CN114358386A - Double-trip-mode ride-sharing site generation method based on reserved trip demand - Google Patents
Double-trip-mode ride-sharing site generation method based on reserved trip demand Download PDFInfo
- Publication number
- CN114358386A CN114358386A CN202111487717.XA CN202111487717A CN114358386A CN 114358386 A CN114358386 A CN 114358386A CN 202111487717 A CN202111487717 A CN 202111487717A CN 114358386 A CN114358386 A CN 114358386A
- Authority
- CN
- China
- Prior art keywords
- passenger
- travel
- initial
- ride
- information
- 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.)
- Pending
Links
Images
Abstract
The invention provides a method for generating a double-trip mode ride-sharing station based on a reserved trip demand, which comprises the steps of firstly collecting reserved trip information of passengers, and selecting the passengers with similar trip paths as an initial passenger group according to the trip demand of the passengers; then performing space-time clustering analysis on the boarding place and the trip time of the initial passenger group to determine initial ride-sharing site information; then, according to the selection of the commuting mode of the passengers, a double-trip mode co-taking initial scheme with two vehicles is formed; and finally, issuing a ride-sharing initial scheme to the passenger, confirming whether the information of ride-sharing and driving is participated in, arranging the information of the vehicle type, the vehicle number and the route according to the feedback information of the passenger, generating a ride-sharing station, and issuing the information of the ride-sharing station to the passenger. The invention can improve the use efficiency of the vehicle and can improve the use efficiency of the vehicle.
Description
Technical Field
The invention belongs to the technical field of intelligent public transportation, and particularly relates to a method for generating a double-trip-mode car-sharing station based on a reserved trip demand.
Background
In recent years, with the continuous acceleration of urbanization process in China, the quantity of automobiles kept in China is rapidly increased, and urban infrastructure is difficult to meet the travel requirements in the peak travel time period, especially in the early peak and late peak time periods of working days. In large and medium-sized cities in China, the phenomenon of traffic congestion is increasingly serious, inconvenience is brought to the life of people, social production cost is improved, and adverse effects are caused on economic development and urbanization construction in China. Meanwhile, only one driver exists in many private commuting cars, and in the non-trip peak period, the public transport vehicles have serious idle load phenomenon, so that the resource waste is caused. The phenomenon shows that a method is urgently needed to reasonably utilize traffic resources and relieve the current situation of urban traffic jam.
With the rise of the customized public transportation and online booking industry, the trip mode of car-assembling type is more and more favored by travelers. As far as the present in China, 29 provinces such as Beijing, Hebei, Shanxi and the like open customized public transportation services, and the number of operation lines is 5400, and the annual passenger capacity is close to 1.8 hundred million people. The travel modes such as net car booking and windward driving are also one of the main travel modes of people. Foreign scholars mostly concentrate on analyzing the spatial position of the bus stop and determine the position of the customized bus stop by using a clustering algorithm; at present, the research on a synthetic station is mainly based on hierarchical clustering, a DBSCAN algorithm and a K-means algorithm to perform clustering analysis on spatial geographic positions to obtain coordinate points, the research on generating customized bus stations is developed by analyzing and utilizing GPS data, SP (service provider) and RP (remote protocol) survey data and network appointment data, and abundant research is performed on the aspects of path planning, carpooling information matching and the like; however, by integrating the domestic travel commuting characteristics, the above research results only consider using the bus as a single vehicle, and do not discuss the ride-sharing problem under the dual modes of private cars and buses, and the selection of ride-sharing stations lacks importance on time, and mostly focuses on the cluster analysis of spatial positions.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a method for generating a double-trip-mode ride-sharing station based on a reserved trip demand, which can improve the use efficiency of vehicles, relieve road congestion and improve the trip experience of passengers.
The present invention achieves the above-described object by the following technical means.
A double-trip-mode ride-sharing site generation method based on reserved trip demands specifically comprises the following steps:
collecting reserved travel information of passengers, and selecting the passengers with similar travel paths as an initial passenger group according to travel requirements of the passengers;
performing space-time clustering analysis on boarding places and trip times of initial passenger groups to determine initial ride-sharing site information; according to the selection of the commuting mode of the passengers, a double-trip mode co-taking initial scheme with two vehicles is formed; and issuing a ride-sharing initial scheme to the passenger, confirming whether the information of ride-sharing and driving is participated in, arranging the information of the vehicle type, the vehicle number and the route according to the feedback information of the passenger, generating a ride-sharing station, and issuing the ride-sharing station information to the passenger.
