CN109523064B - Intelligent micro-hub based on multi-network fusion - Google Patents
Intelligent micro-hub based on multi-network fusion Download PDFInfo
- Publication number
- CN109523064B CN109523064B CN201811258830.9A CN201811258830A CN109523064B CN 109523064 B CN109523064 B CN 109523064B CN 201811258830 A CN201811258830 A CN 201811258830A CN 109523064 B CN109523064 B CN 109523064B
- Authority
- CN
- China
- Prior art keywords
- micro
- data
- user
- taxi
- 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.)
- Active
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
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention particularly relates to an intelligent micro-hub based on multi-network fusion, which integrates information and facility intelligent comprehensive hubs comprising rail transit, buses, taxis, shared single cars and other transportation modes along the line. The method comprises the steps of carrying out information analysis processing on bus operation data, taxi operation data and shared bicycle release data by establishing a micro-junction information system, providing transfer information for users, and generating an operation scheduling optimization scheme for an operator to adjust an operation scheme by analyzing the user transfer data. And realizing the intensive land layout of the micro hub by optimizing the operation scheduling. The invention creates a comprehensive data processing platform fusing data of various different traffic modes by establishing a novel intelligent micro-hub system, establishes a corresponding intelligent information processing system of a machine learning and deep learning model taking an algorithm as a core, and finally achieves the purposes of land occupation concentration, continuous improvement of user transfer efficiency and continuous optimization of passenger flow distribution efficiency of the whole micro-hub.
Description
Technical Field
The invention belongs to the field of transportation hubs, and particularly relates to a micro-hub integrated data processing platform which can fuse and anchor various transportation modes such as rail transit, ground public transportation, taxis, non-motor vehicles and the like, simultaneously furthest reduces occupied land resources outside a road infrared line, builds a micro-hub information center, and realizes the continuous efficient transfer of passengers in the micro-hub by means of a corresponding algorithm.
Background
The passenger transport transportation hub is a complex of various transportation facilities (including lines, stations, vehicles, information and the like) which are used for handling passenger transfer, sending and arrival at transportation nodes and providing a parking space for the transportation nodes.
According to the comprehensive passenger transport transportation hub layout planning of Shanghai city, the layout of rail transportation stations is used as a basic framework, 145 city-level comprehensive passenger transport transportation hubs are formed in the whole city by means of transportation facilities such as airports, railways, highways, ground buses and the like, and the passenger transport transportation hubs in the city are classified into A, B, C, D types according to the transportation functions and scale born by the hubs.
Class A hub: large-scale external transportation facilities such as aviation, railways and the like are taken as main facilities, and urban transportation facilities such as rail transit stations, ground bus stations, social parking lots, taxi stations and the like are arranged in a matched manner to form internal and external transportation hubs;
class B hub: an urban comprehensive transportation hub which is mainly formed by a rail transit station and is combined with other transportation facilities such as ground bus stations, social parking lots (storehouses), taxi stations, long-distance bus stations and the like;
class C hub: a P + R junction for providing preferential transfer for large and medium-sized social parking lots is arranged in the region near the outer ring and outside the outer ring and near the main road and rail transit stations for properly intercepting motor vehicles entering the city
Class D hub: the multi-line ground bus first and last stations are intensively distributed in the region far away from the rail transit station, and the small hub mainly comprising the pure ground bus transfer station is changed in the next year.
The four types of hubs are characterized by occupying the land outside the road, or being independently constructed or comprehensively developed. However, due to the limitation of shortage of land resources, the construction of passenger transportation hubs is generally difficult to implement, the project amount is large, the construction period is long, the lack of regional bus transfer capacity cannot be quickly and effectively relieved, various public transportation modes are not smoothly connected, and the bus attraction is reduced.
The micro junction is a small junction which is different from the conventional concept of a traffic junction, is a collection of traffic flow, information flow and people flow in space, does not occupy or occupies little land resources outside a red line of a road, only organically combines a rail station, a bus station, a taxi station, a non-motor vehicle parking point and the like, and realizes the connection and transfer among different modes, different directions and different levels, thereby providing traffic facilities for passengers for sending, transferring, standardizing various vehicle parking and integrating intelligent information functions.
