CN113283714A - Traffic jam suppression method based on group decision - Google Patents

Traffic jam suppression method based on group decision Download PDF

Info

Publication number
CN113283714A
CN113283714A CN202110502780.XA CN202110502780A CN113283714A CN 113283714 A CN113283714 A CN 113283714A CN 202110502780 A CN202110502780 A CN 202110502780A CN 113283714 A CN113283714 A CN 113283714A
Authority
CN
China
Prior art keywords
route
data
traffic
attribute
evaluation
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
CN202110502780.XA
Other languages
Chinese (zh)
Other versions
CN113283714B (en
Inventor
安吉尧
付志强
刘韦
郭亮
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.)
Hunan University
Original Assignee
Hunan University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hunan University filed Critical Hunan University
Priority to CN202110502780.XA priority Critical patent/CN113283714B/en
Publication of CN113283714A publication Critical patent/CN113283714A/en
Application granted granted Critical
Publication of CN113283714B publication Critical patent/CN113283714B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/7867—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • 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
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • 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)
  • Strategic Management (AREA)
  • Educational Administration (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Tourism & Hospitality (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • General Business, Economics & Management (AREA)
  • Computational Mathematics (AREA)
  • Operations Research (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Game Theory and Decision Science (AREA)
  • Quality & Reliability (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Computing Systems (AREA)
  • Library & Information Science (AREA)
  • Primary Health Care (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Algebra (AREA)
  • Software Systems (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention relates to a traffic jam suppression method based on group decision, and belongs to the technical field of traffic jam suppression. And collecting and uploading the data to a cloud platform data center by using a terminal sensor system in a network connection environment, and collecting intersection traffic data by using traffic monitoring systems at all intersections in a city to form an evaluation matrix. And re-assigning and normalizing the attribute weight of each trunk to obtain a comprehensive evaluation matrix. And calculating the comprehensive ranking value of each trunk according to the user preference and the comprehensive evaluation matrix data, and continuously uploading the latest data to the cloud platform data center. The method has an obvious effect on the aspect of road congestion inhibition in an information physical system, effectively reduces the road congestion and blockage, and further reduces the accident rate of driving; the parking rate and the waiting time in traffic driving are effectively reduced, and the travel cost is further reduced. The congestion of a traffic network is inhibited, and the traffic condition of urban main roads is improved.

