CN118114954A - Intelligent bus scheduling method and system based on multi-source data analysis - Google Patents

Intelligent bus scheduling method and system based on multi-source data analysis Download PDF

Info

Publication number
CN118114954A
CN118114954A CN202410521430.1A CN202410521430A CN118114954A CN 118114954 A CN118114954 A CN 118114954A CN 202410521430 A CN202410521430 A CN 202410521430A CN 118114954 A CN118114954 A CN 118114954A
Authority
CN
China
Prior art keywords
data
congestion
bus
scheduling
area
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.)
Granted
Application number
CN202410521430.1A
Other languages
Chinese (zh)
Other versions
CN118114954B (en
Inventor
张革初
詹宸
朱永扬
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.)
Didi Advantage Technology Shenzhen Co ltd
Original Assignee
Didi Advantage Technology Shenzhen Co ltd
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 Didi Advantage Technology Shenzhen Co ltd filed Critical Didi Advantage Technology Shenzhen Co ltd
Priority to CN202410521430.1A priority Critical patent/CN118114954B/en
Publication of CN118114954A publication Critical patent/CN118114954A/en
Application granted granted Critical
Publication of CN118114954B publication Critical patent/CN118114954B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
    • 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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06315—Needs-based resource requirements planning or analysis
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/123—Traffic control systems for road vehicles indicating the position of vehicles, e.g. scheduled vehicles; Managing passenger vehicles circulating according to a fixed timetable, e.g. buses, trains, trams
    • 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

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Data Mining & Analysis (AREA)
  • Strategic Management (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • General Business, Economics & Management (AREA)
  • Mathematical Physics (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Development Economics (AREA)
  • Computational Linguistics (AREA)
  • Educational Administration (AREA)
  • Biophysics (AREA)
  • Game Theory and Decision Science (AREA)
  • Biomedical Technology (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses an intelligent bus dispatching method and system based on multi-source data analysis, which relate to the technical field of intelligent bus dispatching, wherein a trained pattern recognition model is used for judging whether a previous period is a peak period or not, and if so, a speed analysis instruction is sent out; monitoring and obtaining bus running speeds in all subareas, constructing constructed congestion coefficients in the subareas, and distinguishing whether the subareas are congested according to the congestion coefficients; taking prediction data in the subareas as input by using the area state model, scheduling buses in each subarea by using the trained vehicle scheduling model, and outputting a scheduling scheme; and (3) performing simulation test on each scheduling scheme by using a vehicle scheduling digital twin model, constructing the change of the congestion coefficient in the subarea according to the test result, constructing the improvement degree, and rapidly screening the optimal scheme with the highest feasibility from a plurality of scheduling schemes according to the improvement degree, so that the risk cost when scheduling the bus is reduced.