Disclosure of Invention
The invention aims to provide an intelligent micro hub based on multi-network fusion, which utilizes the least land resources to realize the maximum bus transfer, distribution, parking and guidance benefits, really realizes seamless connection, multi-network fusion anchoring and intelligent efficient transfer from the aspect of service passenger and facility connection, improves the public transportation attraction by improving the operation efficiency of regional public transportation, and finally realizes the bus priority development strategy.
In order to achieve the purpose, the invention adopts the technical scheme that: an intelligent micro junction based on multi-network fusion integrates transfer modes among rail transit, buses, taxis and shared buses to form an urban public transit micro junction, integrates information of a bus subsystem, a taxi subsystem and a shared bus subsystem by establishing a micro junction information center, provides transfer information for users, and provides transfer information for users for a bus platform, a taxi platform and a shared bus platform for operation capacity adjustment.
Further, the micro-junction information center inquires real-time information of the bus platform, the taxi platform and the shared single-car platform from each subsystem, estimates the possibility that a user successfully reserves a network appointment according to the real-time information of the shared single-car platform, selects different transfer modes by the user, updates and updates a micro-junction information center database according to the outbound condition of the user, estimates the personnel distribution of each transfer mode according to a statistical algorithm by utilizing historical data in the database, and provides the estimated data for the corresponding bus platform, the taxi platform and the shared single-car platform.
Further, the real-time information of the bus platform, the taxi platform and the shared single-bus platform comprises the real-time bus stop time of the bus platform, the real-time stop condition of arriving taxies and the real-time supply condition of shared single-buses.
Furthermore, the statistical algorithm of rail-switched buses adopts a sample distribution algorithm as a core, and the information of the personnel who transfer the buses and the card swiping information of the buses along each route are selected from the micro-junction information center and are provided for the bus operator to optimize after being counted by the sample distribution algorithm.
Furthermore, a statistical algorithm for rail-exchanging taxi taking adopts a poisson algorithm as a core, estimates the distribution of taxi transfer personnel according to the data of the micro-hub information center database based on poisson distribution, provides the taxi platform with the taxi driver for decision making, and updates the real-time taxi parking situation according to the decision result of the taxi driver.
Furthermore, a statistical algorithm of the rail-switched ride sharing single vehicle adopts a binomial distribution algorithm as a core, provides the possibility that the user successfully reserves the network car appointment according to the data of the database of the micro-hub information center based on binomial distribution, and provides the possibility for the user and the sharing single vehicle platform at the same time.
Further, the traffic micro-hub is divided into two types, namely a straight type and an estuary type: the linear type, namely a taxi station and a bus station are arranged on a slow lane and are arranged by using the space of a non-motor lane; the harbor type is characterized in that a concave part is arranged in a sidewalk and used for setting a taxi station and a bus station, or the sidewalk is partially outwards protruded and is provided with the taxi station and the bus station along the convex part.
The micro junction is a collection of traffic flow, information flow and people flow in space, does not occupy or occupies little land resources outside the red line of the road, and provides transportation facilities for passengers to send and transfer, standardize various vehicle parking and integrate intelligent information. By establishing a micro-junction information center, creating a comprehensive data processing platform (taking rail transit, public transportation, shared bicycle and taxi as examples) combining various transfer traffic mode data, establishing a corresponding intelligent information processing system of a machine learning and deep learning model taking an algorithm as a core, continuously self-optimizing along with the accumulation of data, predicting more and more accurately along with the time, dispatching and distributing buses more efficiently based on the personalized transfer requirements of users, sharing the transport capacity of the bicycle and the taxi, and finally achieving the aims of continuously improving the time efficiency of the four transportation means transfer modes of the users and continuously optimizing the passenger flow distribution efficiency of the whole micro-junction. The invention can realize the linked transfer of various traffic modes by using the minimum land resources, and provides a novel intelligent micro hub with the shortest transfer distance and the minimum transfer time for passengers.
Drawings
FIG. 1 is a general block diagram of an intelligent micro-hub of the present invention.
Fig. 2 is an assistant decision process of rail-switching by bus.
Fig. 3 is an aid decision flow for a rail-swap taxi ride.
