CN111126681B - Bus route adjusting method based on historical passenger flow - Google Patents

Bus route adjusting method based on historical passenger flow Download PDF

Info

Publication number
CN111126681B
CN111126681B CN201911274222.1A CN201911274222A CN111126681B CN 111126681 B CN111126681 B CN 111126681B CN 201911274222 A CN201911274222 A CN 201911274222A CN 111126681 B CN111126681 B CN 111126681B
Authority
CN
China
Prior art keywords
passenger flow
correlation
stations
time
subsequence
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
CN201911274222.1A
Other languages
Chinese (zh)
Other versions
CN111126681A (en
Inventor
李海波
翁邵源
陈文韵
高悦尔
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.)
Huaqiao University
Original Assignee
Huaqiao 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 Huaqiao University filed Critical Huaqiao University
Priority to CN201911274222.1A priority Critical patent/CN111126681B/en
Publication of CN111126681A publication Critical patent/CN111126681A/en
Application granted granted Critical
Publication of CN111126681B publication Critical patent/CN111126681B/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"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Game Theory and Decision Science (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention relates to a public transport route adjusting method based on historical passenger flow, which is characterized in that through traversing passenger flow time subsequences with all lengths and adopting correlation coefficients to perform clustering, clustering results form two time sequence sets, namely a positive correlation cluster set and a negative correlation cluster set, which respectively represent route stations with the same riding behavior trend and route stations with opposite riding behavior trend, and macroscopic passenger traveling behavior characteristics between routes and between stations can be obtained through analyzing the clustering results. The invention can not only understand the riding demand of passengers, help to find stations with the same or similar passenger flow change, arrange public traffic network for a city, set new line, adjust the existing line, provide decision support for line adjustment, but also improve public traffic scheduling and operate more scientifically and reasonably. The invention is independent of specific scenes, is more universal and can be used for public transport networks of large and medium-sized cities.