Description

Intelligent bus scheduling method and system based on multi-source data analysis
Technical Field
The invention relates to the technical field of intelligent bus dispatching, in particular to an intelligent bus dispatching method and system based on multi-source data analysis.
Background
The intelligent bus scheduling is an emerging traffic management concept, and is used for intelligently and finely managing and scheduling urban buses by applying a modern information technology, so that the aim of improving the bus running efficiency, reducing traffic jam, reducing energy consumption and improving urban environment is achieved. The implementation of intelligent bus dispatching is not only beneficial to improving bus operation efficiency, but also can provide more accurate bus taking service for passengers and improve passenger satisfaction.
Meanwhile, in order to increase the reliability of dispatching, the source of collected data is increased, for example, information such as the position, the speed and the passenger flow of a bus is collected in real time by means of GPS/Beidou positioning technology, communication technology, GIS geographic information system technology and the like. According to the actual demand and the characteristics of the bus route, an appropriate scheduling algorithm is adopted, such as actual scheduling, optimal scheduling, fuzzy scheduling and the like, so that optimal vehicle departure time and interval are calculated, and optimal scheduling of the bus is realized.
In the chinese patent application publication No. CN115018319 a, a smart city bus dispatching system is disclosed, which includes: the system comprises a driving difficulty acquisition module, a driving characteristic value estimation module and a matching module. The driving difficulty of each bus line is obtained through a driving difficulty obtaining module; the driving characteristic value of a driver on each line is obtained through a driving characteristic value obtaining module; constructing an adjacent probability matrix of each line through a driving characteristic value estimation module, acquiring a familiar line of a driver through the similarity between the unknown line and the adjacent probability matrix of each known line, and further estimating the driving characteristic value of the unknown line; and matching the driving characteristic value with the driving difficulty through a matching module.
Combining the contents of the above applications and prior art:
When a bus is in an operation stage, particularly in a rush hour of working or working, a large number of passengers usually select the bus as a travel mode, but if the weather state of the rush hour is poor, such as raining, road surface slippery, or a great holiday is considered in the present period, the congestion situation in the bus operation area is more serious, and a large unbalance exists in the bus quota in each area, in this case, if the bus cannot be reasonably scheduled in time, more passengers may not be sat, and then stay, in the existing bus scheduling method, a large number of passengers may be present, and the bus is scheduled, so that the current or possible congestion situation is not considered in the scheduling mode, and the due relieving effect is difficult to fully play for the congestion area.
Therefore, the invention provides an intelligent bus dispatching method and system based on multi-source data analysis.
Disclosure of Invention
(One) solving the technical problems
Aiming at the defects of the prior art, the invention provides an intelligent bus scheduling method and system based on multi-source data analysis, which are used for distinguishing whether a sub-area is congested according to congestion coefficients; taking prediction data in the subareas as input by using an area state model, scheduling buses in each subarea by using a trained vehicle scheduling model, and outputting a scheduling scheme; the vehicle dispatching digital twin model is used for carrying out simulation test on each dispatching scheme, the change of the congestion coefficient in the subarea is constructed by test data, the improvement degree is constructed, the optimal scheme with the highest feasibility is rapidly screened from a plurality of dispatching schemes according to the improvement degree, and the risk cost when dispatching the bus is reduced, so that the technical problem in the background technology is solved.
(II) technical scheme
In order to achieve the above purpose, the invention is realized by the following technical scheme:
an intelligent bus dispatching method based on multi-source data analysis comprises the following steps,
Building uniformity of bus distribution by number of buses in each areaIf uniformity/>Sending a data acquisition instruction to the outside when the uniformity threshold is lower than the uniformity threshold;
Collecting bus running and dispatching related data in the subareas, judging whether the period before training is a peak period or not according to the collected data by using a mode identification model after training, and if so, sending a speed analysis instruction;
Monitoring and obtaining the bus running speed in each subarea, and constructing the congestion coefficient in the subarea by combining the fluctuation of the running speed According to the congestion coefficient/>Distinguishing whether the sub-area is congested; wherein, the congestion factor/>The construction mode of (2) is as follows:
，
wherein, K, k is the number of subregions,/>Is a qualified standard value of a speed stable value,/>Is the stable speed value of the ith sub-area,/>Is the average value thereof; weight coefficient: /(I)，/>And (2) and；
Taking prediction data in the subareas as input by using an area state model, scheduling buses in each subarea by using a trained vehicle scheduling model, and outputting a scheduling scheme;
Using a vehicle dispatching digital twin model to carry out simulation test on each dispatching scheme, and constructing congestion coefficients in subareas by test data Variable construction improvement degree/>According to the improvement degree/>And screening out an optimal scheme.