Further, the travel demand of the passenger comprises the longitude and latitude coordinates lo of the departure point coordinate of each passenger iiLongitude and latitude coordinates of the destination coordinate ldiTime to end TdiStandard passenger capacity Q of whether or not it is willing to drive, and if it is willing to drive, the vehicle it is supposed to drivei。
Further, the travel path similarity calculation formula is as follows:
wherein: p denotes the last stop of the initial path of the first passenger, q denotes the last stop of the initial path of the second passenger, Head (tr)i) Indicates the first station in the initial travel path of passenger i, Rest (tr)i) A subsequence consisting of all stations except the first point of the initial travel path of the passenger i is represented;
the initial travel path is as follows:
wherein: p is a radical ofiAnd (lat, lng, t) represents that the passenger is located at the geographic coordinate position { lat, lng) at the time point t, and lat and lng respectively represent the latitude and longitude of the bus stop.
Further, the double-trip mode ride-sharing initial scheme specifically includes: judging whether to drive a private car according to the information of the co-passenger, counting the number of passengers except for the driver to obtain the number C of vacant seats of the private car if the co-passenger drives the private car, and counting the total number V of non-driving passengers if the co-passenger does not drive the private car; judging whether C is larger than V, when C is larger than V, not increasing the passenger car, and planning the private car driving line by utilizing Dijkstra algorithm according to the preliminary trip line, otherwise, selecting the passenger car type according to the difference value, and formulating the passenger car line by utilizing Dijkstra algorithm according to the preliminary trip line.
Further, the commute mode includes private cars and buses.
Further, the ride-sharing site information includes position information, departure time, and vehicle information on which each passenger rides.
Further, clustering information extraction is carried out on the initial group of the carpooled passengers, and the lo is selectediTravel time ti, generating passenger information set pi：pi＝{loi，ti}；
Constructing a loss function J (u, r):
wherein: uk is the clustering center;
when the passengers are divided into K clusters, the corresponding loss function is noted as DKAnd calculating: gap (k) ═ E (logD)k)-logDkWherein, E (logD)k) Is logDk(iii) a desire;
when the maximum value is obtained by gap (K), the corresponding K value is the optimal classification cluster number, and an initial co-multiplication site set { u/u } is generated by using K-means clustering according to the K valuei}; each initial ride-sharing site comprises a geographic coordinate clustering center lsiTravel time demand clustering center tsiAnd co-passenger information xsiWherein x issiThe method belongs to X, and X is a set of travel demands of the whole passengers in the region.
The invention has the beneficial effects that:
(1) according to the selection of the commuting mode of the passenger, the invention forms a double-travel mode ride-sharing initial scheme with two vehicles, and the double-travel mode ride-sharing initial scheme specifically comprises the following steps: judging whether to drive a private car according to the information of the co-passenger, counting the number of passengers except for the driver to obtain the number C of vacant seats of the private car if the co-passenger drives the private car, and counting the total number V of non-driving passengers if the co-passenger does not drive the private car; judging whether C is larger than V, when C is larger than V, not increasing the passenger car, and planning the private car driving line by utilizing Dijkstra algorithm according to the preliminary trip line, otherwise, selecting the passenger car type according to the difference value, and formulating the passenger car line by utilizing Dijkstra algorithm according to the preliminary trip line. According to the initial scheme of the double-trip-mode ride sharing, the use efficiency of vehicles is improved, and road congestion is relieved;
(2) the method and the device extract the clustering information of the initial group of the co-passengers, determine the information of the initial co-passenger station, ensure that the walking distance from the whole co-passenger to the bus station is shortest and the waiting time is shortest, and improve the traveling experience of the passengers.
Drawings
Fig. 1 is a flow chart of a method for generating a double-trip-mode ride station based on a reserved trip demand according to the present invention;
fig. 2 is a flow chart of vehicle model matching according to the present invention.