Fig. 4 is an assistant decision flow of the rail-switched shared vehicle.
Detailed Description
The invention aims to provide an intelligent micro hub based on multi-network fusion, which utilizes the least land resources to realize the maximum bus transfer, distribution, parking and guidance benefits, really realizes seamless connection, multi-network fusion anchoring and intelligent efficient transfer from the aspect of service passenger and facility connection, improves the public transportation attraction by improving the operation efficiency of regional public transportation, and finally realizes the bus priority development strategy. On one hand, the micro hub and the passenger transport hub are interdependent, have similar functions with the passenger transport hub, are continuation and supplement of the passenger transport hub, and guide and drive the development of peripheral areas in a micro hub construction mode under the condition that the passenger transport hub is constructed without independent land in the urban construction area range; on the other hand, the major difference between the micro hub and the passenger transport hub is represented in the occupation scale of land outside the red line of the road, the passenger transport hub needs to completely occupy the urban construction land outside the red line, and the micro hub mainly takes the land used in the red line of the road as the main point; in the third aspect, a comprehensive data processing center (taking rail transit, public transportation, shared bicycle and taxi as examples) combining various transfer traffic mode data is created by establishing a micro-junction information center, a machine learning and deep learning intelligent information processing model taking a corresponding algorithm as a core is established, continuous self-optimization is carried out along with the accumulation of historical data, the prediction accuracy is improved, the operation energy of different types of transportation tools is adjusted and utilized more efficiently based on the personalized requirements of users, and finally the purpose of continuously improving the transfer efficiency of the users is achieved.
The intelligent traffic micro-junction has the advantages that the reasonable transfer adjustment is carried out on the public transport modes, the taxi modes and the sharing single-car modes, the urban public traffic micro-junction is built, the matching connection of various public traffic modes is promoted, and the urban public traffic outgoing sharing rate is improved. Traffic micro-hubs fall into two categories, linear and estuary: the linear type, namely a taxi station/bus station is arranged on a slow lane and is arranged by using the space of a non-motor lane; the bay type, i.e., the station, is set up by compressing the width of a sidewalk or a non-motorized lane.
In principle, each entrance and exit of the newly-built and the existing railway stations is the optimal section of a micro-junction which is mutually connected in multiple modes, and if the entrance and exit are positioned on the same side of a road and are close to each other, the selection is carried out according to the direction of passenger transfer; under the condition of limited conditions, the traffic mode distribution conditions can be dispersedly arranged in combination, but corresponding sign signs are arranged in the rail station, the bus station, the taxi station and the shared single vehicle for guidance.
Taxi stations and bus stations connected with the rail transit station are mainly in a bay type, taxi stations and bus stations are required to be arranged, but the bay type can not be arranged, and a linear type can be arranged according to the situation; when the rail transit station exit and entrance is positioned at the intersection, the micro-junction is arranged by combining the rail transit exit and entrance at the downstream of the intersection; the non-motor vehicle parking points in the micro hub can be integrated with urban side corner zones or building retreat red lines; one end of a taxi station or a conventional bus station close to the entrance and exit of the track station is not more than 50 meters, and the farthest distance is not more than 100 meters under the condition of limited conditions; the parking scales of taxis, non-motor vehicles and buses are comprehensively determined by combining passenger flow prediction and peripheral conditions, the number of the taxi parking positions is generally not less than 2, when a plurality of taxis park, the parking position length is calculated according to 6m x n (n is the number of the taxi parking positions), and the number of the non-motor vehicle parking positions is generally not less than 50.
The parking scales of the taxi, the non-motor vehicle and the bus are comprehensively determined by combining passenger flow prediction and peripheral conditions, the number of the taxi berths is generally not less than 2, and the number of the non-motor vehicle parking berths is generally not less than 20; the added taxi stop form is consistent with the bus stop form; ensuring sufficient sidewalk space, and ensuring that the pedestrian passing width is not less than 4 m after the micro-junction is arranged; the overall coordination with the surrounding environment is emphasized, and the environment quality is improved.