Further, marking an operation area of the bus as a dispatching area, dividing the dispatching area into a plurality of subareas, and inquiring and acquiring the number of the buses in each subarea; wherein, build uniformityThe way of (2) is as follows: taking the ratio of the buses in the subarea to the total number of buses currently as the actual duty ratio/>And the average proportion obtained from the historical data is taken as a preset duty ratio/>According to the following mode:
，
wherein, Is weight,/>，/>And/>，/>N is the number of subregions,/>For the actual duty cycle of the ith sub-region,/>For a corresponding preset duty cycle.
Further, after receiving the data acquisition instruction, acquiring the buses and the operation data thereof in each sub-area, at least including: real-time passenger flow data, vehicle position data, road condition data, weather conditions and large-scale activity arrangement in the subareas; preprocessing the collected data respectively, and summarizing the preprocessed data to construct a bus running scheduling data set; and extracting part of data from the bus operation scheduling data set as sample data, and training a pattern recognition model by the sample data.
Further, after receiving the analysis instruction, acquiring data of each bus position and road condition, periodically acquiring the running speed of the bus, and constructing a speed stabilizing value of the busThe mode is as follows:，
wherein, ，/>For the acquisition times of the driving speed,/>For the i-th monitored travel speed,/>Is the average value thereof.
Further, the speed stabilizing value of each bus in the summarized subareaCongestion factor/>, built from several speed stable valuesIf the obtained congestion coefficient/>If the congestion threshold value is exceeded, marking the congestion area with the corresponding subarea, and taking other areas as non-congestion areas; and if the congestion area reaches half of the non-congestion area, sending a scheduling instruction to the outside.
Further, training and generating a regional state prediction model by using sample data, predicting each item of data in the subarea by using the trained regional state model, and acquiring corresponding prediction data;
constructing a vehicle scheduling model by a simulated annealing algorithm; by taking the obtained prediction data as input to reduce the congestion coefficient of the congestion area As a dispatch target, to maintain the current passenger carrying, that is, to ensure that the passengers waiting for the vehicle can have the vehicle riding as a constraint condition; and dispatching buses in each subarea by using the trained vehicle dispatching model, and obtaining and outputting a plurality of dispatching schemes.
Further, a digital twin model of vehicle dispatching is obtained after training by using sample data, and each dispatching scheme is simulated and tested by the digital twin model of vehicle dispatching, wherein: setting a test period including a plurality of sub-periods, outputting corresponding test data in each sub-period, and constructing congestion coefficients corresponding to each sub-area by the test dataMarking the sub-areas by using the congestion coefficients obtained by testing.
Further, the degree of improvementThe construction mode of (2) is as follows: sequencing all sub-periods before and after optimization according to a time axis, wherein the sub-periods correspond to each other one by one, and the obtained congestion coefficients/>, in the sub-periods before and after optimizationConstruction improvement degree/>Wherein:
，
wherein, />Congestion coefficients on the ith subcycle before and after optimization,/>, respectivelyAnd (3) withFor the corresponding mean value, i is the number of the sub-period,/>N is the number of sub-periods,/>Is the congestion factor intermediate value in the ith subcycle,/>Is the corresponding mean value.
Further, if the degree of improvement isIf the number exceeds the expected number, the improved area is improved, and the scheduling scheme with the most improved area is output as an optimal scheme; and constructing a dispatching instruction by the optimal scheme, and issuing the dispatching instruction to each bus.
An intelligent bus scheduling system based on multi-source data analysis, comprising:
vehicle distribution analysis unit for constructing uniformity of bus distribution according to number of buses in each area If uniformity/>Sending a data acquisition instruction to the outside when the uniformity threshold is lower than the uniformity threshold;
The mode identification unit is used for collecting related data of bus operation and scheduling in the subareas, judging whether the period before training is a peak period or not according to the collected data by the mode identification model after training, and if so, sending out a speed analysis instruction;
The congestion analysis unit monitors and acquires the running speed of buses in each subarea, and constructs congestion coefficients in the subareas by combining the fluctuation of the running speed According to the congestion coefficient/>Distinguishing whether the sub-area is congested;
The vehicle dispatching unit takes the prediction data in the subareas as input by using the area state model, dispatches buses in each subarea by using the trained vehicle dispatching model, and outputs a dispatching scheme;
The scheme test unit is used for carrying out simulation test on each scheduling scheme by using a vehicle scheduling digital twin model, and constructing congestion coefficients in the subareas by test data Variable construction improvement degree/>According to the improvement degree/>And screening out an optimal scheme.
(III) beneficial effects
The invention provides an intelligent bus scheduling method and system based on multi-source data analysis, which have the following beneficial effects:
1. According to uniformity Judging whether the current bus distribution is reasonable, if so, not entering a dispatching process currently, and if not, reallocating the buses for each subarea to ensure that passengers of a traffic route can be all transported away when the traffic condition is poor.
2. According to the collected data as input, whether the vehicle running in the dispatching area is in a peak period or a non-peak period is further judged, whether the bus dispatching is carried out or not is further determined according to whether the traffic period is the peak period or not through judging the traffic period, and the travel pressure is reduced.
3. According to the congestion factorJudging whether the current congestion exists in each subarea or not, and further screening out a congestion area and a non-congestion area respectively; at this time, according to the constructed congestion coefficient/>The method can realize rapid screening of the congestion area, can clearly schedule the target when the buses are required to be scheduled, improves the scheduling efficiency, and can relieve the current passing pressure and slow down the subsequent passenger retention pressure by dispatching the vehicles when the current peak time period is determined.
4. The buses in the congestion areas and the non-congestion areas are scheduled by using the trained vehicle scheduling model, the speed of the buses is subjectively increased by increasing the shifts of the buses, the stop points are reduced, and the like, the passing speed of the buses is increased, the current congestion areas are relieved, and after the congestion areas are determined, notification can be sent to the outside, so that further congestion is avoided.
5. According to the degree of improvementThe method can judge and evaluate the effectiveness of the current dispatching scheme, thereby guaranteeing the effect and the corresponding feasibility of the dispatching scheme, rapidly screening the dispatching scheme with the highest feasibility, reducing the risk cost when dispatching the bus and guaranteeing the dispatching success rate.
Drawings
FIG. 1 is a flow chart of an intelligent bus dispatching method of the invention;
FIG. 2 is a schematic diagram of the intelligent bus dispatching system.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the invention provides an intelligent bus scheduling method based on multi-source data analysis, which comprises the following steps:
step one, constructing uniformity of bus distribution according to the number of buses in each area If uniformity/>Sending a data acquisition instruction to the outside when the uniformity threshold is lower than the uniformity threshold;
the first step comprises the following steps:
Step 101, before a bus is required to be scheduled, marking an operation area of the bus as a scheduling area, constructing an electronic map covering the scheduling area, dividing the scheduling area into a plurality of sub-areas on the electronic map, numbering and marking each sub-area respectively, and inquiring and obtaining the number of the buses in each sub-area;
102, after the current number of buses in each subarea is obtained, constructing uniformity according to the distribution state of buses The mode is as follows: taking the ratio of the buses in the subarea to the total number of buses currently as the actual duty ratio/>And the average proportion obtained from the historical data is taken as a preset duty ratio/>According to the following mode:
，
wherein, Is weight,/>，/>And/>，/>N is the number of subregions,/>For the actual duty cycle of the ith sub-region,/>A corresponding preset duty cycle;
presetting a uniformity threshold according to bus management expectation and historical data; if the uniformity is obtained If the uniformity threshold is lower than the uniformity threshold, the current buses are unevenly distributed in each subarea, the passenger carrying requirements in the subareas may not be completely corresponding, data acquisition is performed if the passenger carrying requirements in the subareas are needed, the buses in each subarea are scheduled on the basis of the data acquisition, and at the moment, a data acquisition instruction is sent to the outside;
In use, the contents of steps 101 and 102 are combined:
When considering whether to schedule the buses in the operation area, firstly acquiring the number of buses in each subarea, and calculating the uniformity of the acquired distribution state according to the difference between the current bus distribution and the expected bus distribution Thereby according to uniformity/>Judging whether the current bus distribution is reasonable, if so, not entering a dispatching process currently, and if not, reallocating the buses for each subarea to ensure that passengers of a traffic route can be all transported away when the traffic condition is poor.
Combining the contents of the above applications and prior art:
When a bus is in an operation stage, particularly in a rush hour of working or working, a large number of passengers usually select the bus as a travel mode, but if the weather state of the rush hour is poor, such as raining, road surface slippery, or a great holiday is considered in the present period, the congestion situation in the bus operation area is more serious, and a large unbalance exists in the bus quota in each area, in this case, if the bus cannot be reasonably scheduled in time, more passengers may not be sat, and then stay, in the existing bus scheduling method, a large number of passengers may be present, and the bus is scheduled, so that the current or possible congestion situation is not considered in the scheduling mode, and the due relieving effect is difficult to fully play for the congestion area.
Collecting bus running and dispatching related data in the subareas, judging whether the period before training is a peak period or not according to the collected data by using the mode identification model after training, and if so, sending out a speed analysis instruction;
the second step comprises the following steps:
Step 201, after receiving a data acquisition instruction, acquiring buses and operation data thereof in each sub-area, at least including: real-time passenger flow data, vehicle position data, vehicle speed, road condition data, weather conditions, large-scale activity arrangements and the like in the subareas; preprocessing the collected data respectively, and summarizing the preprocessed data to construct a bus running scheduling data set;
Step 202, extracting part of data from a bus operation scheduling data set as sample data, constructing an initial model by a convolutional neural network, training a pattern recognition model by the sample data, recognizing data acquired in real time by the trained pattern recognition model, such as real-time passenger flow data and the like, recognizing whether the current time period is a peak time period or not, and if so, sending a speed analysis instruction;
in use, the contents of steps 201 and 202 are combined:
when buses among all subareas are considered to be scheduled, data related to bus driving in the scheduling area are collected first, after a mode identification model is trained, the collected data is used as input, whether the bus driving in the scheduling area is in a peak period or a non-peak period is judged, whether the bus is scheduled or not is further determined according to whether the traffic period is the peak period through judging the traffic period, and travel pressure is reduced.
Step three, monitoring and obtaining the running speed of buses in each subarea, constructing congestion coefficients in the subareas by combining the fluctuation of the running speed, and distinguishing whether the subareas are congested according to the congestion coefficients;
The third step comprises the following steps:
Step 301, after receiving an analysis instruction, acquiring data of each bus position and road condition, periodically acquiring the running speed of the bus, acquiring the running speed of the bus in each monitoring period, and constructing a speed stabilizing value of the bus according to the fluctuation of the speed of the bus The mode is as follows: /(I)，
Wherein,，/>For the acquisition times of the driving speed,/>For the i-th monitored travel speed,/>Is the average value thereof;
Step 302, summarizing the speed stability values of each bus in the subarea Congestion factor/>, built from several speed stable valuesThe mode is as follows:
，
wherein, K, k is the number of subregions,/>Is a qualified standard value of a speed stable value,/>Is the stable speed value of the ith sub-area,/>Is the average value thereof; weight coefficient: /(I)，/>And (2) and; The weight coefficient is determined by reference analytic hierarchy process or is consistent with the previous value;
Presetting a congestion threshold according to historical data and management expectations of bus running speeds; if the obtained congestion coefficient If the congestion threshold value is exceeded, marking the congestion area with the corresponding subarea, and taking other areas as non-congestion areas; if the congestion area reaches half of the non-congestion area, a scheduling instruction is sent to the outside;
in use, the contents of steps 301 and 302 are combined:
After the peak period of traffic is determined, the running speed and the state of each bus are monitored, and a speed stabilizing value is built according to the fluctuation of the running speed Thereby constructing the congestion coefficient/>According to the congestion coefficient/>Judging whether the current congestion exists in each subarea or not, and further screening out a congestion area and a non-congestion area respectively; at this time, according to the constructed congestion coefficient/>The method can realize rapid screening of the congestion area, can clearly schedule the target when the buses are required to be scheduled, improves the scheduling efficiency, and can relieve the current passing pressure and slow down the subsequent passenger retention pressure by dispatching the vehicles when the current peak time period is determined.
Step four, taking prediction data in the subareas as input by using the area state model, scheduling buses in each subarea by using the trained vehicle scheduling model, and outputting a scheduling scheme;
The fourth step comprises the following steps:
Step 401, extracting part of data from a bus operation scheduling data set as sample data, constructing an initial model by a convolutional neural network, training the initial model by using the sample data to generate a regional state prediction model, predicting various data in a subarea by using the trained regional state model, such as vehicle speed, vehicle data, weather conditions, passenger number and the like, and acquiring corresponding prediction data;
Step 402, constructing an initial model by a simulated annealing algorithm, and constructing a vehicle scheduling model after training the initial model by using the number of samples; by taking the obtained prediction data as input to reduce the congestion coefficient of the congestion area As a dispatch target, to maintain the current passenger carrying, that is, to ensure that the passengers waiting for the vehicle can have the vehicle riding as a constraint condition;
Scheduling buses in each subarea by using a trained vehicle scheduling model, acquiring a plurality of scheduling schemes and outputting the scheduling schemes; the dispatching scheme comprises the steps of adjusting bus departure frequency, vehicle speed, route optimization and the like;
In use, the contents of steps 401 and 402 are combined:
After the congestion area and the non-congestion area are screened out, each item of data in each sub-area is predicted through a trained area state model, buses in the congestion area and the non-congestion area are scheduled through a trained vehicle scheduling model on the basis of the obtained predicted data, the speed of the buses is subjectively improved by increasing the shifts of the buses, stop stations are reduced, the passing speed of the buses is increased, the current congestion area is relieved, and meanwhile, after the congestion area is determined, notification can be sent to the outside to avoid further congestion.