Description

Bus route adjusting method based on historical passenger flow
Technical Field
The invention relates to the technical field of data analysis, in particular to a public transport line adjusting method based on historical passenger flow.
Background
The setting and optimization of the public traffic network is a complex system project, and is often influenced by a plurality of factors, such as the limitation of objective factors such as hardware infrastructure and the like, and the design defects of the public traffic network, such as unreasonable structure, unbalanced network development, mismatching of line capacity and passenger flow and the like, are far less than expected for the throughput of the passenger flow. This very big influence the performance of city public transit overall efficiency. Therefore, the method has important significance for improving the passenger flow throughput of the whole public transport network by mining the passenger flow change trend between stations and between lines and finding the difference in the passenger flow change trend.
In the current method, a bus network is optimized mainly by means of modeling, such as building a station coverage model, dynamically designing a road network model according to bus demands, optimizing the network based on passenger flow distribution, and the like. Other methods include: the method comprises the steps of network optimization based on a genetic algorithm, a public transportation distribution method considering OD requirements, a cluster analysis method and the like.
At present, the data analysis method can better reflect objective passenger flow trend, and the relation between each line and each site can be objectively reflected through the statistics and clustering of passenger flow. However, under the influence of data volume, the time complexity of the conventional clustering method is often very high, and improvement is needed according to data characteristics so as to better meet the real analysis requirements.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a public transport line adjusting method based on historical passenger flow, which can not only know the riding requirements of passengers according to the increasing and decreasing change relation of the passenger flow among stations, lay a public transport network for a city, set a new line and adjust the existing line, but also improve public transport scheduling and operate more scientifically and reasonably.
The technical scheme of the invention is as follows:
a public transport route adjusting method based on historical passenger flow obtains the passenger flow correlation among stops, and adjusts the density of the public transport route passing through the stops based on the passenger flow correlation among the stops;
the steps of obtaining the passenger flow volume correlation between the stations are as follows:
1) counting historical card swiping data of all bus stops according to time intervals to form a passenger flow time sequence;
2) acquiring a passenger flow time subsequence by using a multi-scale sliding window;
3) calculating the sum of the passenger flow time subsequences;
4) computing arbitrary two sites Fi、FjCorrelation coefficient of passenger flow time subsequences;
when the correlation coefficient is positive, the passenger flow volume increasing and decreasing trends among the stations are the same; when the correlation coefficient is negative, the passenger flow volume increase and decrease trends among the stations are opposite; when the correlation coefficient is zero, the passenger flow volume increasing and decreasing trend between the stations is uncertain;
and selecting the stations with positive or negative correlation coefficients, and adjusting the density of the bus route according to the passenger flow increasing and decreasing trend among the stations.
Preferably, step 1) is specifically:
1.1) dividing time intervals for any station according to fixed time intervals, counting historical card swiping data of each station in each time interval, and forming a passenger flow time sequence F ═ (F)1,f2,f3,…,fn)；
1.2) repeating the step 1.1) to obtain the passenger flow time sequence set D of all the stations.
Preferably, step 2) is specifically:
2.1) extracting a passenger flow time subsequence S (F) from the time sequence F e D with the length of a sliding window being l and the step length index of the sliding window being 1k＝(fk,fk+1,…,fk+l-1)，k∈(1≤k≤n-l+1)；
2.2) repeating the step 2.1) until all passenger flow time subsequence set Sl of the passenger flow time sequence F are obtainedF；
2.3) repeating the step 2.1) and the step 2.2) until a subsequence set SD of all sequences in the passenger flow time sequence set D is obtained.
Preferably, the time of day is divided into a peak period and a non-peak period, the time interval of the peak period is shorter than the time interval of the non-peak period, and the length of the sliding window of the peak period is longer than the length of the sliding window of the non-peak period.
Preferably, step 3) is specifically:
3.1) calculating the sum of the passenger flow when the length of the subsequence is l as follows:
Figure BDA0002315099850000021
3.2) calculating the sum of the passenger flow when the length of the subsequence is l +1 as follows:
Suml+1(S(F)k)＝Suml(S(F)k)+fk+1。
preferably, the step 4) is specifically:
4.1) initializing positively correlated clusters
Figure BDA0002315099850000031
Initializing negative correlation clusters
Figure BDA0002315099850000032
4.2) separately calculating
Figure BDA0002315099850000033
All of the subsequences S (F)i)k、S(Fj)kAverage value of (2)
Figure BDA0002315099850000034