The user can access the micro-junction information center where the destination is located through app application of the micro-junction mobile phone terminal integrating the bus card data, real-time operation conditions of the user and the destination are inquired by combining data platforms of different traffic modes, meanwhile, the micro-junction information center provides inquiry results and transfer suggestions of the different traffic modes for the user through different algorithms based on requirements of the user and the capacity of a specific supplier, and provides capacity adjustment decision suggestions for the supplier, so that the user and the supplier are helped to improve transfer efficiency. Meanwhile, data created by a user can enrich a historical overall database and is fed back to the micro-hub information center to help optimize a corresponding machine learning model, so that the service capability of the micro-hub information center is continuously improved.
The user can access the micro-junction information center where the destination is located through app application of the micro-junction mobile phone terminal integrating bus card data, and query respective real-time operation conditions of the user and the destination by combining data platforms of different traffic modes, meanwhile, the micro-junction information center provides query results of different traffic modes based on the requirements of the user and the capacity of a specific supplier, provides transfer suggestions for the user through machine learning-deep learning model judgment results based on different core algorithms, provides transportation energy adjustment decision suggestions for the supplier, and helps the two parties to improve transfer efficiency. Meanwhile, data created by a user can enrich a historical total database and is fed back to the micro-hub information center to help optimize a corresponding algorithm, so that the service capability of the micro-hub information center is continuously improved.
FIG. 1 is a general block diagram of an intelligent micro-hub of the present invention. The method comprises the following steps that a user collects real-time taxi conditions, real-time shared bicycle conditions and real-time bus conditions through a micro-junction mobile phone terminal APP and other multimedia modes through a micro-junction information center, and the possibility of successful reservation of the shared bicycle is estimated according to a statistical algorithm, the user makes a transfer decision, the user records user data and updates a micro-junction information center database after the user leaves the station through an AFC (automatic ticketing and checking system), and if the user transfers the bus, the number of transfer persons of the buses along different time periods is estimated according to the card swiping data of the bus card and the statistical algorithm and provided for a bus platform; and if the user takes the taxi, estimating the demand distribution of the transfer taxi at different collecting and distributing ports of the micro-junction according to the data of the information center database of the micro-junction station, and providing the data to a taxi platform so that a taxi driver can decide whether to stop to receive the taxi or not.
Fig. 2 is an assistant decision process of rail-switching by bus. A user inquires public transportation real-time information through a micro-junction mobile phone terminal APP through a micro-junction information center, the user makes a decision according to the public transportation real-time information and other transfer information, the user records the user data and updates a micro-junction information center database after the user leaves a station through an AFC (automatic ticketing and ticket checking system), the user who transfers the public transportation arrives at a public transportation station, the distribution of transfer demands of the buses along the line in different time periods is estimated through a sample distribution algorithm according to the card swiping data of the buses and the user decision data of selecting taxis on the APP, and the distribution is provided for a public transportation platform data center, and the operation scheme is optimized by a public transportation platform.
Fig. 3 is an aid decision flow for a rail-swap taxi ride. A user inquires about the real-time parking condition of a taxi arriving at a station through a micro-junction mobile phone terminal APP through a micro-junction information center, the user makes a decision according to the real-time parking condition of the taxi arriving at the station and other transfer information, the user records the user data and updates a micro-junction information center database after going out of the station through an AFC (automatic ticket selling and checking system), meanwhile, the transfer requirement of the transfer taxi is estimated through a poisson algorithm according to the data of the micro-junction information center database, a taxi platform is sent out for the taxi driver to make a decision on whether to park or not, and if the taxi driver makes a parking decision, the real-time parking condition of the taxi arriving at the station is updated.
Fig. 4 is an assistant decision flow of the rail-switched shared vehicle. A user inquires real-time supply conditions of a shared bicycle and the possibility of successful reservation of the shared bicycle through a micro-hub mobile phone terminal APP through a micro-hub information center, the user makes a decision according to the information and other transfer information, if the user selects a networked reservation, the user makes the reservation of the shared bicycle and selects a specific position, if the reservation is unsuccessful, other transfer modes are selected, after the user leaves the station through an AFC (automatic ticket selling and checking system), the user data is recorded and the database of the micro-hub station information center is updated, meanwhile, the possibility of successful reservation of the shared bicycle is estimated based on a two-item distribution algorithm according to the data of the database of the micro-hub station information center, the user is provided for inquiry to the micro-hub information center, and the user can be provided to a shared bicycle platform for adjusting the delivery amount and optimizing the delivery layout of the shared bicycle platform.