Step five, using a vehicle dispatching digital twin model to carry out simulation test on each dispatching scheme, and constructing congestion coefficients in the subareas according to test dataVariable construction improvement degree/>According to the improvement degree/>Screening out an optimal scheme;
The fifth step comprises the following steps:
Step 501, constructing an initial model by a machine learning algorithm, training by using sample data, obtaining a vehicle dispatching digital twin model, and performing simulation test on each dispatching scheme by the vehicle dispatching digital twin model, wherein: setting a test period including a plurality of sub-periods, outputting corresponding test data in each sub-period, and constructing congestion coefficients corresponding to each sub-area by the test data Marking the sub-areas by using the congestion coefficients obtained by testing;
step 502, according to the congestion coefficients in each sub-area Variable construction improvement degree/>The mode is as follows: sequencing all sub-periods before and after optimization according to a time axis, wherein the sub-periods correspond to each other one by one, and the obtained congestion coefficients/>, in the sub-periods before and after optimizationConstruction improvement degree/>The mode is as follows:
，
wherein, />Congestion coefficients on the ith subcycle before and after optimization,/>, respectivelyAnd (3) withFor the corresponding mean value, i is the number of the sub-period,/>N is the number of sub-periods,/>Is the congestion factor intermediate value in the ith subcycle,/>Is the corresponding average value;
Presetting an improvement degree threshold according to historical data and effect expectation of a scheduling scheme, and if the obtained improvement degree is If the number exceeds the expected number, the improved area is improved, and the scheduling scheme with the most improved area is output as an optimal scheme; constructing a dispatching instruction by the optimal scheme, and issuing the dispatching instruction to each bus;
In use, the contents of steps 501 and 502 are combined;
On the basis of obtaining a plurality of scheduling schemes, a vehicle scheduling digital twin model is built, the scheduling schemes are tested by using the vehicle scheduling digital twin model, and the improvement degree is built by test data According to the improvement degree/>The method can judge and evaluate the effectiveness of the current dispatching scheme, thereby guaranteeing the effect and the corresponding feasibility of the dispatching scheme, rapidly screening the dispatching scheme with the highest feasibility, reducing the risk cost when dispatching the bus and guaranteeing the dispatching success rate.
Referring to fig. 2, the present invention provides an intelligent bus dispatching system based on multi-source data analysis, comprising,
Vehicle distribution analysis unit for constructing uniformity of bus distribution according to number of buses in each areaIf uniformity/>Sending a data acquisition instruction to the outside when the uniformity threshold is lower than the uniformity threshold;
The mode identification unit is used for collecting related data of bus operation and scheduling in the subareas, judging whether the period before training is a peak period or not according to the collected data by the mode identification model after training, and if so, sending out a speed analysis instruction;
The congestion analysis unit monitors and acquires the running speed of buses in each subarea, and constructs congestion coefficients in the subareas by combining the fluctuation of the running speed According to the congestion coefficient/>Distinguishing whether the sub-area is congested;
the vehicle dispatching unit takes the prediction data in the subareas as input by using the area state model, dispatches buses in each subarea by using the trained vehicle dispatching model, and outputs the obtained dispatching scheme;
The scheme test unit is used for carrying out simulation test on each scheduling scheme by using a vehicle scheduling digital twin model, and constructing congestion coefficients in the subareas by test data Variable construction improvement degree/>According to the improvement degree/>And screening out an optimal scheme.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
The analytic hierarchy Process (AHP for short) is a decision analysis method, which decomposes elements related to decision into levels of targets, criteria, schemes, etc., and then performs qualitative and quantitative analysis on the basis of the decomposition. The analytic hierarchy process decomposes a complex multi-objective decision problem into multiple objectives or criteria by treating it as a system and decomposing the objectives into several levels of multi-objectives (or criteria, constraints). The method calculates the hierarchical single ordering (weights) and the total ordering by a qualitative index fuzzy quantization method, thereby being used as a system method for optimizing decision of a target (multiple indexes) and multiple schemes.
The basic principle of the analytic hierarchy process is: decomposing the decision problem into different hierarchical structures according to the sequence of the total target, each layer of sub-target and the evaluation criterion until a specific spare power switching scheme, then solving the priority weight of each element of each layer to a certain element of the previous layer by using a method for solving and judging matrix characteristic vectors, and finally merging the final weights of each alternative scheme to the total target in a hierarchical mode by using a weighted sum method, wherein the final weight with the maximum value is the optimal scheme.
In the several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the elements is merely a division of some logic functions, and there may be additional divisions in actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (10)