Description

Traffic jam suppression method based on group decision
Technical Field
The invention relates to a traffic jam suppression method based on group decision, and belongs to the technical field of traffic jam suppression.
Background
At present, urbanization in China is accelerated, and economy is also developed at a high speed, so that more and more motor vehicles are used. The urban roads are more and more complex, the traffic jam phenomenon is more and more serious, and the living standard of citizens is influenced. Traffic congestion gradually becomes a prominent problem in the traffic field. Meanwhile, along with the continuous development of information technology and intelligent manufacturing, the city is endowed with higher requirements, and new development opportunities are brought to the construction of smart cities. Urban traffic systems are an important component enabling human activities. At present, two thirds of cities in China have traffic jam, and the phenomenon is particularly prominent in the first-line city. However, when a large city is congested, once the traffic jam is not solved in time, more and more congested areas can be caused, even traffic paralysis can be caused, and the traffic jam can cause air pollution, travel time, fuel cost, harm to life health of people and other problems. On the one hand, traffic congestion increases the emission of carbon dioxide and several other pollutants in the air. Moreover, the fuel consumption is greatly increased, and the travel cost of travelers is increased. On the other hand, traffic accidents are easily caused by traffic jam, the life health of travelers is harmed, and the rescue time of patients is endangered due to the fact that the ambulance runs and is blocked. Although the intelligent transportation system becomes a new field in the transportation research, with the aid of ITS (intelligent transportation system), the current transportation facilities can be utilized, reasonable traffic control strategies can be formulated, the traffic jam is relieved, and the environmental pollution is reduced, so that the intelligent transportation system is widely concerned by various scholars, and is one of the ways for solving the problems in the transportation field at present. However, the existing research is a single decision-making individual when making decisions. The single individual inevitably has knowledge blind areas or professional limitations, the intelligence and experience of multiple people cannot be fully considered, the opinion or experience of the multiple people can be fully considered by the group decision theory, and the decision risk can be greatly reduced in the decision process. Few researchers consider how to suppress traffic congestion from the group decision level. The decision making of the intelligent traffic system rarely uses a group decision making idea, so that unreasonable and unfair decision making results possibly exist, the user can possibly keep the contrary psychology on the suggestions given by the traffic intelligent system, the user excessively refuses the decision-making suggestions given by the traffic system, the control strategy of the traffic system is invalid, and the traffic jam is not relieved. At present, most managers give road strength inducing schemes by analyzing the influence and preference degree of various factors on the driver path selection from the whole. There is an inevitable conflict between the manager and the traveler. Because modern human beings relate to a wide range of information and a plurality of influencing factors, the scientization of decision is not possible to be well completed only by the ability of one person, and the modern human beings need to concentrate the advantages of the groups and the wisdom of the people to make the best decision. The group decision idea is proposed to make an assistant decision for the traveler, and make the decision-making persons agree during the decision-making process, so that the result is more convincing.
Disclosure of Invention
The invention aims to provide a traffic jam suppression method based on group decision in an intelligent networking environment, relates to multiple technologies of multi-data acquisition (sensor, GPS, radar), data processing and group decision, is suitable for the field of traffic jam suppression in the field of automobile information physical systems (CPS), and particularly is suitable for a multi-attribute group decision scene, so that the defects in the prior art are overcome.
The invention is realized by the following technical scheme, which comprises the following steps:
step 1: and analyzing the decision problem, and selecting an alternative route set S, wherein S is 1,2,3, … and S. Collecting road network traffic data by using a terminal sensor system in a network connection environment and uploading the road network traffic data to a cloud platform data center, and simultaneously collecting intersection traffic data by using traffic monitoring systems of all intersections in a city, wherein the intersection traffic data comprises vehicle-mounted sensors, traffic system data and GPS (global positioning system) position information;
step 2: the cloud platform data center carries out data preprocessing on received data, wherein the data preprocessing comprises data dimension reduction and data redundancy removal to obtain standardized original data, and a scheme evaluation attribute set P is determined as { P [1], P [2], … and P [ d ] };
and step 3: the cloud platform scores standardized original data by means of a plurality of road traffic auxiliary systems to form an evaluation matrix, d attributes of standardized S routes are evaluated and scored through M intelligent traffic systems, and an attribute set p of the routes is { p [1], p [2], …, p [ d ] }, and p [ i ] represents the ith attribute of the route; respectively storing the scoring results in a dXM matrix to finally obtain S evaluation matrices, and carrying out forward and dimension elimination processing on the attribute evaluation indexes;
and 4, step 4: the cloud platform data center performs reassignment and normalization processing on the attribute weight of each trunk by using the proposed iterative Reverse Top-k algorithm based on the evaluation matrix of each route to obtain a comprehensive decision matrix;
step 4.1: for evaluation matrix AsS1, 2,3, …, S, by Top-k, usually k 2 to d-1, and the result is reported as Top (S) -k(m)Gathering Top-k results under the mth expert evaluation system corresponding to the mth route;
step 4.2: for all of the routes Top(s) -k(m)The set forms a Top(s) -k set table, the Top(s) -k set table is subjected to Reverse Top-k operation, and the obtained result is recorded as RTop(s) -kp[i]Reverse Top-k result corresponding to i-th attribute of s-th route, RTop(s) -k for all routesp[i]An RTop(s) -k set table is formed by the sets, and the evaluation similarity among the d attributes of the route s can be calculated according to the RTop(s) -k set table obtained by Reverse Top-k operation, so that the dominance of each attribute is obtained;
step 4.3: d-2 times of circulation are carried out in sequence (k is 2-d-1), and the attribute dominance degree obtained by each circulation is normalized to obtain an attribute weight value;
step 4.4: the cloud platform data center obtains the evaluation row vector A of the s-th route again according to the attribute weight values', S is 1,2,3, …, S, obtaining the evaluation matrix of S routes in turn and merging the evaluation matrix into an S x d comprehensive evaluation matrix A, and returning the comprehensive evaluation matrix to the terminal side;
