The points in S include vertices.
Step 3.3: let CL＝CL\{ciAnd for the alternative clustering site set CiMake a judgment if CiIf the set is empty, storing the ci in the Y, and returning to the step 3.1 to perform the repeated judgment operation; if it is
Figure BDA0001632545200000104
Returning to the step 2.2, obtaining the initial clustering co-multiplication station by using a fuzzy c-means clustering method; otherwise, C is addediThe elements in the cluster are sequentially taken out and recorded as a, and then the current cluster and ride station set Y' is:
Y'＝{Y,a,CL}。 (7)
the formula for calculating the value of the cost function is as follows:
Figure BDA0001632545200000111
Figure BDA0001632545200000112
wherein d isijRepresenting the euclidean distance of the cluster center i from the demand point j.
If it is
Figure BDA0001632545200000113
Selecting the CiThe point with the minimum value function value is stored as a clustering site in Y, otherwise, C is madei'＝Ci-Cin.Y, selecting CiIn this method, the point at which the value function value is the smallest is stored as a cluster point in Y, and the corresponding value function value is denoted as F.
Step 3.4: and (5) updating the rows i +1 to c of the U matrix according to the equations (10) and (11), and returning to the step 3.1.
The calculation formula is as follows:
Figure BDA0001632545200000114
Figure BDA0001632545200000115
s140: and judging the generation result of the clustering ride-sharing site through the updated value function value.
And calculating the change quantity of the finally updated value function value F relative to the last value function value.
And if the change amount is greater than or equal to the preset threshold epsilon, making the initialized U matrix the U matrix finally updated by the passenger clustering center correction operation method in the step S130, returning to the step 2.2 to obtain the initial clustering carpooling station set again, and continuing iteration.
And if the change amount is smaller than a preset threshold epsilon, stopping the calculation, wherein Y is the set of the customized public transportation cluster and carpooling stations.
Example two
The embodiment provides a processing flow of a customized bus stop-and-ride station generation method based on cluster center correction, as shown in fig. 1, and the method comprises the following processing steps:
step 1: and classifying the passenger reservation demand points by using a hierarchical clustering method, and determining the number of the best bus station to be customized.
The classification means that all demand points (including an entering point and a leaving point) are classified into several types according to spatial positions, and the number of the classified types is the number of the station points; the getting-on point and the getting-off point respectively refer to a starting point and a destination submitted by each passenger when the passenger selects the customized bus, and the getting-on point and the getting-off point are both demand points.
Step 2: and according to the obtained optimal station number and the space coordinates of the demand points, the passenger demand set and the sample characteristic vector set, and acquiring an initial clustering carpooling station set.
And step 3: and carrying out a series of correction and clustering operations on the initial clustering centers which are not overlapped with the alternative carpooling sites of the road network in the initial clustering carpooling site set, so that the clustering centers fall on the road network.
And 4, step 4: and judging the result generated by the clustering and multiplying station.
As shown in fig. 2, a flow chart of a method for determining the number of best stations of a customized bus includes the following specific steps:
(1) and acquiring position coordinate data of the appointed passenger demand point pair, including a boarding point and a alighting point.
(2) The distances between all the demand points are calculated.
(3) And performing hierarchical clustering according to the maximum distance measurement criterion until all the demand points are clustered, and generating a clustering tree.
(4) And (3) determining a maximum distance value, and determining the number c of the best bus stations customized according to the clustering tree generated in the step (3) by taking the maximum distance value as a classification standard.
Fig. 3 shows a flowchart of a method for constructing an initial clustering ride-sharing site set, which includes the following steps:
(1) constructing a passenger demand set X ═ X1,x2,...,xnN is the number of requirements, each requirement xiWith sample feature vector (x)i1,xi2,...,xim)TAnd m is the index number of classification research. The method mainly relates to the aggregation of demand points in space, so m is 2, and the related index is the horizontal and vertical coordinates of the demand points.
(2) And acquiring an initial clustering carpooling station point set by using a fuzzy c-means clustering method. The initial clustering ride-sharing station set obtaining method based on the fuzzy c-means clustering comprises the following specific processes:
A. defining fuzzy c space to obtain c × n matrix U, and initializing membership matrix U by using random numbers with values between [0,1] so as to satisfy the following formula:
Figure BDA0001632545200000131
B. and constructing a clustering co-multiplication station coordinate set Y { (x, Y) | (x, Y) ∈ S }, wherein S represents a road network candidate co-multiplication station set and is the sum of intermediate points which can be used as stations in road network nodes and road sections. Initialize Y, order
Figure BDA0001632545200000132
C. Calculating the following formula by using a matrix U generated by initialization to obtain initial clustering centers of C ride-sharing sites, and constructing a temporary clustering ride-sharing site set CL＝{ci|i＝1,...,c}，Card(CL)＝c，CLThe initial clustering carpooling station set is obtained; the calculation formula is as follows:
Figure BDA0001632545200000133