1. An intelligent bus scheduling method based on multi-source data analysis is characterized in that: comprising the steps of (a) a step of,
Building uniformity of bus distribution by number of buses in each areaIf uniformity/>Sending a data acquisition instruction to the outside when the uniformity threshold is lower than the uniformity threshold;
Collecting bus running and dispatching related data in the subareas, judging whether the period before training is a peak period or not according to the collected data by using a mode identification model after training, and if so, sending a speed analysis instruction;
Monitoring and obtaining the bus running speed in each subarea, and constructing the congestion coefficient in the subarea by combining the fluctuation of the running speed According to the congestion coefficient/>Distinguishing whether the sub-area is congested; wherein, the congestion factor/>The construction mode of (2) is as follows:
，
wherein, K, k is the number of subregions,/>Is a qualified standard value of a speed stable value,/>Is the stable speed value of the ith sub-area,/>Is the average value thereof; weight coefficient: /(I)，/>And (2) and；
Taking prediction data in the subareas as input by using an area state model, scheduling buses in each subarea by using a trained vehicle scheduling model, and outputting a scheduling scheme;
Using a vehicle dispatching digital twin model to carry out simulation test on each dispatching scheme, and constructing congestion coefficients in subareas by test data Variable construction improvement degree/>According to the improvement degree/>And screening out an optimal scheme.
2. The intelligent bus scheduling method based on multi-source data analysis according to claim 1, wherein the intelligent bus scheduling method is characterized in that:
Marking an operation area of a bus as a dispatching area, dividing the dispatching area into a plurality of sub-areas, and inquiring and acquiring the number of buses in each sub-area; wherein, build uniformity The way of (2) is as follows: taking the ratio of the buses in the subarea to the total number of buses currently as the actual duty ratio/>And the average proportion obtained from the historical data is taken as a preset duty ratio/>According to the following mode:
，
wherein, Is weight,/>，/>And/>，/>N is the number of subregions,/>For the actual duty cycle of the ith sub-region,/>For a corresponding preset duty cycle.
3. The intelligent bus scheduling method based on multi-source data analysis according to claim 1, wherein the intelligent bus scheduling method is characterized in that:
After receiving the data acquisition instruction, acquiring buses and operation data thereof in each subarea, wherein the data acquisition instruction at least comprises the following steps: real-time passenger flow data, vehicle position data, vehicle speed, road condition data, weather conditions, and large-scale activity arrangements within the sub-area; preprocessing the collected data respectively, and summarizing the preprocessed data to construct a bus running scheduling data set; and extracting part of data from the bus operation scheduling data set as sample data, and training a pattern recognition model by the sample data.
4. A multi-source data analysis-based intelligent bus scheduling method as set forth in claim 3, wherein:
After receiving the analysis instruction, acquiring data of each bus position and road condition, periodically acquiring the running speed of the bus, and constructing a speed stabilizing value of the bus The mode is as follows: /(I)，
Wherein,，/>For the acquisition times of the driving speed,/>For the travel speed at the i-th monitoring,Is the average value thereof.
5. The intelligent bus scheduling method based on multi-source data analysis according to claim 4, wherein the intelligent bus scheduling method is characterized in that:
summarizing speed stabilizing values of buses in subareas Congestion factor/>, built from several speed stable valuesIf the obtained congestion coefficient/>If the congestion threshold value is exceeded, marking the congestion area with the corresponding subarea, and taking other areas as non-congestion areas; and if the congestion area reaches half of the non-congestion area, sending a scheduling instruction to the outside.
6. The intelligent bus scheduling method based on multi-source data analysis according to claim 5, wherein the intelligent bus scheduling method is characterized in that:
training to generate a regional state prediction model by using sample data, predicting each item of data in the subareas by using the trained regional state model, and acquiring corresponding prediction data;
constructing a vehicle scheduling model by a simulated annealing algorithm; by taking the obtained prediction data as input to reduce the congestion coefficient of the congestion area As a dispatch target to maintain current passenger delivery as a constraint; and dispatching buses in each subarea by using the trained vehicle dispatching model, and obtaining and outputting a plurality of dispatching schemes.
7. The intelligent bus scheduling method based on multi-source data analysis according to claim 1, wherein the intelligent bus scheduling method is characterized in that:
Obtaining a vehicle dispatching digital twin model after training by using sample data, and performing simulation test on each dispatching scheme by the vehicle dispatching digital twin model, wherein: setting a test period including a plurality of sub-periods, outputting corresponding test data in each sub-period, and constructing congestion coefficients corresponding to each sub-area by the test data To test the obtained congestion coefficient/>The sub-regions are marked.
8. The intelligent bus scheduling method based on multi-source data analysis according to claim 7, wherein the intelligent bus scheduling method is characterized in that:
Degree of improvement The construction mode of (2) is as follows: sequencing all sub-periods before and after optimization according to a time axis, wherein the sub-periods correspond to each other one by one, and the obtained congestion coefficients/>, in the sub-periods before and after optimizationConstruction improvement degree/>Wherein:
，
wherein, />Congestion coefficients on the ith subcycle before and after optimization,/>, respectivelyAnd/>For the corresponding mean value, i is the number of the sub-period,/>N is the number of sub-periods,/>Is the congestion factor intermediate value in the ith subcycle,/>Is the corresponding mean value.
9. The intelligent bus scheduling method based on multi-source data analysis according to claim 8, wherein the intelligent bus scheduling method is characterized in that:
if the degree of improvement is improved If the number exceeds the expected number, the improved area is improved, and the scheduling scheme with the most improved area is output as an optimal scheme; and constructing a dispatching instruction by the optimal scheme, and issuing the dispatching instruction to each bus.
10. An intelligent bus dispatching system based on multi-source data analysis is characterized in that: comprising the steps of (a) a step of,
Vehicle distribution analysis unit for constructing uniformity of bus distribution according to number of buses in each areaIf uniformity/>Sending a data acquisition instruction to the outside when the uniformity threshold is lower than the uniformity threshold;
The mode identification unit is used for collecting related data of bus operation and scheduling in the subareas, judging whether the period before training is a peak period or not according to the collected data by the mode identification model after training, and if so, sending out a speed analysis instruction;
The congestion analysis unit monitors and acquires the running speed of buses in each subarea, and constructs congestion coefficients in the subareas by combining the fluctuation of the running speed According to the congestion coefficient/>Distinguishing whether the sub-area is congested;
The vehicle dispatching unit takes the prediction data in the subareas as input by using the area state model, dispatches buses in each subarea by using the trained vehicle dispatching model, and outputs a dispatching scheme;
The scheme test unit is used for carrying out simulation test on each scheduling scheme by using a vehicle scheduling digital twin model, and constructing congestion coefficients in the subareas by test data Variable construction improvement degree/>According to the improvement degree/>And screening out an optimal scheme.
CN202410521430.1A 2024-04-28 2024-04-28 Intelligent bus scheduling method and system based on multi-source data analysis Active CN118114954B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410521430.1A CN118114954B (en) 2024-04-28 2024-04-28 Intelligent bus scheduling method and system based on multi-source data analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410521430.1A CN118114954B (en) 2024-04-28 2024-04-28 Intelligent bus scheduling method and system based on multi-source data analysis