and 5: the terminal side equipment receives data returned by the cloud platform data center, calculates comprehensive ranking values of all routes according to user preferences and comprehensive evaluation matrix data to obtain an optimal decision, and continuously uploads latest data to the cloud platform data center; step 6: and the cloud platform center acquires new traffic data acquired by the terminal side sensor system and traffic data in the intersection monitoring system in real time, and performs data fusion processing with historical data.
In step2, the route average speed
Figure BDA0003057081320000031
n is the total number of vehicles traveling on the route, t is the number of road segments in the route; total length of travel of route
Figure BDA0003057081320000032
Average waiting total time of route traffic lights
Figure BDA0003057081320000033
The road condition evaluation As of the road is based on the video images collected by the vehicle-mounted recorder and the traffic system, and the million-vehicle-kilometer accident rate of the road
Figure BDA0003057081320000034
The step3 comprises the following scoring rules:
the congestion level of the route is as follows: when in use
Figure BDA0003057081320000035
When the running state is smooth;
Figure BDA0003057081320000036
when the running state is basically smooth;
Figure BDA0003057081320000037
when the vehicle is in a light congestion state, the running state is light congestion;
Figure BDA0003057081320000038
when the running state is moderate congestion;
Figure BDA0003057081320000039
when the running state is severe congestion;
total length of route grade: the total length of the route L and the reverse data index need to be normalized into a forward evaluation index, and two methods are generally adopted:
mode 1, let U be (total length L × number S of routes)/sum of all route lengths
Figure BDA00030570813200000310
Then
Figure BDA00030570813200000311
Wherein S is the number of route solutions;
mode 2, million-vehicle-kilometer accident rate of route road
Figure BDA00030570813200000312
Wherein N isacIs the number of accidents occurring on the route; when P is larger than 3, the per million vehicle kilometer accident rate grade of the route road is high; when 1 < TwWhen the accident rate is less than or equal to 3, the per million vehicle kilometer accident rate grade of the route road is medium; when T iswAnd when the accident rate is less than or equal to 1, the per million vehicle kilometer accident rate grade of the route road is low.
The detailed calculation steps of the step4 are as follows:
step 4.1: for evaluation matrix AsS1, 2,3, …, s, performs a Top-k operation involving the number of attributes in the evaluation matrix, the number of evaluation expert scoring systems, and for a given set of attributes, p, d, { p [1]，p[2]，…，p[d]}，p[i]Representing the ith attribute of the attribute p, wherein the value range of k is 2-d-1;
evaluation matrix A1：
A1＝[a11 a12 a13 … a1(M-1) a1M a21 a22 a23 … a2(M-1) a2M a31 a32 a33 … a3(M-1)a3M………………ad1 ad2 ad3 … ad(M-1) adM]
A1An evaluation matrix representing route 1, wherein the number of attributes is d, the number of expert scoring systems is M, a11Representative expert System number 1 vs. 1 AttributeA value of credit of; in the same way, ad1Is the scoring of the d-th attribute by scoring system 1;
step 4.2: obtaining a top (S) -k set table via step 4.1, where S ═ 1,2,3, …, S; the Top(s) -k set table contains M Top(s) -k(m)Set, where M ═ 1,2,3, …, M;
performing Reverse Top-k operation on the set table Top(s) -k, wherein for an attribute set p with k being 2 and the number of attributes being d, { p [1], p [2], …, p [ d ] }, the RTop-k set of the ith attribute number p [ i ] is:
Figure BDA0003057081320000041
in the formula: s is the number of the route (S is 1,2,3, …, S), and p [ i ] is the ith attribute number of the route (i is 1,2,3, …, d); sequentially calculating the evaluation similarity between each two attributes, and finally obtaining the dominance of each attribute;
Figure BDA0003057081320000042
and j ≠ i;
when k takes 2, the ith attribute p [ i [ ]]Has a dominance of | D(s) -2p[i]|；
Step 4.3: for evaluation matrix AsSequentially circulating the steps from S4.1 to S4.2 until k is d-1; the attribute dominance degree obtained by each circulation is simply normalized to obtain an attribute p [ i]Final dominance value | D(s)p[i]|；
Figure BDA0003057081320000043
The final dominance of the d attributes of the route s may form a d × M dominance matrix d(s);
d × M-dimensional attribute dominance matrix d(s):
D(s)＝[D(s)p[1] 0 0 … 0 0 0 D(s)p[2] 0 … 0 0 0 0 D(s)p[3] … 0 0………………0 0 0 … 0 D(s)p[d]]
the cloud platform data center calculates a new evaluation matrix A of the route according to the dominance matrixs′；
Figure BDA0003057081320000044
Wherein
Figure BDA0003057081320000045
Thereby obtaining
Figure BDA0003057081320000046
Step 4.4: for different evaluation matrixes A in sequencesS is 1,2,3, …, S, and step 4.1 to step 4.3 are repeated to obtain S1 × d evaluation matrices as', S-1, 2,3, …, S; the cloud platform data center evaluates S evaluation matrixes AsCombining into an S x d comprehensive evaluation matrix A, and transmitting the comprehensive evaluation matrix back to the terminal side;
obtaining a comprehensive evaluation matrix A ═ A1′，A2′，A3′，…，AS′]。
And 5, calculating the comprehensive ranking value of each route in the step as follows:
defining user U1Preference of (1) { u [1]]，u[2]，u[3]，…，u[d]In which u [ i ]]Representing user U1Preference weight on ith attribute;
each route corresponds to one row of the comprehensive evaluation matrix, and the comprehensive value f of each route is calculatedw(s)：
Figure BDA0003057081320000051
Wherein f isw(s)Represents the sum of the s-th route, p [ i ]]' score value of i-th column representing s-th row of comprehensive matrix A, calculating fw(s)And comparing the values, and finally, selecting the route with the highest score for driving.
The data fusion mode in the step6 is as follows:
classifying according to the collection date field in the collected data according to the seasons, months, weeks and days; and fusing data collected at any time during decision making and historical data corresponding to the time interval to serve as new original data to increase data reliability.
The method has the advantages that the method has obvious effect on the aspect of road congestion inhibition in the information physical system, effectively reduces the road congestion and further reduces the accident rate of driving; the parking rate and the waiting time in traffic driving are effectively reduced, and the travel cost is further reduced. The congestion of a traffic network is inhibited, and the traffic condition of urban main roads is improved.
Drawings
Fig. 1 is a flow chart of a traffic congestion suppression method based on group decision in an intelligent networking environment.
Fig. 2 is a schematic algorithm flow chart.
Fig. 3 is a scene model diagram based on traffic CPS.
Detailed Description