In the invention, transfer suggestions are provided for users through machine learning-deep learning model judgment results based on different core algorithms.
1) Taxi
The method comprises the steps of establishing a machine learning-deep learning model to generate probability for deciding a parking passenger-waiting suggestion as a target function, wherein the tag function comprises but is not limited to sub data in a micro-hub AFC total database, derivative data of various users of the micro-hub app, decision-making data after a taxi receives the parking suggestion of the micro-hub app, taxi parking data, data whether the taxi successfully receives passengers or not, data converted from other modes into taxi facilities, layout data of the taxi facilities and the like.
2) Bus with a movable rail
Establishing probability generated by a machine learning-deep learning model for deciding different results of bus operation optimization into different objective functions, including but not limited to bus model selection, bus operation route adjustment and bus operation shift adjustment; the tag function includes, but is not limited to, various sub-data in an AFC total database, various derivative data of a micro-hub app user, various sub-data of the current operation condition of a bus line, bus proportion data selected by a micro-hub app passenger, data converted from other modes into buses by the passenger, bus platform facility layout data and the like.
3) Sharing bicycle
Establishing a target function which takes a machine learning-deep learning model generation probability to help a user to decide whether to select a sharing bicycle or not, wherein the target function mainly takes a sharing bicycle platform release decision, and the target function comprises but is not limited to the release amount of the sharing bicycle at different positions, the release amount of the sharing bicycle at different times and the decision of the user on whether to reserve the sharing bicycle or not; the tag function includes, but is not limited to, subdata in an AFC total database, derivative data of users of the micro-hub app, reserved quantity of shared vehicles, usage quantity of the shared vehicles, data of transition from passengers to shared vehicles in other modes, layout data of shared vehicle facilities, and the like.
Basis of algorithm
1) Poisson distribution
Within a certain time range, a certain thing occurs "x" times, and
(1) are independent of each other;
(2) within the same time range, the occurrence probability is the same;
(3) the above rule applies to the average number of occurrences of an event being "u".
Then
The number of people leaving the station in a specific time period on a working day is independent of each other and has approximately the same occurrence probability.
x: the number of taxi taking at the specific entrance/exit (E (x) reference) in each specific time period
u: average number of taxi-taking per specific time period (statistical data)
P (x): and judging a threshold value, and continuously improving the threshold value in the future by the initial discussion.
2) Distribution of two terms
n: number of reserved persons (platform data reference)
x: successful appointment data (platform data reference for specific use)
p: user data survey references, platform data references.
3) theoretical-F function of sample distribution
Estimating the distribution condition of the whole sample by using the using condition of the micro-junction APP as the sample
(1) Is generally normally distributed
(2) Reducing the limitations of Z and T distributions
(3) Variance consistency of app number of users in micro hub
(4) The variation has additive property
(5) And estimating the distribution condition of the whole bus taking based on the sample distribution so as to help the bus group to continuously optimize the bus operation mode.
4) Machine learning-deep learning
With the establishment of the three algorithms, the increased total database is continuously updated based on the accumulation of the time lapse, wherein the micro-pivot analysis judgment result and the actual passenger carrying result of the taxi are target functions, a data set with the judgment result consistent with the actual passenger carrying result is screened as a target function training set, and a newly added data set is used as a target function verification set. And the micro-junction analysis judgment result and the actual car booking result of the shared single car are target functions, a data set with the judgment result consistent with the actual car booking result is screened and used as a target function training set, and a newly added data set is used as a target function verification set. The bus takes the number of actual passengers (micro-hub app holders) as a target function.
The user behavior data set is used as a label function alternative set and comprises but is not limited to operation tendency judgment data in user apps, bus card AFC outbound data, user bus waiting time data, bus card swiping data of buses, bus O-D data, user average booking times data, user successful booking proportion data of single cars, existing single car putting quantity utilization efficiency data, taxi driver decision data, taxi waiting time data, taxi space vacancy data and the like, and basic data such as related time, place and the like.