where p is a weighted index and p ∈ (1, ∞).
Fig. 4 is a flowchart of a passenger cluster center correction operation method, which includes the following steps:
(1) the following operations were repeated until
Figure BDA0001632545200000134
(2) Judgment ciWhether the epsilon is true or not is determined, if yes, the initial clustering and carpooling station set needs to be obtained again; otherwise, judging the clustering center ciPosition, generating corresponding candidate clustering site set Ci。
ciCorresponding candidate clustered site set CiThe generation method comprises the following steps:
and setting the UNIT as the minimum closed polygon which can be formed by the nodes of the road network. In this embodiment, a grid-shaped road network is taken as an example for explanation, and all road network nodes form a road network candidate station set. The modified operation is schematically shown in fig. 5, where points A, B, C, D are road network nodes, which together form a road network UNIT.
A. If the cluster center c is generatediLocated within the road network UNIT UNIT, as shown in FIG. 5 (1), the vertex A, B, C, D of the UNIT is selected as ciAlternative clustering sites for points, this time Ci＝{A,B,C,D}。
B. If the cluster center c is generatediIf the road network is located, c is selectediThe vertex of the road section is taken as ciAlternative clustering sites for points, in the case shown in FIG. 5.(2), ciThe corresponding set of candidate clustered sites is denoted Ci＝{A,B}。
C. If the cluster center c is generatediCoincident with the top of the UNIT UNIT of the road network, then
Figure BDA0001632545200000141
(3) Let CL＝CL\{ci}. Judgment CiWhether it is an empty set, if it is an empty set, ciStoring the obtained product in Y, and returning to the step 1; if it is
Figure BDA0001632545200000142
Re-acquiring the initial clustering carpooling station set; otherwise, sequentially taking CiIf the element (denoted as a) is the current cluster co-multiplication site set, Y' ═ Y, a, CLCalculating the value of the cost function according to the formulas (3) and (4), if
Figure BDA0001632545200000143
Selection CiThe point with the minimum value function value is stored as a clustering site in Y, otherwise C is madei'＝Ci-Cin.Y, selecting CiIn this method, the point at which the cost function value is the smallest is stored as a cluster site in Y, and the corresponding cost function value is denoted as F.
The calculation formula is as follows:
Figure BDA0001632545200000144
Figure BDA0001632545200000151
wherein d isijRepresenting the euclidean distance of the cluster center i from the demand point j.
(4) Updating rows i +1 to c of the U matrix according to the following two formulas, and returning to the step (1);
Figure BDA0001632545200000152
Figure BDA0001632545200000153
the method for judging the generation result of the clustering ride-sharing site in the embodiment is specifically realized as follows:
calculating the change quantity of the finally updated value function value F relative to the last value function value, and judging whether the change quantity is greater than or equal to a preset threshold value epsilon or not; if yes, the initialized membership matrix is a U matrix finally obtained by updating the correction operation, and the step of the initial clustering and multiplying station set construction method is returned; otherwise, stopping the algorithm, wherein Y is the set of the required customized public transportation cluster and ride-sharing stations.
EXAMPLE III
The embodiment provides a customized bus station generation method based on cluster center correction, which is specifically realized as follows:
(1) in the embodiment, a grid network environment is selected, and the network nodes are all the alternative ride-sharing sites.
(2) The number of the obtained optimal stops is 12 calculated by the method for determining the number of the optimal stops of the customized bus.
(3) Based on the optimal number of the sites, constructing an initial clustering and ride-sharing site set, and generating results as shown in fig. 6, it can be seen that the initial clustering and ride-sharing sites all deviate from a road network and are not overlapped with the alternative ride-sharing sites.
(4) And correcting the initial clustering center, judging the generation result, and obtaining the final station generation result as shown in fig. 7, wherein the passenger ride-sharing stations are distributed on the road network and coincide with the alternative ride-sharing stations.
Fig. 8 is a schematic diagram of solving the membership matrix, which is used for evaluating the classification effect of the samples, and it can be seen that the membership matrix has a large difference between each type, the discrimination is high, and the clustering effect is good.
In summary, the embodiment of the invention provides a customized bus stop-pooling generation method based on cluster center correction, which can reasonably arrange the customized bus stop-pooling based on passenger reservation data, and overcomes the defect that the traditional bus stop planning needs to be adjusted manually according to experience, overcomes the defect that the current customized bus stop arrangement research is not based on the actual road network environment, ensures the scientificity, accuracy, rationality and effectiveness of arrangement of the stop-pooling stations, and provides corresponding reference for site selection of the customized bus stop at the current stage.
Those of ordinary skill in the art will understand that: the figures are merely schematic representations of one embodiment, and the blocks or flow diagrams in the figures are not necessarily required to practice the present invention.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for apparatus or system embodiments, since they are substantially similar to method embodiments, they are described in relative terms, as long as they are described in partial descriptions of method embodiments. The above-described embodiments of the apparatus and system are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (4)