The invention aims to solve the technical problem of providing a new technical method in the field of traffic congestion inhibition by utilizing a multi-attribute group decision theory idea, relating to multiple technologies of multi-data acquisition (sensors, GPS, radar), data processing and group decision and being applicable to the field of automobile information physical systems (CPS). Preferred embodiments of the present invention will be further described with reference to fig. 1 to 3; the flow chart of the method of the invention is shown in figure 1. The invention utilizes a terminal sensor system in the networking environment to collect and upload the data to a cloud platform data center, and simultaneously, utilizes a traffic monitoring system of each crossing in a city to collect crossing traffic data. The cloud platform data center carries out data preprocessing on received data, including data dimension reduction, data redundancy removal and the like, obtains standardized original data, and scores the standardized original data by means of a plurality of road traffic auxiliary systems to form an evaluation matrix. And the cloud platform data center performs reassignment and normalization processing on the attribute weight of each trunk by using the proposed algorithm based on the evaluation matrix of each trunk to obtain a comprehensive evaluation matrix. And finally, the cloud platform data center transmits the comprehensive evaluation matrix back to the terminal side equipment. And the terminal side equipment receives the data returned by the cloud platform data center, calculates the comprehensive ranking value of each trunk according to the user preference and the comprehensive evaluation matrix data to obtain an optimal decision, and continuously uploads the latest data to the cloud platform data center. In addition, the cloud platform center acquires new traffic data acquired by the terminal side sensor system and traffic data in the intersection monitoring system in real time, and performs fusion processing with historical data. The method has an obvious effect on the aspect of road congestion inhibition in the Cyber Physical System (CPS), effectively reduces the road congestion and further reduces the accident rate of driving; the parking rate and the waiting time in traffic driving are effectively reduced, and the travel cost is further reduced. The congestion of a traffic network is inhibited, and the traffic condition of urban main roads is improved.
The algorithm flow chart of the method is shown in FIG. 2, and the steps are as follows:
step 1: and collecting and uploading the data to a cloud platform data center by using a terminal sensor system in the networking environment, and simultaneously collecting intersection traffic data by using traffic monitoring systems at all intersections in the city. Including vehicle mounted sensors, traffic system data, and GPS location information. Generally, it relates to: road section number ID and road section average speed VidDate of data collection DcoDate of data upload DupThe number N of the road passing through the intersectionsWaiting time T of red, yellow and green light on road sectionidTime H of travel of road sectionidRoad section running length SidAnd road segment video image data;
the raw data is stored in the following format:
name: link ID, code: SEGMENT _ ID, data type: string;
name: average speed V of road sectionidAnd the code is as follows: SEGMENT _ Vid, data type: number;
name: date of data acquisition DcoAnd the code is as follows: SEGMENT _ Dco, data type: date;
name: date of data upload DupAnd the code is as follows: SEGMENT _ Dup, data type: date;
name: number N of road crossingssAnd the code is as follows: SEGMENT _ Ns, data type: number;
name: waiting time T of red, yellow and green lightidAnd the code is as follows: SEGMENT _ Tid, data type: date;
name: road section driving time HidAnd the code is as follows: SEGMENT _ Hid, data type: date;
name: road section travel length SidAnd the code is as follows: SEGMENT _ Sid, data type: number;
name: road conditions video image path VI, code: SEGMENT _ PATH, data type: string;
name: valid sample, code: USED _ SAMPLE, data type: bool;
name: invalid sample, code: UNUSED _ SAMPLE, data type: bool.
Step 2: the cloud platform data center performs data preprocessing on the received data, including data dimension reduction, data redundancy removal and the like, so as to obtain standardized original data. The processed data metrics generally include: average speed of route
Figure BDA0003057081320000061
Total running length L of route and number L of traffic lights in road sectionSAverage waiting total time T of traffic lights of routewNumber of accidents on route NacThe road condition evaluation As (video image data dimension reduction) of the route road, the million vehicle kilometer accident rate P of the route road and the like;
the preprocessed data is stored in the following format:
name: average speed of route
Figure BDA00030570813200000712
Code: SEGMENT _ V, data type: number;
name: total length of route travel L, code: SEGMENT _ L, data type: number;
name: number of traffic lights LSAnd the code is as follows: SEGMENT _ Ls, data type: number;
name: average waiting total for route traffic lightsDuration TwAnd the code is as follows: SEGMENT _ Tw, data type: date;
name: number of accidents on route NacAnd the code is as follows: SEGMENT _ Nac, data type: number;
name: road condition assessment of the route As, code: SEGMENT _ As, data type: number;
name: million vehicle kilometers accident rate P of route roads, code: SEGMENT _ P, data type: number;
wherein the route average speed
Figure BDA0003057081320000071
n is the total number of vehicles traveling on the route, t is the number of road segments in the route; total length of travel of route
Figure BDA0003057081320000072
Average waiting total time of route traffic lights
Figure BDA0003057081320000073
The road condition evaluation As of the road is based on the video images collected by the vehicle-mounted recorder and the traffic system, and the million-vehicle-kilometer accident rate of the road
Figure BDA0003057081320000074
Step 3: the cloud platform scores the standardized raw data by means of a plurality of road traffic auxiliary systems to form an evaluation matrix (for incomplete evaluation matrices, filling by adopting an averaging method). D attributes (attribute set p of route is { p [1] of route) of standardized S routes (number of route options according to user position) by M intelligent transportation systems]，p[2]，…，p[d]}，p[i]I-th attribute representing route) and storing the scoring results in a matrix A of d × MsIn (S ═ 1,2,3, …, S), S evaluation matrices are finally obtained.
The basic scoring rules are as follows:
the congestion level of the route is as follows: when in use
Figure BDA0003057081320000075
When the running state is smooth;
Figure BDA0003057081320000076
when the running state is basically smooth;
Figure BDA0003057081320000077
when the vehicle is in a light congestion state, the running state is light congestion;
Figure BDA0003057081320000078
when the running state is moderate congestion;
Figure BDA0003057081320000079
when the vehicle is in a heavy congestion state, the running state is in a heavy congestion state.
Total length of route grade: the total length of the route L and the reverse data index need to be normalized into a forward evaluation index, and two methods are generally adopted: mode 1, let U be (total length L × number S of routes)/sum of all route lengths
Figure BDA00030570813200000710
Then
Figure BDA00030570813200000711
Where S is the number of route plans. When the U is more than 1.5, the route length grade is far; when U is more than 1.1 and less than or equal to 1.5, the route length grade is far; when U is more than 0.9 and less than or equal to 1.1, the route length grade is moderate; when U is more than 0.5 and less than or equal to 0.9, the route length grade is closer; when U is less than or equal to 0.5, the route length grade is close. Mode 2, when the total length L of the route is more than 15km, the route length grade is far; when L is more than 12km and less than or equal to 15km, the length grade of the route is far; when L is more than 8km and less than or equal to 12km, the length grade of the route is moderate; when L is more than 5km and less than or equal to 8km, the length grade of the route is closer; and when L is less than or equal to 5km, the route length grade is close.