Publications (2)

Publication Number Publication Date
CN118114954A true CN118114954A (en) 2024-05-31
CN118114954B CN118114954B (en) 2024-07-12

Family

ID=91212331

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410521430.1A Active CN118114954B (en) 2024-04-28 2024-04-28 Intelligent bus scheduling method and system based on multi-source data analysis

Country Status (1)

Country Link
CN (1) CN118114954B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116704771A (en) * 2023-06-21 2023-09-05 中咨数据有限公司 Real-time positioning command processing system based on traffic information of congested road section
CN117634792A (en) * 2023-11-24 2024-03-01 招商局金融科技有限公司 Port congestion scheduling method, device, equipment and medium
CN117671992A (en) * 2024-01-30 2024-03-08 山东恒宇电子有限公司 Intelligent bus dispatching method and system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116704771A (en) * 2023-06-21 2023-09-05 中咨数据有限公司 Real-time positioning command processing system based on traffic information of congested road section
CN117634792A (en) * 2023-11-24 2024-03-01 招商局金融科技有限公司 Port congestion scheduling method, device, equipment and medium
CN117671992A (en) * 2024-01-30 2024-03-08 山东恒宇电子有限公司 Intelligent bus dispatching method and system

Also Published As

Publication number Publication date
CN118114954B (en) 2024-07-12

Similar Documents

Publication Publication Date Title
CN111951549B (en) Self-adaptive traffic signal lamp control method and system in networked vehicle environment
CN105894847B (en) The real-time bus dynamic dispatching system and method for unsupervised learning under a kind of cloud platform environment
CN111564053B (en) Vehicle scheduling method and device, vehicle scheduling equipment and storage medium
CN110276474A (en) A kind of track traffic station passenger flow forecasting in short-term
CN107194491A (en) A kind of dynamic dispatching method based on Forecasting of Travel Time between bus passenger flow and station
CN114023073B (en) Expressway congestion prediction method based on vehicle behavior analysis
CN109376906B (en) Travel time prediction method and system based on multi-dimensional trajectory and electronic equipment
CN113538067B (en) Inter-city network vehicle-closing demand prediction method and system based on machine learning
CN112950128A (en) Intelligent logistics transportation system and method based on big data fusion
CN114418606B (en) Network vehicle order demand prediction method based on space-time convolution network
CN112949987B (en) Taxi scheduling and matching method, system, equipment and medium based on prediction
CN115860594A (en) Simulation system and method applied to intelligent bus scheduling
CN117371596A (en) Public transport comprehensive regulation and control system for smart city based on multi-source data
CN115953186A (en) Network appointment demand pattern recognition and short-time demand prediction method
CN112669595B (en) Network taxi booking flow prediction method based on deep learning
CN118114954B (en) Intelligent bus scheduling method and system based on multi-source data analysis
CN114463978A (en) Data monitoring method based on rail transit information processing terminal
CN113592246B (en) PCA-TOPSIS model-based road construction period bus network evaluation method
Deng et al. Research on bus passenger traffic forecasting model based on gps and ic card data
CZ20011464A3 (en) Method for determining driving times and driving time prognoses in a transport system by making use of cellular telephone positioning and a system for making the same
CN117196266B (en) Unmanned shared automobile area scheduling method and device based on neural network
Bao et al. How do metro station crowd flows influence the taxi demand based on deep spatial-temporal network?
CN118396196B (en) Intelligent city planning management method based on big data
CN118114995B (en) Full-link monitoring system suitable for high-frequency acquisition of energy data
Niu Application of Genetic Algorithm to Optimize Transit Schedule under Time-Dependent Demand

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