1. A customized bus station generation method based on cluster center correction is characterized by comprising the following steps:
step 1: classifying the passenger reservation demand points by utilizing a hierarchical clustering method, and determining the number of the best bus station;
step 2: constructing a passenger demand set and a sample characteristic vector set according to the number of the optimal stations and the space coordinates of the demand points, and acquiring an initial clustering carpooling station set;
and step 3: correcting and clustering the initial clustering centers which are not overlapped with the alternative ride-sharing sites in the initial clustering ride-sharing site set to enable the clustering centers to fall on a road network to obtain clustering ride-sharing sites;
and 4, step 4: judging the generated result of the clustering ride-sharing site according to the updated value function value;
according to the number of the optimal stations and the space coordinates of the demand points, a passenger demand set and a sample characteristic vector set are constructed, and an initial clustering carpooling station set is obtained, which comprises the following steps:
step 2.1: the passenger demand set X is constructed as:
X＝{x1,x2,...,xn}， (1)
wherein n is the required number;
each demand x in the set of passenger demandsiThe sample feature vector of (a) is:
(xi1,xi2,...,xim)T， (2)
wherein m is the index number of classification research, aggregation is carried out on a demand point space, and m is 2;
step 2.2: acquiring an initial clustering carpooling station set by using a fuzzy c-means clustering method;
the method for acquiring the initial clustering ride-sharing station set by using the fuzzy c-means clustering method comprises the following steps:
step 2.2.1: defining fuzzy c space to obtain c × n matrix U, and initializing membership matrix U by using random numbers with values between [0,1] so as to satisfy the following formula:
Figure FDA0003369683590000021
step 2.2.2: constructing a clustering co-multiplication site coordinate set Y:
Y＝{(x,y)|(x,y)∈S}， (4)
the method comprises the following steps that S represents a set of alternative station points for co-taking in the road network, and is the sum of road network nodes and intermediate points which can serve as stations in road sections;
initialize Y, order
Figure FDA0003369683590000022
Step 2.2.3: substituting the matrix U into the following equation:
Figure FDA0003369683590000023
wherein p is a weighted index, and p ∈ (1, ∞);
obtaining initial clustering centers of the c ride-sharing sites, and simultaneously constructing temporary clustering ride-sharing sites as follows:
CL＝{ci|i＝1,...,c}，Card(CL)＝c， (6)
wherein, CLNamely the initial clustering carpooling station set;
the method for correcting and clustering the initial clustering centers which are not overlapped with the alternative ride-sharing sites in the initial clustering ride-sharing site set to enable the clustering centers to fall on the road network to obtain the clustering ride-sharing sites comprises the following steps:
step 3.1: repeating the judging operation until
Figure FDA0003369683590000024
Step 3.2: the judging operation is as follows:
if c isiIf the new cluster belongs to Y, the initial cluster carpooling station set is obtained again, and if the new cluster carpooling station set does not belong to Y, the cluster center c is judgediTo generate a corresponding candidate clustered site set Ci；
Step 3.3: let CL＝CL\{ciAnd for the alternative clustering site set CiMake a judgment if CiFor the empty set, ciStoring the data into Y, and returning to the step 3.1 to perform the repeated judgment operation; if it is
Figure FDA0003369683590000025
Returning to the step 2.2, obtaining the initial clustering co-multiplication station by using a fuzzy c-means clustering method; otherwise, C is addediThe elements in the cluster are sequentially taken out and recorded as a, and then the current cluster and ride station set Y' is:
Y'＝{Y,a,CL}， (7)
the formula for calculating the value of the cost function is as follows:
Figure FDA0003369683590000031
Figure FDA0003369683590000032
wherein d isijRepresenting the Euclidean distance between the clustering center i and the demand point j;
if it is
Figure FDA0003369683590000033
Selecting the CiThe point with the minimum value function value is stored as a clustering site in Y, otherwise, C is madei'＝Ci-Cin.Y, selecting CiThe point with the minimum value function value is used as a clustering site and stored into Y, and the corresponding value function value is recorded as F;