Average waiting total duration grade of route traffic lights: because the traffic basic engineering development of each city is extremely unbalanced, the traffic network of different cities is different. China main city traffic analysis published together with the traffic bureau according to the Gaode mapThe report reported that the delay time of the traffic in Shanghai is 39 seconds/vehicle, and the delay time of the fertilizer combination is 25 seconds/vehicle. Therefore, the division of the average waiting total duration grade of the route traffic lights should be dynamically adjusted to correspond to the corresponding city. Here we take the Changsha city as an example: average waiting total time T of traffic lights on routewWhen the waiting time is more than 150 hours, the average waiting total time length grade of the route traffic lights is long; when the average waiting total time of the traffic lights of the route is 110 < TwWhen the waiting time is less than or equal to 150, the average waiting total time length grade of the route traffic lights is longer; when the average waiting total time length of the traffic lights of the route is 70 < TwWhen the waiting time is less than or equal to 110, the average waiting total time of the route traffic lights is moderate; when the average waiting total time of the traffic lights of the route is 30 < TwWhen the waiting time is less than or equal to 70, the average waiting total time length grade of the route traffic lights is shorter; average waiting total time T of traffic lights on routewAnd when the waiting time is less than or equal to 30, the average waiting total time duration grade of the route traffic lights is short.
Route road condition assessment As: the road condition evaluation grade is simple, and urban roads are generally processed by cement or asphalt. However, some roads over a long period of time may have potholes, cracks, and the like on the road surface. And performing data dimensionality reduction on video image data of the vehicle-mounted sensor, the automobile data recorder and the traffic system by using a learned convolutional neural network to obtain a road condition evaluation value. The higher the evaluation value, the flatter the road is.
Million-vehicle-kilometer accident rate of route roads
Figure BDA0003057081320000081
Wherein N isacIs the number of accidents occurring on the route. When P is larger than 3, the per million vehicle kilometer accident rate grade of the route road is high; when 1 < TwWhen the accident rate is less than or equal to 3, the per million vehicle kilometer accident rate grade of the route road is medium; when T iswAnd when the accident rate is less than or equal to 1, the per million vehicle kilometer accident rate grade of the route road is low.
Step 4: and the cloud platform data center performs reassignment and normalization processing on the attribute weight of each trunk by using the proposed algorithm based on the evaluation matrix of each trunk to obtain a comprehensive decision matrix. First, the evaluation matrix A is respectively aligneds(S ═ 1,2,3, …, S) to perform Top-k operations (usually k ═ 2 ∞d-1), the result is denoted as Top(s) -k(m)Set (Top-k results under the mth expert evaluation system corresponding to the s-th route). Secondly, for all of the routes Top(s) -k(m)The set forms a Top(s) -k set table, the Reverse Top-k operation is carried out on the Top(s) -k set table, and the obtained result is recorded as RTop(s) -kp[i](Reverse Top-k result for ith attribute corresponding to s-th route). All of RTop(s) -k for the routep[i]The sets form an RTop(s) -k set table. According to the RTop(s) -k set table obtained by Reverse Top-k operation, the evaluation similarity among the d attributes of the route s can be calculated, and therefore the dominance of each attribute is obtained. And (d-2) circulating for d-2 times in sequence (k is 2-d-1), and normalizing the attribute dominance degree obtained by each circulation to obtain an attribute weight value. Finally, the cloud platform data center obtains the evaluation matrix A of the s-th route again according to the attribute weight values' (S is 1,2,3, …, S), the evaluation matrices that have sequentially obtained S routes are combined into an S × d comprehensive evaluation matrix a, and the comprehensive evaluation matrix is returned to the terminal side.
The detailed calculation steps are as follows:
s4.1 evaluation matrix As(S-1, 2,3, …, S) performing Top-k operations involving the number of attributes in the evaluation matrix, the number of evaluation expert scoring systems, and for a given set of attributes, p-p [1] for a given number of attributes, d]，p[2]，…，p[d]}，p[i]Represents the ith attribute of the attribute p, wherein the value range of k is 2-d-1,
evaluation matrix a 1:
A1＝[a11 a12 a13 … a1(M-1) a1M a21 a22 a23 … a2(M-1) a2M a31 a32 a33 … a3(M-1)a3M………………ad1 ad2 ad3 … ad(M-1) adM]
A1an evaluation matrix representing route 1, wherein the number of attributes is d, the number of expert scoring systems is M, a11The value of the 1 st attribute is given by representative expert system number 1. In the same way, ad1Is the scoring system 1 for the d-th attributeScoring of (4);
if the expert number is M, k and 2 is selected, the top 2 attributes with the highest score of the route are selected for calculation, and M Top(s) -2 can be obtained from the evaluation matrix of the route(m)And (wherein M is 1,2,3, …, M; S is 1,2,3, …, S). When the route number is 4, the expert scoring system number is 1, and the Top-k set when k takes 2 is:
Figure BDA0003057081320000091
if the evaluation matrix A of route 44In the expert system number 1, the highest 2 items with scores are the 1 st and 3 rd attributes, and then Top (4) -2(1)＝{a11，a31}；
Assume now the evaluation matrix A of the number 4 route4Including M expert scoring systems, we can obtain a Top (4) -2 set including M Top (4) -2(m)；
Figure BDA0003057081320000092
Top(s)-k(m)The mth Top-k set for the S route (S ═ 1,2,3, …, S; M ═ 1,2,3, …, M; k ═ 2,3, …, d-1). M Top(s) -k(m)Forming a Top(s) -k set table;
top (4) -2 set table is { Top (4) -2(1)，Top(4)-2(2)，Top(4)-2(3)，…，Top(4)-2(M)}。
S4.2 after the above step S4.1, we can obtain 1 top (S) -k set table (where S ═ 1,2,3, …, S). A Top(s) -k set table containing M Top(s) -k(m)Set (where M ═ 1,2,3, …, M). Then, a Reverse Top-k operation is performed on the set table Top(s) -k, where for k being 2, the attribute set p with the number of attributes d is { p [1]]，p[2]，…，p[d]H, i-th attribute number p [ i ]]The RTop-k set of (A) is:
Figure BDA0003057081320000101
in the formula: s is the number of the route (S ═ 1,2,3, …, S), p [ i [ n ] ]]The ith attribute for the route is numbered (i ═ 1,2,3, …, d). RTop(s) -kp[i]P [ i ] th of route numbered s]An RTop-k set of attributes, finding the existence of p [ i ] from a Top(s) -k set table]The expert rating system number m of (a) constitutes RTop(s) -kp[i]. d RTops(s) -kp[i]Forming an RTop(s) -k set table;
the RTop (4) -2 set table is { RTop (4) -2p[1]，RTop(4)-2p[2]，RTop(4)-2p[3]，…，RTop(4)-2p[d]}；
Calculating the dominance degree among the d attributes of the route s according to the RTop(s) -k set table;
property p [ i ] in route s]And attribute p [ j]The evaluation similarity between them is | sim(s) -k(ij)|。
Figure BDA0003057081320000102
In the formula: i, j ∈ 1,2,3, …, d. When k is taken to be 2, the number of bits is,
Figure BDA0003057081320000103
the evaluation similarity between each two attributes is calculated in sequence, and finally the dominance of each attribute can be obtained,
Figure BDA0003057081320000104
and j is not equal to i,
when k takes 2, the ith attribute p [ i [ ]]Has a dominance of | D(s) -2p[i]|。
S4.3 evaluation matrix AsAnd sequentially circulating the steps from S4.1 to S4.2 until k is d-1. Finally, the attribute dominance degree obtained by each circulation is simply normalized to obtain an attribute p [ i [ ]]Final dominance value | D(s)p[i]|；
Figure BDA0003057081320000105
The final dominance of the d attributes of the route s may form a d × M dominance matrix d(s);
d × M-dimensional attribute dominance matrix d(s):