On the basis, the label function weight ratio of each of the three machine learning models is randomly set, the weight is preliminarily set, the machine learning models corresponding to different transfer modes are respectively based on a preliminary loss function, the learning pace is established and adjusted, a gradient descent method is used as a core to perform data training, and the training of each group of data helps the machine learning system to judge the influence degree of all label functions on a target function so as to readjust the weight until continuous convergence is formed to further improve the model, so that the prediction for continuously improving the accuracy is provided for upcoming users and suppliers. Meanwhile, with the enrichment and accumulation of data, the newly added objective function and the tag function are grouped into training data and verification data for machine learning training, so that the limit of the traditional algorithm can be broken through, the objective function can be predicted more accurately, and the continuous improvement service of the micro-hub multi-mode overall transfer efficiency is provided.
Claims (4)
1. An intelligent micro-junction based on multi-network fusion is characterized in that rail transit, buses along the line, taxies and shared single vehicles are integrated to form an urban public transit micro-junction, information analysis processing is carried out on bus operation data, taxi operation data and shared single vehicle release data by establishing a micro-junction information center, transfer information is provided for users, user transfer information is provided for a bus platform, a taxi platform and a shared single vehicle platform, and the operation can be adjusted by an operator,
the micro-junction information center inquires real-time information of buses, taxis and shared single cars, estimates the possibility that a user successfully reserves network appointment according to the real-time information of a shared single car platform, selects different transfer modes by the user, updates a micro-junction information center database according to the outbound condition of the user, estimates the personnel distribution of each transfer mode according to a statistical algorithm by utilizing historical data in the database, and provides the estimated data for an operator to perform operation scheduling optimization;
the statistical algorithm of the rail-switched buses adopts a sample distribution algorithm as a core, selects information of personnel who transfer the buses and card swiping information of all buses along the line according to a micro-junction information center, and provides the information for a bus operator to optimize a scheme after statistics by the sample distribution algorithm; the method comprises the steps that a user inquires real-time bus information through a micro-junction information center, the user makes a decision according to the real-time bus information and other transfer information, the user records user data and updates a micro-junction information center database after the user leaves a bus through an automatic ticketing and ticket checking system, the user decision data of a taxi selected on an APP are obtained according to card swiping data of the bus when the user transfers the bus, the transfer demand distribution of each bus along the line in different time periods is estimated through a sample distribution algorithm and provided for a bus platform data center, and an operation scheme is optimized through a bus platform;
the statistical algorithm for the rail-exchanging taxi taking adopts a poisson algorithm as a core, estimates the distribution of taxi transfer personnel according to the data of a micro-hub information center database and based on poisson distribution, provides the distribution for a taxi platform for taxi drivers to make decisions, and updates the real-time parking condition of the taxi according to the decision result of the taxi drivers; a user inquires about the real-time parking condition of a taxi arriving at a station through a micro-junction information center, the user makes a decision according to the real-time parking condition of the taxi arriving at the station and other transfer information, the user records the user data and updates a micro-junction information center database after leaving the station through an automatic ticket selling and checking system, meanwhile, the transfer requirement of the transfer taxi is estimated through a poisson algorithm according to the data of the micro-junction information center database, a taxi platform is sent out for a taxi driver to make a decision whether to park, and if the taxi driver makes a parking decision, the real-time parking condition of the taxi arriving at the station is updated;
the statistical algorithm of the rail-switched shared bicycle adopts a binomial distribution algorithm as a core, provides the possibility that a user successfully reserves the network appointment according to the data of the micro-hub information center database based on binomial distribution, and provides the possibility for the user and the shared bicycle platform at the same time; the user inquires real-time supply conditions of the shared bicycle and the possibility of successful reservation of the shared bicycle through the micro-hub information center, the user makes a decision according to the information and other transfer information, if the user selects a networked reservation, the user makes the reservation of the shared bicycle and selects a specific position, if the reservation is unsuccessful, other transfer modes are selected, after the user leaves the station through the automatic ticket selling and checking system, the user data is recorded and the database of the micro-hub station information center is updated, meanwhile, the possibility of successful reservation of the shared bicycle is estimated based on a two-item distribution algorithm according to the data of the database of the micro-hub station information center, the user is provided for the micro-hub information center to inquire, and meanwhile, the user can be provided for the shared bicycle platform to adjust the input amount of the shared bicycle and optimize the layout and input.