The method comprises the following specific steps:
Figure BDA0002315099850000035
Figure BDA0002315099850000036
4.3) pairs
Figure BDA0002315099850000037
Middle arbitrary subsequence S (F)i)kGo through
Figure BDA0002315099850000038
Respectively obtain S (F)i)kAnd
Figure BDA0002315099850000039
positive correlation of the collector sequences and Nε(S(Fi)k) And a set of negatively correlated collector subsequences Nε-(S(Fi)k)；
Figure BDA00023150998500000310
Figure BDA00023150998500000311
Wherein epsilon is a critical value of the site correlation coefficient,
Figure BDA00023150998500000312
is the degree of correlation between subsequences;
4.4) to N, respectivelyε(S(Fi)k) And Nε-(S(Fi)k) All sub-sequence traversals in
Figure BDA00023150998500000313
Respectively obtaining the sub-sequence sets related to the sub-sequences, and respectively adding the sub-sequence sets into the set Nε(S(Fi)k) And Nε-(S(Fi)k) Performing the following steps;
4.5) repeating step 4.3), step 4.4) until the set N is reachedε(S(Fi)k) And Nε-(S(Fi)k) No new subsequence is added, and positive correlation clusters C are formed respectivelykAnd negative correlation cluster
Figure BDA00023150998500000314
If a cluster C is positively correlatedkAnd negative correlation cluster
Figure BDA00023150998500000315
The geographical positions among the sites represented by the subsequences contained in the cluster are in accordance with a preset threshold value, and the related time intervals are in accordance with actual meanings, then the cluster set is added: c ← Ck，
Figure BDA00023150998500000316
4.6)
Figure BDA00023150998500000317
Repeating the step 4.3), the step 4.4) and the step 4.5) to form a final positive correlation cluster C and a final negative correlation cluster C-Respectively representing all time periods with positive correlation and all time periods with negative correlation of passenger flow of two stations.
Preferably, in step 4.3), the covariance
Figure BDA0002315099850000041
The invention has the following beneficial effects:
according to the method for adjusting the bus route based on the historical passenger flow, through traversing the passenger flow time subsequences with all lengths, clustering is carried out by adopting the correlation coefficient, and when the correlation coefficient is positive, the bus taking behaviors of passengers have positive correlation trends among different stations in the time period, namely the passenger flow increasing and decreasing trends among the stations are the same; when the correlation coefficient is negative, the passenger flow volume increase and decrease trends among the stations are opposite in the time period; and when the correlation coefficient is zero, the passenger flow volume increasing and decreasing trend among the stations in the time period is uncertain. And the final clustering result forms two time sequence sets, namely a positive correlation cluster set and a negative correlation cluster set, which respectively represent line stations with the same riding behavior trend and line stations with opposite riding behavior trends, so that macroscopic passenger travel behavior characteristics between lines and between stations can be obtained by analyzing the clustering result.
The invention can not only understand the riding demand of passengers, help to find stations with the same or similar passenger flow change, arrange public traffic network for a city, set new line, adjust the existing line, provide decision support for line adjustment, but also improve public traffic scheduling and operate more scientifically and reasonably.
The invention is independent of specific scenes, is more universal and can be used for public transport networks of large and medium-sized cities.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a schematic diagram of a process for finding related subsequences;
FIG. 3 is a schematic diagram of peak hours of the placing station in the example;
FIG. 4 is a schematic diagram of the passenger flow during the peak period of the pond side station in the embodiment;
FIG. 5 is a schematic diagram of obtaining the 1 st subsequence of the pond edge station with a length of 4 by using a sliding window in the example;
FIG. 6 is a schematic diagram of obtaining the 2 nd subsequence of the pond edge station with a length of 4 using a sliding window in the embodiment;
FIG. 7 is a schematic diagram of obtaining the 3 rd subsequence of the pond edge station with a length of 4 by using a sliding window in the embodiment;
FIG. 8 is a schematic diagram of obtaining the 4 th subsequence of the pond edge station with a length of 4 using a sliding window in the example;
FIG. 9 is a schematic diagram of the sum of the pond-edge station subsequences in the example;
FIG. 10 is a schematic representation of the summation of the screwing station subsequences in the example;
FIG. 11 is a diagram showing the subsequences in positive correlation with the connectable stations (subsequence length 4) in this example;
FIG. 12 is a schematic of the subsequence with negative correlation to the aluminium site (subsequence length 4, correlation period 0) in the example;
FIG. 13 is a diagram of the subsequences that are negatively correlated with the ligation station (subsequence length 5, correlation period 0) in the example;
FIG. 14 is a diagram of the subsequences that are negatively correlated with the screwing station (subsequence length 6, correlation period 0) in the example.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