D(s)＝[D(s)p[1] 0 0 … 0 0 0 D(s)p[2] 0 … 0 0 0 0 D(s)p[3] … 0 0………………0 0 0 … 0 D(s)p[d]]
the cloud platform data center calculates a new evaluation matrix A of the route according to the dominance matrixs′；
Figure BDA0003057081320000106
Wherein
Figure BDA0003057081320000107
Thereby obtaining
Figure BDA0003057081320000108
S4.4 successively comparing different evaluation matrixes As(S ═ 1,2,3, …, S) by looping through steps S4.1 to S4.3, S1 × d evaluation matrices a can be obtaineds' (S-1, 2,3, …, S). The cloud platform data center evaluates S evaluation matrixes As' combining into an S x d comprehensive evaluation matrix A, and returning the comprehensive evaluation matrix to the terminal side, wherein the comprehensive evaluation matrix A is [ A ═ A1′，A2′，A3′，…，AS′]。
Step 5: the terminal side equipment receives data returned by the cloud platform data center, calculates comprehensive ranking values of all routes according to user preferences and comprehensive evaluation matrix data, and assumes a user U1Preference of (1) { u [1]]，u[2]，u[3]，…，u[d]In which u [ i ]]Representing user U1Preference weight on ith attribute. Each route corresponds to one row of the comprehensive evaluation matrix, so that the comprehensive value f of each route can be calculatedw(s)，
Figure BDA0003057081320000111
Wherein f isw(s)Represents the sum of the s-th route, p [ i ]]' represents the value of credit of the ith column of the s-th row of the comprehensive matrix A. Calculating fw(s)And comparing the values, and finally, selecting the route with the highest score for driving.
Step 6: the method comprises the steps that a cloud platform center obtains new traffic data collected by a terminal side sensor system and traffic data in a crossing monitoring system in real time, and performs data fusion processing with historical data;
the collected data is classified according to the date of collection field by season, month, week and day. And fusing data collected at any time during decision making and historical data corresponding to the time interval to serve as new original data to increase data reliability.
The method has an obvious effect on the aspect of road congestion inhibition in the Cyber Physical System (CPS), effectively reduces the road congestion and further reduces the accident rate of driving; the parking rate and the waiting time in traffic driving are effectively reduced, and the travel cost is further reduced. The congestion of a traffic network is inhibited, and the traffic condition of urban main roads is improved.
As shown in FIG. 3, the invention establishes a vehicle decision auxiliary model under a traffic CPS scene, acquires information of vehicles in the driving process through the sensors in FIG. 3, and each vehicle passes through the acquired real-time information and uploads the information to a cloud data platform. The resulting preprocessed data is assumed to be scored by four expert systems e1，e2，e3，e4The three routes { s } are scored according to respective scoring rules1，s2，s3And is given as an attribute set p ═ p [1]]，p[2]，p[3]，p[4]Score, p [ i ]]Four attributes representing the set of attributes p
Figure BDA0003057081320000112
L、TwAnd P. The results are shown in tables 1,2 and 3, respectively.
Table 1:
Figure BDA0003057081320000113
table 2:
Figure BDA0003057081320000121
TABLE 3
Figure BDA0003057081320000122
First, S.1 converts the above tables into evaluation matrices A, respectivelys(s＝s1，s2，s3) Evaluation matrix of Table 1
Figure BDA0003057081320000123
Comprises the following steps:
Figure BDA0003057081320000124
accordingly, the evaluation matrix of Table 2
Figure BDA0003057081320000125
Comprises the following steps:
Figure BDA0003057081320000126
evaluation matrix of Table 3
Figure BDA0003057081320000127
Comprises the following steps:
Figure BDA0003057081320000128
Figure BDA00030570813200001311
for the above s1Is evaluated by the evaluation matrix
Figure BDA00030570813200001310
Top(s) is obtained by carrying out Top-k operation (k takes values from 2 to d-1 in sequence)1) -2 set table:
Figure BDA00030570813200001312
Figure BDA00030570813200001313
s.2 at Top(s)1) -2 performing Reverse Top-k operation on the basis of the set table to obtain RTop(s)1) -2 a table of sets of the data,
RTop(s1)-2p[1]＝{e1，e3}；RTop(s1)-2p[2]＝{e3，e4}；
RTop(s1)-2p[3]＝{e2}；RTop(s1)-2p[4]＝{e1，e2，e4}。
s.3 according to RTop(s)1) -2 the aggregate table can calculate the route s1The dominance between the 4 attributes of (1),
attribute p [1]]And attribute p [2]]The evaluation similarity is
Figure BDA0003057081320000131
Similarly, the evaluation similarity between two attributes can be calculated,
Figure BDA0003057081320000132
Figure BDA0003057081320000133
s.4 calculating dominance | D(s) of each attribute when k takes 21)-2p[i]|，
Figure BDA0003057081320000134
Figure BDA0003057081320000135
S.5 evaluation matrix A1And sequentially circulating the steps from S.1 to S.4 until k is d-1. As above, when k takes 3, the dominance | D(s) of each attribute is obtained1)-3p[i]|，
Figure BDA0003057081320000136
Figure BDA0003057081320000137
Normalizing the attribute dominance degree obtained by each circulation to obtain an attribute p [ i]Final dominance value | D(s)1)p[i]|，
Figure BDA0003057081320000138
In the same way, can obtain
Figure BDA0003057081320000139
Route s1The final dominance of the 4 attributes of (a) may form a 4 x 4 dominance matrix D(s)1)，
Figure BDA0003057081320000141
The cloud platform data center calculates a new evaluation matrix of the route according to the dominance matrix
Figure BDA0003057081320000142
Figure BDA0003057081320000143
From the preceding formula
Figure BDA0003057081320000144
Namely, it is
Figure BDA0003057081320000145
The following can be obtained:
Figure BDA0003057081320000146
s.6 successively comparing different evaluation matrixes As(s＝s1，s2，s3) And circulating the steps from S1 to S5. It is possible to obtain,
Figure BDA0003057081320000147
finally, a 3 multiplied by 4 comprehensive evaluation matrix is obtained
Figure BDA0003057081320000148
1 st behavior route s of comprehensive evaluation matrix A1Evaluation matrix
Figure BDA0003057081320000149
Column i as p [ i ] of the way]The score of the attribute is calculated,
A＝[19.313 24.180 11.440 22.470 15.150 16.636 24.032 21.000 24.462 18.434 12.923 21.360]。
s.7 suppose user U1For four attributes
Figure BDA00030570813200001410
P has a preference of {0.25, 0.35, 0.3, 0.1}, and we can calculate the comprehensive value f of each routew(s)，
Figure BDA00030570813200001411
Wherein s ═ s1，s2，s3. Can obtain
Figure BDA00030570813200001412
From the result of the calculation
Figure BDA00030570813200001413
To see, route s is selected1Is the optimal solution.
The invention provides an auxiliary decision for travelers by using a group decision idea based on the problem of traffic jam inhibition in an intelligent networking environment, and the travelers are far away from a traffic jam area, so that the maximum throughput of a traffic trunk can be obtained without losing fairness. Under the background of the situation, the group decision idea is utilized to exert the intelligence of most people to guide the driving of the driving vehicle. The group decision is an integral process that multiple decision-makers participate in decision analysis and make decisions together in order to give full play to collective intelligence, and is more convincing. The decision-maker can reach consensus in the group decision process, so that the result is more convincing.
The present invention is illustrated by the above examples of the practice and advantages of the method of the present invention, but not limited thereto, and one skilled in the art can vary the details of the method without substantially departing from the scope of the invention as defined in the claims.
While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims (6)