Detailed Description
The invention will be further described with reference to the following figures and specific examples, but the scope of the invention is not limited thereto.
As shown in fig. 1, a method for generating a double-trip-mode ride station based on a reserved trip demand specifically includes the following steps:
step (1), node division is carried out on the road network of the selected area, and reserved travel information is gathered to passengers in the selected area
Step (1.1), node division is carried out on a road network in a selected area, a road network layer is established according to the existing roads, and road network nodes are established according to the existing bus stops on the road network, wherein the road network nodes comprise longitude and latitude coordinates of the bus stops and lane driving directions; labeling the road network nodes to generate a road network node set { D };
step (1.2), the longitude and latitude coordinates lo of the departure point coordinates of each passenger i in the selected areaiLongitude and latitude coordinates of the destination coordinate ldiTime to end TdiWhether or not it is willing to drive, ifStandard passenger capacity Q of vehicle willing to driveiCounting is carried out;
step (1.3), constructing a whole passenger trip demand set in the region
X＝{x1，x2，···，xn} (1)
Wherein: n is the number of passengers with travel demands;
demand x of a passenger in the passenger travel demand setiThe sample feature vector of (i ∈ (1, 2,. n)) is:
xi＝{loi，ldi，ti} (2)
sample characteristics include the spatial coordinates of start and end points (including lo)iAnd ldi) And a trip time ti (based on the time to end Td)iObtained by reverse interpolation).
Step (1.4), according to Dijkstra algorithm (Dixtera algorithm), according to the coordinates lo of the departure point and the destination point of a certain passenger ii、ldiPreliminarily planning the path of the passenger i and calculating the initial trip path tr of the passenger iiInitial travel route triThe method comprises the steps that nodes in a road network node set { D }; initial travel route of a certain passenger i
Wherein p isiAnd (lat, lng, t) represents the geographic coordinate position of the passenger i at the time t, and lat and lng respectively represent the latitude and longitude of the bus stop.
Step (2), selecting the passengers with higher coincidence degree of the primary travel path as an initial ride-sharing passenger group
The statistical method of the initial group of carpooling passengers is as follows: according to the initially planned passenger initial travel path, similarity of the passenger initial travel path is calculated by using a Dynamic Time Warping (DTW) algorithm, and passengers with similar travel paths are selected to form an initial group of co-passenger passengers.
The similarity calculation process is shown in formula (3):
wherein: p denotes the last stop of the initial path of one of the passengers, q denotes the last stop of the initial path of the other passenger, Head (tr)i) Indicates the first station in the initial travel path of passenger i, Rest (tr)i) A subsequence consisting of all stations except the first point of the initial travel path of the passenger i is represented;
step (3), performing space-time clustering analysis on boarding places and travel times of the initial pool passenger groups to determine initial pool station information
Step (3.1) of extracting clustering information of the initial group of passengers co-joined, and selecting the departure point coordinate lo of each passenger iiTravel time ti, and generating passenger information set p required by K-means cluster analysisiAs shown in equation (4).
pi＝{loi，ti} (4)
Step (3.2), assume u1, u 2.., uk as K cluster centers, rnkE 0, 1 represents piAnd (4) constructing a loss function J (u, r) according to the formula (5) if the K cluster with uk as the clustering center belongs to the K cluster.
When the passengers are divided into K clusters, the corresponding loss function is noted as DKWhen the Gap (K) has the maximum value according to the Gap statistical quantity, the corresponding K value is the optimal number of classification clusters, and the calculation method is shown in formula (6):
Gap(K)＝E(logDk)-logDk (6)
wherein, E (logD)k) Is logDkThe expectations of (1) are typically generated using monte carlo simulations.
Step (3.3), clustering by using K-means according to the K value in the step (3.2) to generate an initial co-multiplying site set { u } ui}; each initial ride-share site contains a cluster of geographic coordinatesCenter lsiTravel time demand clustering center tsiAnd co-passenger information xsiWherein x issi∈X。
And (4) forming a double-trip-mode ride-sharing initial scheme with two vehicles according to the selection of the commuting mode of the passengers.