The invention provides a public transport line adjusting method based on historical passenger flow, aiming at solving the defects in the prior art. When the correlation coefficient is positive, the passenger riding behavior in the time period has a positive correlation trend among different stations, namely the passenger flow volume increasing and decreasing trends among the stations are the same; when the correlation coefficient is negative, the passenger flow volume increase and decrease trends among the stations are opposite in the time period; and when the correlation coefficient is zero, the passenger flow volume increasing and decreasing trend among the stations in the time period is uncertain. And the final clustering result forms two time sequence sets, namely a positive correlation cluster set and a negative correlation cluster set, which respectively represent line stations with the same riding behavior trend and line stations with opposite riding behavior trends, so that macroscopic passenger travel behavior characteristics between lines and between stations can be obtained by analyzing the clustering result.
The invention relates to a public transport route adjusting method based on historical passenger flow, which comprises the following steps of firstly, obtaining the passenger flow correlation between stations; then, adjusting the density of the bus route passing through the stops based on the passenger flow correlation among the stops;
as shown in fig. 1, the steps of obtaining the passenger flow volume correlation between the stations are as follows:
1) counting historical card swiping data of all bus stops according to time intervals to form a passenger flow time sequence; the method specifically comprises the following steps:
1.1) dividing time intervals for any station according to fixed time intervals, counting historical card swiping data of each station in each time interval, and forming a passenger flow time sequence F ═ (F)1,f2,f3,…,fn)；
1.2) repeating the step 1.1) to obtain a passenger flow time sequence set D of all the stations.
In the passenger flow volume time sequence extraction process, working days and non-working days are considered separately, time of a day is divided into a peak period and a non-peak period, the peak period and the non-peak period need to be divided into periods according to different time intervals, and in specific implementation, the peak period is set according to each city situation, for example: 7:00-9:30. In the invention, the time interval of the peak period is shorter than that of the non-peak period, and according to the common condition, the peak period is divided by 10 minutes according to the departure interval of most buses, and the passenger flow time sequence of the peak period is extracted. And dividing the off-peak period by 30 minutes, and extracting the passenger flow time sequence in the off-peak period.
2) Acquiring a passenger flow time subsequence by using a multi-scale sliding window; the method specifically comprises the following steps:
2.1) extracting a passenger flow time subsequence S (F) from the time sequence F e D with the length of a sliding window being l and the step length index of the sliding window being 1k＝(fk,fk+1,…,fk+l-1)，k∈(1≤k≤n-l+1)；
2.2) repeating the step 2.1) until all passenger flow time subsequence set Sl of the passenger flow time sequence F are obtainedF；
2.3) repeating the step 2.1) and the step 2.2) until obtaining the subsequence set SD of all sequences in the passenger flow time series set D.
If the passenger flow volume is influenced mutually among the stations, the passenger flow volume can be reflected only in a certain period, and is reflected in 40-60 minutes on average according to the urban scale. In the present invention, the length of the sliding window during the peak period is greater than the length of the sliding window during the non-peak period. In specific implementation, the length l of the shortest subsequence of the passenger flow volume sequence in the peak period is 4; the length l of the shortest subsequence in off-peak periods is 2;
3) calculating the sum of the passenger flow time subsequences; the method specifically comprises the following steps:
3.1) calculating the sum of the passenger flow when the length of the subsequence is l as follows:
Figure BDA0002315099850000061
3.2) calculating the sum of the passenger flow when the length of the subsequence is l +1 as follows:
Suml+1(S(F)k)＝Suml(S(F)k)+fk+1。
4) computing arbitrary two sites Fi、FjThe correlation coefficient of the passenger flow time subsequences. When calculating the correlation degree between any two sites, only the numerical correlation degree is calculated, and the physical meaning represented by the subsequence is not considered. Therefore, when the invention is used for solving the correlation degree of the subsequence, the geographical position between the sites also needs to be considered, andwhether the relevant time period is in accordance with practical significance. The invention solves the correlation degree between the subsequences by using the Pearson correlation coefficient.
In step 4), starting from a certain subsequence, traversing
Figure BDA0002315099850000062
Solving a positive correlation set and a negative correlation set to finally form a cluster, which specifically comprises the following steps:
4.1) initializing positively correlated clusters
Figure BDA0002315099850000063
Initializing negative correlation clusters
Figure BDA0002315099850000064
4.2) separately calculating
Figure BDA0002315099850000065
All of the subsequences S (F)i)k、S(Fj)kAverage value of (2)
Figure BDA0002315099850000066
The method comprises the following specific steps:
Figure BDA0002315099850000071
Figure BDA0002315099850000072
4.3) pairs
Figure BDA0002315099850000073
Middle arbitrary subsequence S (F)i)kGo through