1. A traffic jam restraining method based on group decision is characterized by comprising the following steps:
step 1: and analyzing the decision problem, and selecting an alternative route set S, wherein S is 1,2,3, … and S. Collecting road network traffic data by using a terminal sensor system in a network connection environment and uploading the road network traffic data to a cloud platform data center, and simultaneously collecting intersection traffic data by using traffic monitoring systems of all intersections in a city, wherein the intersection traffic data comprises vehicle-mounted sensors, traffic system data and GPS (global positioning system) position information;
step 2: the cloud platform data center carries out data preprocessing on received data, wherein the data preprocessing comprises data dimension reduction and data redundancy removal to obtain standardized original data, and a scheme evaluation attribute set P is determined as { P [1], P [2], … and P [ d ] };
and step 3: the cloud platform scores standardized original data by means of a plurality of road traffic auxiliary systems to form an evaluation matrix, d attributes of standardized S routes are evaluated and scored through M intelligent traffic systems, and an attribute set p of the routes is { p [1], p [2], …, p [ d ] }, and p [ i ] represents the ith attribute of the route; respectively storing the scoring results in a dXM matrix to finally obtain S evaluation matrices, and carrying out forward and dimension elimination processing on the attribute evaluation indexes;
and 4, step 4: the cloud platform data center performs reassignment and normalization processing on the attribute weight of each trunk by using the proposed iterative Reverse Top-k algorithm based on the evaluation matrix of each route to obtain a comprehensive decision matrix;
step 4.1: for evaluation matrix AsS1, 2,3, …, S, by Top-k, usually k 2 to d-1, and the result is reported as Top (S) -k(m)Gathering Top-k results under the mth expert evaluation system corresponding to the mth route;
step 4.2: for all of the routes Top(s) -k(m)The set forms a Top(s) -k set table, the Top(s) -k set table is subjected to Reverse Top-k operation, and the obtained result is recorded as RTop(s) -kp[i]Reverse Top-k result corresponding to i-th attribute of s-th route, RTop(s) -k for all routesp[i]An RTop(s) -k set table is formed by the sets, and the evaluation similarity among the d attributes of the route s can be calculated according to the RTop(s) -k set table obtained by Reverse Top-k operation, so that the dominance of each attribute is obtained;
step 4.3: d-2 times of circulation are carried out in sequence (k is 2-d-1), and the attribute dominance degree obtained by each circulation is normalized to obtain an attribute weight value;
step 4.4: the cloud platform data center obtains the evaluation row vector A of the s-th route again according to the attribute weight values', S is 1,2,3, …, S, obtaining the evaluation matrix of S routes in turn and merging the evaluation matrix into an S x d comprehensive evaluation matrix A, and returning the comprehensive evaluation matrix to the terminal side;
and 5: the terminal side equipment receives data returned by the cloud platform data center, calculates comprehensive ranking values of all routes according to user preferences and comprehensive evaluation matrix data to obtain an optimal decision, and continuously uploads latest data to the cloud platform data center;
step 6: and the cloud platform center acquires new traffic data acquired by the terminal side sensor system and traffic data in the intersection monitoring system in real time, and performs data fusion processing with historical data.
2. The method as claimed in claim 1, wherein in step2, the average speed of the route is determined according to the group decision
Figure FDA0003057081310000021
n is the total number of vehicles traveling on the route, t is the number of road segments in the route; total length of travel of route
Figure FDA0003057081310000022
Average waiting total time of route traffic lights
Figure FDA0003057081310000023
The road condition evaluation As of the road is based on the video images collected by the vehicle-mounted recorder and the traffic system, and the million-vehicle-kilometer accident rate of the road
Figure FDA0003057081310000024
3. The method of claim 1, wherein the group decision-based traffic congestion suppression method comprises,
the step3 comprises the following scoring rules:
the congestion level of the route is as follows: when in use
Figure FDA0003057081310000025
When the running state is smooth;
Figure FDA0003057081310000026
when the running state is basically smooth;
Figure FDA0003057081310000027
when the vehicle is in a light congestion state, the running state is light congestion;
Figure FDA0003057081310000028
when the running state is moderate congestion;
Figure FDA0003057081310000029
when the running state is severe congestion;
total length of route grade: the total length of the route L and the reverse data index need to be normalized into a forward evaluation index, and two methods are generally adopted:
mode 1, let U be (total length L × number S of routes)/sum of all route lengths
Figure FDA00030570813100000210
Then
Figure FDA00030570813100000211
Wherein S is the number of route solutions;
mode 2, million-vehicle-kilometer accident rate of route road
Figure FDA00030570813100000212
Wherein N isacIs the number of accidents occurring on the route; when P is present>3 hours, the per million vehicle kilometer accident rate grade of the route road is high; when 1 is<TwWhen the accident rate is less than or equal to 3, the per million vehicle kilometer accident rate grade of the route road is medium; when T iswAnd when the accident rate is less than or equal to 1, the per million vehicle kilometer accident rate grade of the route road is low.
4. The method for suppressing traffic congestion based on group decision as claimed in claim 1, wherein the detailed calculation step of step4 is as follows:
step 4.1: for evaluation matrix AsS1, 2,3, …, S, performs Top-k operations involving the number of attributes in the evaluation matrix, the number of evaluation expert scoring systems, and for a given set of attributes, p, d, { p [1],p[2],…,p[d]}，p[i]Representing the ith attribute of the attribute p, wherein the value range of k is 2-d-1;
evaluation matrix A1：
Figure FDA00030570813100000213
A1An evaluation matrix representing route 1, wherein the number of attributes is d, the number of expert scoring systems is M, a11The value of the 1 st attribute is given by the representative expert system number 1; in the same way, ad1Is the scoring of the d-th attribute by scoring system 1;
step 4.2: obtaining a top (S) -k set table via step 4.1, where S ═ 1,2,3, …, S; the Top(s) -k set table contains M Top(s) -k(m)Set, where M ═ 1,2,3, …, M;
performing Reverse Top-k operation on the set table Top(s) -k, wherein for an attribute set p with k being 2 and the number of attributes being d, { p [1], p [2], …, p [ d ] }, the RTop-k set of the ith attribute number p [ i ] is:
Figure FDA0003057081310000031
in the formula: s is the number of the route (S is 1,2,3, …, S), and p [ i ] is the ith attribute number of the route (i is 1,2,3, …, d);
sequentially calculating the evaluation similarity between each two attributes, and finally obtaining the dominance of each attribute;
Figure FDA0003057081310000032
when k takes 2, the ith attribute p [ i [ ]]Has a dominance of | D(s) -2p[i]|；
Step 4.3: for evaluation matrix AsSequentially circulating the steps from S4.1 to S4.2 until k is d-1; the attribute dominance degree obtained by each circulation is simply normalized to obtain an attribute p [ i]Final dominance value | D(s)p[i]|；
Figure FDA0003057081310000033
The final dominance of the d attributes of the route s may form a d × M dominance matrix d(s);
d × M-dimensional attribute dominance matrix d(s):
Figure FDA0003057081310000034
the cloud platform data center calculates a new evaluation matrix A of the route according to the dominance matrixs ′；
Figure FDA0003057081310000035
Wherein
Figure FDA0003057081310000036
Thereby obtaining
Figure FDA0003057081310000037
Step 4.4: for different evaluation matrixes A in sequencesS is 1,2,3, …, S, and step 4.1 to step 4.3 are repeated to obtain S1 × d evaluation matrices as', S-1, 2,3, …, S; the cloud platform data center evaluates the S dataPrice matrix AsCombining into an S x d comprehensive evaluation matrix A, and transmitting the comprehensive evaluation matrix back to the terminal side;
obtaining a comprehensive evaluation matrix A ═ A1′,A2′,A3′,…,AS′]。
5. The method for suppressing traffic congestion based on group decision as claimed in claim 1, wherein the step5 of calculating the comprehensive ranking value of each route comprises the following steps:
defining user U1Preference of (1) { u [1]],u[2],u[3],…,u[d]In which u [ i ]]Representing user U1Preference weight on ith attribute;
each route corresponds to one row of the comprehensive evaluation matrix, and the comprehensive value f of each route is calculatedw(s)：
Figure FDA0003057081310000041
Wherein f isw(s)Represents the sum of the s-th route, p [ i ]]' score value of i-th column representing s-th row of comprehensive matrix A, calculating fw(s)And comparing the values, and finally, selecting the route with the highest score for driving.
6. The method of claim 1, wherein the group decision-based traffic congestion suppression method comprises,
the step6 data fusion mode is as follows:
classifying according to the collection date field in the collected data according to the seasons, months, weeks and days; and fusing data collected at any time during decision making and historical data corresponding to the time interval to serve as new original data to increase data reliability.
CN202110502780.XA 2021-05-09 2021-05-09 Traffic jam suppression method based on group decision Active CN113283714B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110502780.XA CN113283714B (en) 2021-05-09 2021-05-09 Traffic jam suppression method based on group decision