step 3.4: updating the rows i +1 to c of the U matrix according to the formulas (10) and (11), and returning to the step 3.1;
Figure FDA0003369683590000034
Figure FDA0003369683590000035
generating a corresponding candidate clustering site set CiThe method comprises the following steps:
setting UNIT as the minimum closed polygon which can be formed by the road network nodes and is the minimum road network UNIT;
step 3.2.1: if the cluster center c is generatediLocated in the road network UNIT UNIT, then CiA set is formed by points in all S covered by each side of the unit, and the points in all S comprise vertexes;
step 3.2.2: if the cluster center c is generatediOn the road network, respectively selecting the distance c on both sides of the road networkiThe nearest point in S is taken as ciAlternative clustering sites for points, this time CiForming a set for the selected two sites;
step 3.2.3: if the cluster center c is generatediCoincide with the points in S covered by each side of Unit UNIT of road network
Figure FDA0003369683590000041
The points in S include vertices.
2. The cluster center correction-based customized bus stop-pooling generation method according to claim 1, wherein the step of classifying the passenger reservation demand points by using a hierarchical clustering method to determine the number of the best stops of the customized bus comprises the following steps:
step 1.1: collecting spatial position coordinate data of a passenger reservation demand point, wherein the passenger reservation demand comprises: an entering point and a leaving point;
step 1.2: calculating Euclidean distances among all the demand points;
step 1.3: performing hierarchical clustering on all demand points according to the Euclidean distance and a maximum distance measurement criterion to generate a clustering tree;
step 1.4: and determining the value of the maximum distance, and determining the optimal number c of stops of the customized bus by taking the value of the maximum distance as a classification standard according to the clustering tree.
3. The method for generating the customized bus stop-and-station based on cluster center correction as claimed in claim 2, wherein the determining the value of the maximum distance determines the optimal number c of stops of the customized bus by using the value of the maximum distance as a classification standard according to the cluster tree, and comprises the following steps:
the values for the maximum distance are as follows: determining a station service radius R, wherein the station service radius R is the maximum traveling distance of passengers, and the value is as follows: 500 ~ 1000m, the coverage of website is for using this website as the centre of a circle, with R as the circular range of radius external radiation, the maximum distance value between two websites is: 4R;
and determining the optimal station number c of the customized bus by taking the maximum distance value 4R as a classification standard according to the clustering tree.
4. The cluster center correction-based customized bus ride-sharing station generation method according to claim 1, wherein the step of judging the generation result of the cluster ride-sharing station through the updated value function value comprises the steps of:
calculating the change quantity of the finally updated value function value F relative to the last value function value;
if the change amount is larger than or equal to the preset threshold epsilon, the initialized U matrix is made to be the U matrix finally updated by the passenger clustering center correction operation method in the step 3, the step 2.2 is returned, the initial clustering carpooling station set is obtained again, and iteration is continued;
and if the change amount is smaller than a preset threshold epsilon, stopping the calculation, wherein Y is the set of the customized public transportation cluster and carpooling stations.
CN201810348137.4A 2018-04-18 2018-04-18 Customized bus station generation method based on cluster center correction Active CN108734337B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810348137.4A CN108734337B (en) 2018-04-18 2018-04-18 Customized bus station generation method based on cluster center correction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810348137.4A CN108734337B (en) 2018-04-18 2018-04-18 Customized bus station generation method based on cluster center correction