Figure BDA0002315099850000074
Respectively obtain S (F)i)kAnd
Figure BDA0002315099850000075
positive correlation of the collector sequences and Nε(S(Fi)k) And a set of negatively correlated collector subsequences Nε-(S(Fi)k)；
Figure BDA0002315099850000076
Figure BDA0002315099850000077
Wherein epsilon is a critical value of the site correlation coefficient,
Figure BDA0002315099850000078
is the degree of correlation between subsequences; based on the pearson correlation coefficient it can be known that,
Figure BDA0002315099850000079
middle, covariance
Figure BDA00023150998500000710
Variance (variance)
Figure BDA00023150998500000711
Standard deviation of
Figure BDA00023150998500000712
4.4) to N, respectivelyε(S(Fi)k) And Nε-(S(Fi)k) All sub-sequence traversals in
Figure BDA00023150998500000713
Respectively obtaining the related sub-sequence sets, and respectively adding the sets Nε(S(Fi)k) And Nε-(S(Fi)k) As shown in fig. 2;
4.5) repeat step 4.3), step 4.4) Up to set Nε(S(Fi)k) And Nε-(S(Fi)k) No new subsequence is added, and positive correlation clusters C are formed respectivelykAnd negative correlation cluster
Figure BDA00023150998500000714
If a cluster C of positive correlationkAnd negative correlation cluster
Figure BDA00023150998500000715
If the geographic position between the sites represented by the sub-sequences contained in the cluster meets a preset threshold value and the relevant time interval meets the actual significance, adding the cluster set: c ← Ck，
Figure BDA00023150998500000716
4.6)
Figure BDA00023150998500000717
Repeating the step 4.3), the step 4.4) and the step 4.5) to form a final positive correlation cluster C and a final negative correlation cluster C-Respectively representing all time periods with positive correlation and all time periods with negative correlation of passenger flow of two stations.
When the correlation coefficient is positive, the passenger flow volume increasing and decreasing trends among the stations are the same; when the correlation coefficient is negative, the passenger flow volume increase and decrease trends among the stations are opposite; when the correlation coefficient is zero, the passenger flow volume increasing and decreasing trend between the stations is uncertain;
and selecting the stations with positive or negative correlation coefficients, and adjusting the density of the bus route according to the increase and decrease trend of the passenger flow among the stations.
Examples
The historical card swiping data of two sites of the pond side and the aluminum cover of the mansion city are taken out, the card swiping data amount per day of a single line is more than 300, for example, 17 lines stop at the pond side, and the card swiping amount per day of 10 months and 10 days in 2018 is 5248. The card swiping data is composed of fields: line number, license plate number, transaction date, card swiping number, transaction time, transaction amount, train number, station number and driving direction. The card swiping data is shown in table 1.
Table 1: historical line card swiping data
Line number License plate number Date of transaction Card number for swiping card
651 Min DZ9525 20181010 11192285
Transaction time Number of vehicles Site numbering Direction of travel
25800 Min DZ5986 23 0
Taking out the pond side and the aluminum coffin at two stations in peak period 7:00-9: and (3) counting the card swiping amount according to the card swiping data of 30 at intervals of 10 minutes to obtain the passenger flow volume in each interval of the two stations. The peak time station passenger flow volume data are shown in table 2, fig. 3 and fig. 4.
Table 2: peak station passenger flow
Figure BDA0002315099850000081
And according to the initial length of 4-15 and the step length of 1, dividing the passenger flow time sequence in the peak period to obtain passenger flow time subsequences with all lengths. For example, the four subsequence steps for obtaining the pond side station passenger flow time subsequence with the length of 4 and the step size of 1 are shown in fig. 5, fig. 6, fig. 7 and fig. 8.
The sum of each length subsequence is calculated, and the obtained results are shown in fig. 9 and fig. 10.
And (3) evaluating the correlation between the subsequences with the same length of the two sites, and when the correlation is greater than 0, considering that the subsequences are positively correlated, and when the correlation is less than 0, considering that the subsequences are negatively correlated. According to the theory of correlation, when the threshold value of the degree of correlation is set to be 0.6, namely the degree of correlation r | > 0.6, the subsequences are considered to be correlated. The subsequences of the pond side and the screwing cap are shown in FIG. 11, FIG. 12, FIG. 13 and FIG. 14. It can be known that fig. 11 shows the time period of the positive correlation between the station pool side and the hiding house, and the bus line density between the station pool side and the hiding house can be increased in the positive correlation time period 0 and the positive correlation time period 5; fig. 12, 13, and 14 show the time periods when the station side is negatively associated with the aluminum charger, which reduces the bus line density between the station side and the aluminum charger.
The above examples are provided only for illustrating the present invention and are not intended to limit the present invention. Changes, modifications, etc. to the above-described embodiments are intended to fall within the scope of the claims of the present invention as long as they are in accordance with the technical spirit of the present invention.