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110502780.XA CN113283714B (en) 2021-05-09 2021-05-09 Traffic jam suppression method based on group decision

Publications (2)

Publication Number Publication Date
CN113283714A true CN113283714A (en) 2021-08-20
CN113283714B CN113283714B (en) 2023-09-29

Family

ID=77278358

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110502780.XA Active CN113283714B (en) 2021-05-09 2021-05-09 Traffic jam suppression method based on group decision

Country Status (1)

Country Link
CN (1) CN113283714B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107886720A (en) * 2017-11-07 2018-04-06 长沙理工大学 Interval two-type fuzzy set festival holiday-based urban traffic congestion transfer analysis method
CN110570654A (en) * 2019-09-16 2019-12-13 河南工业大学 road section traffic jam dynamic detection method based on immunity

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107886720A (en) * 2017-11-07 2018-04-06 长沙理工大学 Interval two-type fuzzy set festival holiday-based urban traffic congestion transfer analysis method
CN110570654A (en) * 2019-09-16 2019-12-13 河南工业大学 road section traffic jam dynamic detection method based on immunity

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
戢晓峰;黄永忠;何增辉;韩春华;: "面向诱导的交通状态信息提取方法", 计算机工程与应用, no. 25 *
李淼;谷峪;陈默;于戈;: "一种针对反向空间偏好top-k查询的高效处理方法", 软件学报, no. 02 *
陈剑飞;徐良杰;: "基于乘客心理的最佳公交乘车路线选择方法研究", 武汉理工大学学报(交通科学与工程版), no. 03 *

Also Published As

Publication number Publication date
CN113283714B (en) 2023-09-29

Similar Documents

Publication Publication Date Title
Li et al. Analysis of road traffic fatal accidents using data mining techniques
CN105913661B (en) A kind of express highway section traffic state judging method based on charge data
CN105865472B (en) A kind of navigation method based on optimum oil consumption
CN107680401B (en) Parking space satisfaction evaluation method based on interval two-type fuzzy set
CN104778834B (en) Urban road traffic jam judging method based on vehicle GPS data
CN104157139B (en) A kind of traffic congestion Forecasting Methodology and method for visualizing
CN103895649B (en) A kind of driver safety driving warning method
CN108256714A (en) A kind of wheelpath real-time risk assessment model based on car networking big data
CN108665191B (en) Subway station outside association area calculation method based on aggregation effect
CN107564313A (en) Highway traffic congestion notice system and forecasting methods
CN107622678A (en) A kind of intelligent traffic control system and its method based on image procossing
CN108985539B (en) Method and device for evaluating parking lot road planning
CN109272745A (en) A kind of track of vehicle prediction technique based on deep neural network
CN106816009A (en) Highway real-time traffic congestion road conditions detection method and its system
CN112784000A (en) Passenger searching method based on taxi track data
CN115662113A (en) Signalized intersection people-vehicle game conflict risk assessment and early warning method
CN103824450A (en) Special vehicle route planning method based on traffic state rule for large-scale activities
CN106548628A (en) The road condition analyzing method that a kind of view-based access control model space transition net is formatted
CN109489679A (en) A kind of arrival time calculation method in guidance path
CN114550445B (en) Urban area traffic safety state assessment method and device
CN113335293B (en) Highway road surface detection system of drive-by-wire chassis
CN113283714A (en) Traffic jam suppression method based on group decision
CN117173899A (en) Smart city data processing method
CN109034593B (en) Car-sharing priority management and control strategy implementation benefit evaluation method
CN115936541A (en) Safe road recommendation method and device

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