2. The intelligent microtest of claim 1, wherein the real-time information of buses, taxis and shared bikes includes bus real-time stop times, arrival at taxi real-time stops, and shared-bikes real-time supply.
3. The intelligent micro-hub of claim 1, wherein the traffic micro-hub is classified into two categories, straight and estuary: the linear type, namely a taxi station and a bus station are arranged on a slow lane and are arranged by using the space of a non-motor lane; the harbor type is characterized in that a concave part is arranged in a sidewalk and used for setting a taxi station and a bus station, or the sidewalk is partially outwards protruded and is provided with the taxi station and the bus station along the convex part.
4. The intelligent micro-hub according to claim 1, wherein with the initial establishment and operation of the correlation algorithm, the micro-hub information system obtains the judgment result data based on different transfer power models, and various correlation data created by various behavior records of the user are continuously accumulated and enriched;
along with the continuous increase of the total amount of sample data and the creation of stability, accumulation and richness of various behavior data types by a user, a machine learning-deep learning model is established based on the large sample, a combination with consistent results is screened out by combining the judgment results of different algorithms and actual results to serve as a target function training set, the related data types are selected from user behavior record data to serve as a label function training set, the initial influence weight is determined, the learning pace is established on the basis of a primary loss function, a mathematical algorithm mainly based on a gradient descent method is used as a core for data training, and for the result of the primary loss function, a machine learning system readjusts the weight according to all the label functions until continuous convergence is formed to further perfect the model;
the newly created data are respectively distributed into a training set and a verification set in proportion according to 5, so that the machine learning model can avoid conventional errors and work efficiently, and the machine learning model with respective corresponding E-type intelligent micro-hubs as the basis for public transportation, leasing and sharing a single-vehicle transfer mode is created.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811258830.9A CN109523064B (en) | 2018-10-26 | 2018-10-26 | Intelligent micro-hub based on multi-network fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811258830.9A CN109523064B (en) | 2018-10-26 | 2018-10-26 | Intelligent micro-hub based on multi-network fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109523064A CN109523064A (en) | 2019-03-26 |
| CN109523064B true CN109523064B (en) | 2022-02-18 |
Family
ID=65773195
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811258830.9A Active CN109523064B (en) | 2018-10-26 | 2018-10-26 | Intelligent micro-hub based on multi-network fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109523064B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210049460A1 (en) * | 2019-08-15 | 2021-02-18 | Noodle Analytics, Inc. | Deep probabilistic decision machines |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112036621B (en) * | 2020-08-18 | 2023-09-22 | 北京交通大学 | Method and device for setting guiding mark in transportation junction |
| CN113129580B (en) * | 2021-03-09 | 2022-04-26 | 北京航空航天大学 | Vehicle scheduling system based on big dipper big data and face identification |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103034906A (en) * | 2012-12-26 | 2013-04-10 | 上海市城市建设设计研究总院 | Multiplex mode traffic intermodal system based on cloud computing |
| CN103500499A (en) * | 2013-09-25 | 2014-01-08 | 青岛海信网络科技股份有限公司 | Induction method and system for comprehensive passenger transport hub passenger transportation mode selection |
| CN104766474A (en) * | 2015-03-16 | 2015-07-08 | 上海市政工程设计研究总院（集团）有限公司 | Urban comprehensive transportation junction transfer passenger flow volume detecting method based on mobile phone terminals |
| CN105160429A (en) * | 2015-08-25 | 2015-12-16 | 浙江工业大学 | Multi-mode public transportation transfer method with virtual transfer micro-hubs |
| CN105719022A (en) * | 2016-01-22 | 2016-06-29 | 上海工程技术大学 | Real-time rail transit passenger flow prediction and passenger guiding system |
| CN105788260A (en) * | 2016-04-13 | 2016-07-20 | 西南交通大学 | Public transportation passenger OD calculation method based on intelligent public transportation system data |