Claims (6)

1. A public transport route adjusting method based on historical passenger flow is characterized in that the passenger flow correlation among stops is obtained, and the density of a public transport route passing through the stops is adjusted based on the passenger flow correlation among the stops;
the steps of obtaining the passenger flow volume correlation between the stations are as follows:
1) counting historical card swiping data of all bus stops according to time intervals to form a passenger flow time sequence;
2) acquiring a passenger flow time subsequence by using a multi-scale sliding window;
3) calculating the sum of the passenger flow time subsequences;
4) computing arbitrary two sites Fi、FjCorrelation coefficient of passenger flow time subsequences;
when the correlation coefficient is positive, the passenger flow volume increase and decrease trends among the stations are the same; when the correlation coefficient is negative, the passenger flow volume increase and decrease trends among the stations are opposite; when the correlation coefficient is zero, the passenger flow volume increasing and decreasing trend between the stations is uncertain;
selecting stations with positive or negative correlation coefficients, and adjusting the density of the bus route according to the passenger flow increasing and decreasing trend among the stations;
the step 1) is specifically as follows:
1.1) dividing time intervals for any site according to fixed time intervals, counting historical card swiping data of each site in each time interval, and forming a passenger flow time sequence F ═(F)1,f2,f3,…,fn)；
1.2) repeating the step 1.1), and acquiring a passenger flow time sequence set D of all the stations;
the step 2) is specifically as follows:
2.1) extracting a passenger flow time subsequence S (F) from the time sequence F e D with the length of a sliding window being l and the step length index of the sliding window being 1k＝(fk,fk+1,…,fk+l-1)，k∈(1≤k≤n-l+1)；
2.2) repeating step 2.1) until all passenger flow time subsequence sets of the passenger flow time sequence F are obtained
Figure FDA0003582358330000011
2.3) repeating the step 2.1) and the step 2.2) until a subsequence set SD of all sequences in the passenger flow time sequence set D is obtained.
2. The method of claim 1, wherein the time of day is divided into peak time and non-peak time, the time interval of the peak time is shorter than the time interval of the non-peak time, and the length of the sliding window of the peak time is longer than the length of the sliding window of the non-peak time.
3. The method for adjusting the bus route based on the historical passenger flow volume according to claim 2, wherein the step 3) is specifically as follows:
3.1) calculating the sum of the passenger flow when the length of the subsequence is l as follows:
Figure FDA0003582358330000021
3.2) calculating the sum of the passenger flow when the length of the subsequence is l +1 as follows:
Suml+1(S(F)k)＝Suml(S(F)k)+fk+1。
4. the method for adjusting the bus route based on the historical passenger flow according to claim 3, wherein the step 4) is specifically as follows:
4.1) initializing positively correlated clusters
Figure FDA0003582358330000022
Initializing negative correlation clusters
Figure FDA0003582358330000023
4.2) separately calculating
Figure FDA0003582358330000024
All of the subsequences S (F)i)k、S(Fj)kAverage value of (2)
Figure FDA0003582358330000025
The method comprises the following specific steps:
Figure FDA0003582358330000026
Figure FDA0003582358330000027
4.3) pairs
Figure FDA0003582358330000028
Middle arbitrary subsequence S (F)i)kGo through
Figure FDA0003582358330000029
Respectively obtain S (F)i)kAnd
Figure FDA00035823583300000210
positive correlation of the collector sequences and Nε(S(Fi)k) And a set of negatively correlated collector subsequences Nε-(S(Fi)k)；
Figure FDA00035823583300000211
Figure FDA00035823583300000212
Wherein epsilon is a critical value of the site correlation coefficient,
Figure FDA00035823583300000213
is the degree of correlation between subsequences;
4.4) to N, respectivelyε(S(Fi)k) And Nε-(S(Fi)k) All sub-sequence traversals in
Figure FDA00035823583300000214
Respectively obtaining the related sub-sequence sets, and respectively adding the sets Nε(S(Fi)k) And Nε-(S(Fi)k) Performing the following steps;
4.5) repeating step 4.3), step 4.4) until the set N is reachedε(S(Fi)k) And Nε-(S(Fi)k) No new subsequence is added, and positive correlation clusters C are formed respectivelykAnd negative correlation cluster
Figure FDA00035823583300000215
If a cluster C is positively correlatedkAnd negative correlation cluster
Figure FDA00035823583300000216
The geographical positions among the sites represented by the subsequences contained in the cluster are in accordance with a preset threshold value, and the related time intervals are in accordance with actual meanings, then the cluster set is added: c ← Ck，
Figure FDA0003582358330000031
4.6)
Figure FDA0003582358330000032
Repeating the step 4.3), the step 4.4) and the step 4.5) to form a final positive correlation cluster C and a final negative correlation cluster C-Respectively representing all time periods with positive correlation and all time periods with negative correlation of passenger flow of two stations.
5. The method for regulating bus routes based on historical passenger flow according to claim 4, wherein in step 4.3), the covariance
Figure FDA0003582358330000033
6. The method of claim 4, wherein the positive correlation cluster C and the negative correlation cluster C are used for adjusting the bus route based on the historical passenger flow-Respectively representing line stations and ride stations with the same riding behavior trendThe vehicles behave as route stations with opposite trends.
CN201911274222.1A 2019-12-12 2019-12-12 Bus route adjusting method based on historical passenger flow Active CN111126681B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911274222.1A CN111126681B (en) 2019-12-12 2019-12-12 Bus route adjusting method based on historical passenger flow