As shown in fig. 2, whether to drive a private car is judged according to the information of the co-passenger, if the co-passenger drives the private car, the number of passengers except the driver is counted to obtain the number of vacant seats of the private car C, the calculation method of C is shown in formula (7), and if the co-passenger does not drive the private car, the total number V of non-driving passengers (having travel demand but not driving the private car) is counted.
Wherein m is the number of people who choose to drive the private car to go out.
Judging whether C is larger than V, when C is larger than V, not increasing the passenger car, and planning the private car driving line by utilizing Dijkstra algorithm according to the preliminary trip line, otherwise, selecting the passenger car type according to the difference value, and formulating the passenger car line by utilizing Dijkstra algorithm according to the preliminary trip line.
Step (5), passenger-oriented issuing of ride-sharing information, passenger feedback information collection, and generation of ride-sharing station
Issuing an initial plan of ride pooling for passengers, determining whether to participate in ride pooling and driving information, arranging vehicle types, vehicle numbers and route information according to passenger feedback information, and generating ride pooling stations; and issuing the information of the ride-sharing station to the passengers, wherein the information comprises position information, departure time and vehicle information taken by each passenger, and generating a final ride-sharing station.
The present invention is not limited to the above-described embodiments, and any obvious improvements, substitutions or modifications can be made by those skilled in the art without departing from the spirit of the present invention.
Claims (7)
1. A double-trip-mode ride-sharing site generation method based on reserved trip demands is characterized by comprising the following steps:
collecting reserved travel information of passengers, and selecting the passengers with similar travel paths as an initial passenger group according to travel requirements of the passengers;
performing space-time clustering analysis on boarding places and trip times of initial passenger groups to determine initial ride-sharing site information; according to the selection of the commuting mode of the passengers, a double-trip mode co-taking initial scheme with two vehicles is formed; and issuing a ride-sharing initial scheme to the passenger, confirming whether the information of ride-sharing and driving is participated in, arranging the information of the vehicle type, the vehicle number and the route according to the feedback information of the passenger, generating a ride-sharing station, and issuing the ride-sharing station information to the passenger.
2. The method of claim 1, wherein the travel demand of the passenger comprises a longitude and latitude coordinate lo of a departure point coordinate of each passenger iiLongitude and latitude coordinates of the destination coordinate ldiTime to end TdiStandard passenger capacity Q of whether or not it is willing to drive, and if it is willing to drive, the vehicle it is supposed to drivei。
3. The method for generating a double-travel-mode ride station based on reserved travel demands according to claim 1, wherein the travel path similarity calculation formula is as follows:
wherein: p denotes the last stop of the initial path of the first passenger, q denotes the last stop of the initial path of the second passenger, Head (tr)i) Indicates the first station in the initial travel path of passenger i, Rest (tr)i) A subsequence consisting of all stations except the first point of the initial travel path of the passenger i is represented;
the initial dischargeThe row path is:
wherein: p is a radical ofiAnd (lat, lng, t) represents that the passenger is located at the geographic coordinate position (lat, lng) at the time point t, and the lat and the lng respectively represent the latitude and the longitude of the bus stop.
4. The double-travel-mode ride station generation method based on reserved travel demand according to claim 1, wherein the double-travel-mode ride initial plan specifically includes: judging whether to drive a private car according to the information of the co-passenger, counting the number of passengers except for the driver to obtain the number C of vacant seats of the private car if the co-passenger drives the private car, and counting the total number V of non-driving passengers if the co-passenger does not drive the private car; judging whether C is larger than V, when C is larger than V, not increasing the passenger car, and planning the private car driving line by utilizing Dijkstra algorithm according to the preliminary trip line, otherwise, selecting the passenger car type according to the difference value, and formulating the passenger car line by utilizing Dijkstra algorithm according to the preliminary trip line.
5. The reserved travel demand-based double travel mode ride station generation method of claim 1, wherein the commute mode comprises a private car and a bus.
6. The dual travel mode pool stop generation method based on reserved travel demand according to claim 1, wherein the pool stop information includes location information, departure time, and vehicle information on which each passenger rides.