| CN107085620A (en) * | 2017-06-05 | 2017-08-22 | 中南大学 | A kind of taxi and subway are plugged into the querying method and system of travel route |
| CN108009747A (en) * | 2017-12-21 | 2018-05-08 | 北京工业大学 | A kind of multiple dynamic decision process information acquisition method of trip mode |
-
2018
- 2018-10-26 CN CN201811258830.9A patent/CN109523064B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103034906A (en) * | 2012-12-26 | 2013-04-10 | 上海市城市建设设计研究总院 | Multiplex mode traffic intermodal system based on cloud computing |
| CN103500499A (en) * | 2013-09-25 | 2014-01-08 | 青岛海信网络科技股份有限公司 | Induction method and system for comprehensive passenger transport hub passenger transportation mode selection |
| CN104766474A (en) * | 2015-03-16 | 2015-07-08 | 上海市政工程设计研究总院（集团）有限公司 | Urban comprehensive transportation junction transfer passenger flow volume detecting method based on mobile phone terminals |
| CN105160429A (en) * | 2015-08-25 | 2015-12-16 | 浙江工业大学 | Multi-mode public transportation transfer method with virtual transfer micro-hubs |
| CN105719022A (en) * | 2016-01-22 | 2016-06-29 | 上海工程技术大学 | Real-time rail transit passenger flow prediction and passenger guiding system |
| CN105788260A (en) * | 2016-04-13 | 2016-07-20 | 西南交通大学 | Public transportation passenger OD calculation method based on intelligent public transportation system data |
| CN107085620A (en) * | 2017-06-05 | 2017-08-22 | 中南大学 | A kind of taxi and subway are plugged into the querying method and system of travel route |
| CN108009747A (en) * | 2017-12-21 | 2018-05-08 | 北京工业大学 | A kind of multiple dynamic decision process information acquisition method of trip mode |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210049460A1 (en) * | 2019-08-15 | 2021-02-18 | Noodle Analytics, Inc. | Deep probabilistic decision machines |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109523064A (en) | 2019-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Daganzo et al. | Public transportation systems: Principles of system design, operations planning and real-time control | |
| CN107036617B (en) | Travel route planning method and system combining taxi and subway | |
| Mueller et al. | Simulation‐based analysis of personal rapid transit systems: service and energy performance assessment of the Masdar City PRT case | |
| CN111815189A (en) | Modular bus dispatching system | |
| CN109523064B (en) | Intelligent micro-hub based on multi-network fusion | |
| CN107609677A (en) | A kind of customization public bus network planing method based on taxi GPS big datas | |
| Salavati et al. | Applying AHP and clustering approaches for public transportation decisionmaking: a case study of Isfahan city | |
| CN109308574A (en) | A kind of flexible bus dispatching method in internet of real-time response half | |
| CN110084397A (en) | A kind of subway straightforward line planing method | |
| Christoforou et al. | Influencing longitudinal passenger distribution on railway platforms to shorten and regularize train dwell times | |
| CN114707709A (en) | Safety early warning method and system for comprehensive passenger transport hub of railway | |
| CN115936330A (en) | Demand response type BRT vehicle scheduling algorithm based on Lagrangian principle | |
| Bruun et al. | Choosing the right public transport solution based on performance of components | |
| Wang et al. | Multi-objective optimization of customized bus routes based on full operation process | |
| CN112434969B (en) | Regional multi-type rail transit transport capacity resource allocation method | |
| Mátrai et al. | Route planning based on urban mobility management | |
| CN116562538A (en) | Microcirculation public transportation scheduling optimization method and device considering rail transit connection | |
| CN115565400A (en) | Bus operation optimization simulation method based on big data | |
| Jansson et al. | Public Transport in the era of ITS: Forms of Public Transport | |
| Kassens-Noor | Olympic transport | |
| Mobolaji et al. | Concept of advanced personal rapid transit at airports | |
| Niger | Rationalizing public transport: a Euro-Asian perspective | |
| Tang | Designing a personal rapid transit system for the University of Delaware based on student schedule data | |
| Yin et al. | Framework for deployment planning of bus rapid transit systems | |
| XIAO | MULTI-MODAL SYSTEM DESIGN FOR URBAN METRO SYSTEMS TO HANDLE EXCESSIVE DEMAND CAUSED BY MEGA-EVENTS |
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 |