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911274222.1A CN111126681B (en) 2019-12-12 2019-12-12 Bus route adjusting method based on historical passenger flow

Publications (2)

Publication Number Publication Date
CN111126681A CN111126681A (en) 2020-05-08
CN111126681B true CN111126681B (en) 2022-06-07

Family

ID=70499909

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911274222.1A Active CN111126681B (en) 2019-12-12 2019-12-12 Bus route adjusting method based on historical passenger flow

Country Status (1)

Country Link
CN (1) CN111126681B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112465396B (en) * 2020-12-15 2023-04-18 中北大学 Bus scheduling method and system based on station events along line
CN113177681A (en) * 2021-03-15 2021-07-27 蓝海（福建）信息科技有限公司 Intercity travel passenger vehicle line correlation identification method
CN117575684B (en) * 2024-01-15 2024-04-05 杭州路过网络有限公司 Passenger flow volume prediction method and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105868861A (en) * 2016-04-08 2016-08-17 青岛海信网络科技股份有限公司 Bus passenger flow evolution analysis method based on time-space data fusion
CN107316108A (en) * 2017-06-19 2017-11-03 华南理工大学 A kind of citizens' activities public bus network chooses sliding window multiple features Forecasting Methodology
CN109558985A (en) * 2018-12-10 2019-04-02 南通科技职业学院 A kind of bus passenger flow amount prediction technique based on BP neural network
KR20190076267A (en) * 2017-12-22 2019-07-02 한국교통연구원 Method and Apparatus for Dynamically Determining Bus Service Route

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105868861A (en) * 2016-04-08 2016-08-17 青岛海信网络科技股份有限公司 Bus passenger flow evolution analysis method based on time-space data fusion
CN107316108A (en) * 2017-06-19 2017-11-03 华南理工大学 A kind of citizens' activities public bus network chooses sliding window multiple features Forecasting Methodology
KR20190076267A (en) * 2017-12-22 2019-07-02 한국교통연구원 Method and Apparatus for Dynamically Determining Bus Service Route
CN109558985A (en) * 2018-12-10 2019-04-02 南通科技职业学院 A kind of bus passenger flow amount prediction technique based on BP neural network

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
铁路旅客流量规律及预测;胡早鑫等;《内江科技》;20161025(第10期);全文 *

Also Published As

Publication number Publication date
CN111126681A (en) 2020-05-08

Similar Documents

Publication Publication Date Title
CN111126681B (en) Bus route adjusting method based on historical passenger flow
CN110288212B (en) Improved MOPSO-based electric taxi newly-built charging station site selection method
CN109753694A (en) Small and medium-sized cities Transit Network Design method based on overall process trip detecting period
CN107545730B (en) Bus IC card data-based station getting-on/off passenger number estimation method
CN106910018A (en) One kind rescue resource regulating method and system
CN108388970B (en) Bus station site selection method based on GIS
CN112070355A (en) Distribution scheduling method for airport ferry vehicle
CN112309119B (en) Urban traffic system capacity analysis optimization method
CN103956042A (en) Public bike scheduling area intelligent partition method based on graph theory
CN114021883A (en) Dispatching method for subway transfer shared bicycle in peak period
CN116611586B (en) Newly built road network flow prediction method and system based on double-layer heterogeneous network
CN111191817B (en) Bus network topology division method based on transfer passenger flow
CN110543976A (en) Charging station layout optimization method based on genetic algorithm
CN112150802A (en) Urban road grade division method based on ground bus running state reliability
CN108985616A (en) A kind of public transportation lane evaluation of layout method
CN111160722B (en) Bus route adjusting method based on passenger flow competition relationship
CN112749836B (en) Customized passenger transport intelligent transportation capacity allocation method based on passenger flow time sequence
CN113537596A (en) Short-time passenger flow prediction method for new line station of urban rail transit
CN117218836B (en) Urban planning project comprehensive management information platform based on smart city
CN112598328A (en) Optimization method and system for multi-target distribution of transfer boarding gates in satellite hall mode
CN115294773B (en) Bus lane optimal configuration method, electronic device and storage medium
CN107273703B (en) Pavement service performance distribution condition prediction method
CN113723459B (en) Bus operation time interval division and one-way point recommendation method based on time aggregation
AU2021102278A4 (en) LSTM-based method and apparatus for forecasting the demand for DiDi orders
CN112509357A (en) Bus departure time interval optimization method based on genetic algorithm

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