7. The reserved travel demand-based double travel mode pool station generation method according to claim 2, wherein:
extracting clustering information of initial group of passengers, selecting the loiTravel time ti, generating passenger information set pi：pi＝{loi，ti}；
Constructing a loss function J (u, r):
wherein: uk is the clustering center;
when the passengers are divided into K clusters, the corresponding loss function is noted as DKAnd calculating: gap (k) ═ E (logD)k)-logDkWherein, E (logD)k) Is logDk(iii) a desire;
when the maximum value is obtained by gap (K), the corresponding K value is the optimal classification cluster number, and an initial co-multiplication site set { u/u } is generated by using K-means clustering according to the K valuei}; each initial ride-sharing site comprises a geographic coordinate clustering center lsiTravel time demand clustering center tsiAnd co-passenger information xsiWherein x issiThe method belongs to X, and X is a set of travel demands of the whole passengers in the region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111487717.XA CN114358386A (en) | 2021-12-07 | 2021-12-07 | Double-trip-mode ride-sharing site generation method based on reserved trip demand |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111487717.XA CN114358386A (en) | 2021-12-07 | 2021-12-07 | Double-trip-mode ride-sharing site generation method based on reserved trip demand |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114358386A true CN114358386A (en) | 2022-04-15 |
Family
ID=81098396
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111487717.XA Pending CN114358386A (en) | 2021-12-07 | 2021-12-07 | Double-trip-mode ride-sharing site generation method based on reserved trip demand |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114358386A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115186049A (en) * | 2022-09-06 | 2022-10-14 | 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 |
| CN115587657A (en) * | 2022-10-19 | 2023-01-10 | 华中科技大学 | Station determining and route optimizing method for night customized bus |
-
2021
- 2021-12-07 CN CN202111487717.XA patent/CN114358386A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115186049A (en) * | 2022-09-06 | 2022-10-14 | 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 |
| 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 |
| CN115587657A (en) * | 2022-10-19 | 2023-01-10 | 华中科技大学 | Station determining and route optimizing method for night customized bus |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109409599B (en) | Customized bus line opening optimization method based on real-time requirements | |
| CN107036617B (en) | Travel route planning method and system combining taxi and subway | |
| CN105844362B (en) | Urban traffic comprehensive trip decision-making device | |
| Zamir et al. | Effects of transit-oriented development on trip generation, distribution, and mode share in Washington, DC, and Baltimore, Maryland | |
| CN109612488B (en) | Big data micro-service-based mixed travel mode path planning system and method | |
| CN114358386A (en) | Double-trip-mode ride-sharing site generation method based on reserved trip demand | |
| CN110222786B (en) | Dynamic car pooling method and system based on travel information | |
| CN104881710A (en) | Urban express delivering method based on vehicle self-organized network | |
| CN112561249B (en) | Real-time demand-oriented city customized bus scheduling method | |
| CN110414795B (en) | Newly-increased high-speed rail junction accessibility influence method based on improved two-step mobile search method | |
| CN112036757A (en) | Parking transfer parking lot site selection method based on mobile phone signaling and floating car data | |
| CN110909434A (en) | Urban public transport trunk line network design method under low-carbon guidance | |
| CN106373384A (en) | Remote area passenger transport regular bus route real-time generation method | |
| Song et al. | Exploring pareto routes in multi-criteria urban bicycle routing | |
| Wang et al. | Multi-objective optimization of customized bus routes based on full operation process | |
| CN114511143A (en) | Urban rail transit network generation method based on grouping division | |
| CN110288198B (en) | Method for measuring bearing capacity of rental bicycle transportation facility based on traffic subareas | |
| CN109523064B (en) | Intelligent micro-hub based on multi-network fusion | |
| CN106130110B (en) | The electric taxi charging station constant volume method on trip ground is selected based on stratified probability | |
| CN113096429B (en) | Elastic bus area flexibility line generation method based on bus dispatching station distribution | |
| CN110245774A (en) | A method of regular service route optimization is carried out according to employee's home address | |
| CN116090785B (en) | Custom bus planning method for two stages of large-scale movable loose scene | |
| Bao et al. | Spatiotemporal clustering analysis of shared electric vehicles based on trajectory data for sustainable urban governance | |
| Zhang et al. | The research on planning of taxi sharing route and sharing expenses | |
| CN107527105A (en) | Carpooling order combining method |
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 |