Publications (2)

Publication Number Publication Date
CN108734337A CN108734337A (en) 2018-11-02
CN108734337B true CN108734337B (en) 2022-03-18

Family

ID=63939569

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810348137.4A Active CN108734337B (en) 2018-04-18 2018-04-18 Customized bus station generation method based on cluster center correction

Country Status (1)

Country Link
CN (1) CN108734337B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109781130B (en) * 2019-02-01 2023-04-18 上海雷腾软件股份有限公司 Method and equipment for automatically planning bus route
CN110458309B (en) * 2019-06-29 2023-07-11 东南大学 Network about splicing station point location method based on actual road network environment
CN112396324A (en) * 2020-11-19 2021-02-23 北京清研宏达信息科技有限公司 Implementation method of microcirculation network appointment bus Minibus automatic scheduling system
CN113096377B (en) * 2021-02-18 2022-07-29 西南交通大学 Vehicle carpooling planning method based on urban heterogeneity
CN113326989A (en) * 2021-06-15 2021-08-31 北京沃东天骏信息技术有限公司 Method and system for optimizing vehicle route
CN115186049B (en) * 2022-09-06 2023-02-03 深圳市城市交通规划设计研究中心股份有限公司 Intelligent bus alternative station site selection method, electronic equipment and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102592447A (en) * 2011-12-20 2012-07-18 浙江工业大学 Method for judging road traffic state of regional road network based on fuzzy c means (FCM)
CN104318324A (en) * 2014-10-13 2015-01-28 南京大学 Taxi GPS (Global Positioning System) record based airport bus station and path planning method
CN105070044A (en) * 2015-08-17 2015-11-18 南通大学 Dynamic scheduling method for customized buses and car pooling based on passenger appointments
CN107832779A (en) * 2017-12-11 2018-03-23 北方工业大学 Track station classification system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170109764A1 (en) * 2015-10-19 2017-04-20 Xerox Corporation System and method for mobility demand modeling using geographical data

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102592447A (en) * 2011-12-20 2012-07-18 浙江工业大学 Method for judging road traffic state of regional road network based on fuzzy c means (FCM)
CN104318324A (en) * 2014-10-13 2015-01-28 南京大学 Taxi GPS (Global Positioning System) record based airport bus station and path planning method
CN105070044A (en) * 2015-08-17 2015-11-18 南通大学 Dynamic scheduling method for customized buses and car pooling based on passenger appointments
CN107832779A (en) * 2017-12-11 2018-03-23 北方工业大学 Track station classification system

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
城市定制公交合乘站点的布局研究;胡列格 等;《徐州工程学院学报（自然科学版）》;20160331;第31卷(第1期);27-32 *
定制公交线路和站点规划研究;王飞;《中国优秀硕士学位论文全文数据库（电子期刊）》;20170630;26 *
马继辉 等.定制公交线路开行方案研究.《城市公共交通》.2016, *

Also Published As

Publication number Publication date
CN108734337A (en) 2018-11-02

Similar Documents

Publication Publication Date Title
CN108734337B (en) Customized bus station generation method based on cluster center correction
CN108053062B (en) Customized bus route generation method based on multi-source data
CN106651100B (en) Air quality evaluation system and method based on vehicle-mounted monitoring points optimized through Internet of vehicles
CN110458048A (en) Take population distribution Spatio-temporal Evolution and the cognition of town pattern feature into account
CN105206048A (en) Urban resident traffic transfer mode discovery system and method based on urban traffic OD data
CN111126799B (en) Shared network driver and crew matching method based on bipartite graph
CN109785618A (en) Short-term traffic flow prediction method based on combinational logic
CN110909434B (en) Urban public transport trunk line network design method under low-carbon guidance
CN106126816A (en) Repeat the extensive ALS building point cloud modeling method building under automatic sensing
CN108805345A (en) A kind of crime space-time Risk Forecast Method based on depth convolutional neural networks model
CN108256969A (en) A kind of public bicycles lease point dispatcher-controlled territory division methods
CN115936240B (en) Shared bicycle demand prediction and delivery scheduling method
CN110175865A (en) Electric car charging real time pricing method based on ubiquitous cognition technology
CN114386536B (en) Region determination method, device, computing equipment and storage medium
CN111310257A (en) Regional building energy consumption prediction method under BIM environment
CN114511143A (en) Urban rail transit network generation method based on grouping division
Novačko et al. Estimation of origin-destination trip matrices for small cities
CN116796904A (en) Method, system, electronic equipment and medium for predicting new line passenger flow of rail transit
CN106980942A (en) Calculate method of the bicycle free way to the coverage of public bicycles lease point
CN111008730B (en) Crowd concentration prediction model construction method and device based on urban space structure
CN116451963B (en) Multi-factor pension service facility optimal configuration method based on ensemble learning
Rao et al. Comparisons of three‐dimensional urban forms in different urban expansion types: 58 sample cities in China
CN112633591A (en) Space search method and device based on deep reinforcement learning
CN112527935A (en) Gasoline station site selection analysis system based on Voronoi diagram
CN116612633A (en) Self-adaptive dynamic path planning method based on vehicle-road cooperative sensing

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant