CN116151499A - Intelligent multi-mode intermodal route planning method based on improved simulated annealing algorithm - Google Patents

Intelligent multi-mode intermodal route planning method based on improved simulated annealing algorithm Download PDF

Info

Publication number
CN116151499A
CN116151499A CN202211541760.4A CN202211541760A CN116151499A CN 116151499 A CN116151499 A CN 116151499A CN 202211541760 A CN202211541760 A CN 202211541760A CN 116151499 A CN116151499 A CN 116151499A
Authority
CN
China
Prior art keywords
path
point
current
feasible
algorithm
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211541760.4A
Other languages
Chinese (zh)
Inventor
赵林度
李逸龙
孙国豪
瞿子栋
梁艺馨
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.)
Southeast University
Original Assignee
Southeast 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 Southeast University filed Critical Southeast University
Priority to CN202211541760.4A priority Critical patent/CN116151499A/en
Publication of CN116151499A publication Critical patent/CN116151499A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34—Route searching; Route guidance
    • G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34—Route searching; Route guidance
    • G01C21/3407—Route searching; Route guidance specially adapted for specific applications
    • G01C21/3423—Multimodal routing, i.e. combining two or more modes of transportation, where the modes can be any of, e.g. driving, walking, cycling, public transport
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34—Route searching; Route guidance
    • G01C21/3453—Special cost functions, i.e. other than distance or default speed limit of road segments
    • G01C21/3492—Special cost functions, i.e. other than distance or default speed limit of road segments employing speed data or traffic data, e.g. real-time or historical
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34—Route searching; Route guidance
    • G01C21/36—Input/output arrangements for on-board computers
    • G01C21/3697—Output of additional, non-guidance related information, e.g. low fuel level
    • 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
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • 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
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G06Q10/0835—Relationships between shipper or supplier and carriers
    • G06Q10/08355—Routing methods
    • 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)
  • Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Operations Research (AREA)
  • Automation & Control Theory (AREA)
  • Data Mining & Analysis (AREA)
  • Quality & Reliability (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Mathematical Physics (AREA)
  • Development Economics (AREA)
  • Software Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Biophysics (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Game Theory and Decision Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Probability & Statistics with Applications (AREA)

Abstract

The invention discloses an intelligent multi-type intermodal route planning method based on an improved simulated annealing algorithm, and belongs to the field of multi-type intermodal route planning; predicting running time by using a neural network model according to the individual information and the congestion information, constructing a road directed graph, and determining an initial feasible path, an algorithm initial annealing temperature, a minimum annealing temperature and algorithm iteration times; taking the initial feasible path as the current path to change so as to generate a new path, judging whether the path change accords with the tabu rule, and if not, searching the feasible path again; if yes, detecting whether the new path is better than the current path, if yes, accepting the new path, taking the new path as the current path to carry out the next iteration, otherwise, accepting or discarding the new path with probability, cooling, and entering the next iteration; and the new iteration result still changes the current path, judges whether the current path is accepted or not, then continues iteration until the temperature is smaller than the minimum annealing temperature, and ends the iteration, and the current path at the end is the current optimal path.

Description

Intelligent multi-mode intermodal route planning method based on improved simulated annealing algorithm
Technical Field
The invention belongs to the field of multi-mode intermodal route planning, and particularly relates to an intelligent multi-mode intermodal route planning method based on an improved simulated annealing algorithm.
Background
The multi-mode intermodal transportation is one of the transportation modes which are greatly propelled by the China in recent years, and the transportation speed is improved and the operation cost is reduced by organically combining two or more transportation modes, so that the aim of optimizing and configuring transportation capacity resources is fulfilled.
The optimization of the transportation speed and the transportation cost by the multi-type intermodal transportation is realized by reasonably planning a path, dynamically calculating the transportation time and the transportation cost and adjusting the transportation planning in real time. The traditional path optimization algorithm only considers the connectivity among different nodes, but does not consider multiple transportation modes among different nodes, namely more than two transportation modes possibly exist among the same two nodes, and the consumed time and cost of the different transportation modes are different, so that a neural network is needed to be used for considering the influence of the characteristics of the multiple transportation modes (carriers) on the running time when the algorithm is designed. Meanwhile, in the actual situation, as multiple transportation modes exist in multi-mode intermodal transportation, extra time and cost are consumed due to operations such as loading, unloading and transporting when different transportation modes are transported, and factors such as weather, holidays and the like can possibly cause the change of the original transportation speed and transportation cost, so that an algorithm needs to re-plan a transportation scheme according to the latest transportation speed and transportation cost before transportation is executed, and the scientificity and rationality of transportation are ensured. In practice, due to the influence of factors such as numerous nodes, diversity of transportation modes, transportation connection and the like, the traditional simulated annealing algorithm has long calculation time and low convergence speed, and an effective optimal transportation scheme cannot be timely provided.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide an intelligent multi-mode intermodal transportation path planning method based on an improved simulated annealing algorithm.
The aim of the invention can be achieved by the following technical scheme:
an intelligent multi-type intermodal route planning method based on an improved simulated annealing algorithm comprises the following steps:
s1, acquiring real running time, individual information and congestion information;
s2, constructing a travel time prediction model by using a Wide & Deep model according to the real travel time, the individual information and the congestion information;
s3, inputting individual information data provided by a logistics company and congestion information data provided by a traffic department into a running time prediction model, predicting running time of each road section, and obtaining transportation cost of each road section and transportation cost of each city according to quotation of the logistics company;
s4, constructing a road directed graph by using the driving time of each road section, the transportation cost of each road section and the transportation cost of each city obtained in the S3, and initializing various parameters of an algorithm, wherein the method comprises the following steps: initial annealing temperature, minimum annealing temperature, algorithm iteration number, temperature drop rate, maximum tabu length, tabu table;
s5, determining an initial feasible path according to a generation strategy of the initial feasible path, and taking the initial feasible path as a current path;
s6, arbitrarily selecting two nodes in the current path as modification nodes;
s7, judging whether the selection of the modified node accords with the tabu rule, if not, entering S6, and reducing the tabu length; if yes, randomly searching a new path between the two nodes, and calculating objective function values of the new path and the current path;
s8, judging whether to accept the new path according to the acceptance strategy of the new path; if yes, entering S9, otherwise, entering S10 without changing the current path;
s9, taking the new path as the current path to carry out subsequent iteration;
s10, judging whether the current temperature reaches the minimum annealing temperature given in S4, if so, entering S12, otherwise, entering S11;
s11, judging whether the iteration times reach the algorithm iteration times given in S4, if so, reducing the current temperature according to the temperature reduction rate given in S4, and switching to S6;
s12, ending iteration, wherein the current path is the current optimal path and is taken as the driving path;
and S13, in the running process, if the next node has a transfer behavior, entering into S3, recalculating the optimal path, and if the next node is the end point, finishing the running process.
Further, the individual information in S1 includes: load capacity, driving age, mileage, and travel record; the congestion information in S1 includes: road class, date of travel, weather conditions, and historical traffic.
Further, the step of constructing the travel time prediction model includes:
s21, the eigenvector b= [ b ] of the congestion information acquired in S1 1 ,b 2 ,…,b d1 ]Input to the Wide model, the mathematical formula is as follows:
Figure BDA0003978018590000031
Figure BDA0003978018590000032
phi (b) is the cross characteristic of congestion information for the Wide model coefficient;
Figure BDA0003978018590000033
c ki a boolean variable, 1 if feature i belongs to cross feature d, and 0 if it does not;
s22, inputting the feature vector u of the individual information into a Deep model, wherein the Deep model is a cyclic neural network model, and a hidden layer calculation formula is as follows:
a l+1 ＝σ(W l a l +β l )
o＝tanh(W 2 a 2 +β 2 )
l∈[0,1,2]in order to hide the number of layers,
Figure BDA0003978018590000034
for the activation function, o is the output layer, a l ,β l ,W l The activation function, the deviation value and the weight of the hidden layer I are respectively;
s23, merging the Wide model and the Deep model to establish the real driving time e i -s i The minimum loss function with the predicted time f (·) forms a logistic regression problem:
Figure BDA0003978018590000041
Figure BDA0003978018590000042
wherein t is i For road section p i Is used for the vehicle to travel in a vehicle,
Figure BDA0003978018590000043
and beta is a prediction error.
Further, in S5, the step of generating the initial feasible path specifically includes:
s51, S point is the starting point, t point is the end point, x ij To be a feasible direct point, I i For the i-th layer feasible direct point set, y i Is a search point;
s52, starting from the S point, using the S point as the search point y i Searching forAll feasible direct points x ij (can reach directly, the distance and the time are all feasible values, i= … m represents the retrieval layer number, j= … n represents the node sequence), and the method is randomly not repeated (ensures that all x is traversed ij ) Selecting a feasible direct point x ik ∈I i As the next retrieval point y i+1 ；
Repeating the step; up to y i+1 =t, at which a viable path is obtained starting from point s and ending at point t; or if all the feasible direct points are searched, outputting that no feasible paths from the s point to the t point exist currently.
Further, in S7, the method for searching the new path includes: the two nodes are brought into the generation strategy of the feasible paths in S5.
Further, the objective function value solving step of the new path and the current path is as follows:
1) Obtaining the transportation cost of each node of the path according to the road directed graph constructed in S4
Figure BDA0003978018590000044
Figure BDA0003978018590000045
Transfer cost x= { X 1 ,x 2 ,x 3 (ii) and transport time->
Figure BDA0003978018590000046
m epsilon highway, railway, aviation };
2) The specific objective function is as follows:
Figure BDA0003978018590000051
Figure BDA0003978018590000052
wherein min W is an objective function, a q For the transport decision variables, m q For transportation, w u And w c Respectively a time weight coefficient and a cost weight coefficient.
Further, in S8, the acceptance policy of the new path is: if the new path objective function value is better than the current path, accepting the new path; otherwise, calculating a probability function value i according to the current temperature and the objective function value difference, comparing the probability function value i with a random number i 'epsilon (0, 1) generated randomly, and accepting a new path if i > i', otherwise, not accepting the new path.
Further, the solving formula of the probability function value i is as follows:
Figure BDA0003978018590000053
Δ＝f 2 -f 1
wherein f 2 For the objective function value of the new path, f 1 T is the initial annealing temperature given by S4, which is the objective function value of the current path.
A computer storage medium storing a readable program which, when executed, performs the above method.
An apparatus, comprising: one or more processors, memory for storing one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the methods described above.
The invention has the beneficial effects that:
1. starting from the data collection of the bottom layer, a neural network prediction model is established aiming at the complexity and real-time performance of the real traffic condition, so that the effectiveness and the accuracy of constructing road directed graph traffic data are ensured;
2. the invention considers the extra cost brought by the transfer link, and before reaching the transfer node, the current planning path is calibrated according to the current traffic data, so that the scientificity and rationality of the transfer task are ensured;
3. in the invention, the tabu search is used for improving the convergence speed of the heuristic algorithm, so that the path planning task can be completed in a short time.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to those skilled in the art that other drawings can be obtained according to these drawings without inventive effort.
FIG. 1 is a flow chart of the multiple intermodal route planning of the present invention;
FIG. 2 is a flow chart of a Wide & Deep model of the present invention;
fig. 3 is a flow chart of the generation strategy of the initial viable path of the present invention.
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.
As shown in fig. 1, an intelligent multi-mode intermodal route planning method based on an improved simulated annealing algorithm comprises the following steps:
s1, acquiring individual information and congestion information;
obtaining historical travel x from vehicle positioning i At road section p i Is the departure time s of (2) i And time of arrival e i Thereby, the real driving time e can be obtained i -s i I is N, N represents the number of travel samples;
obtaining vehicle and driver information including load capacity, driving age, mileage, travel record, etc. as individual information based on the registered information
Figure BDA0003978018590000061
Representative individual information is d 1 A dimension feature vector; link p i Road (course) grade, date of travel (whether or not it isHoliday), weather conditions (whether bad) and historical traffic as congestion information +.>
Figure BDA0003978018590000071
Representing congestion information as d 2 And (5) maintaining the feature vector.
S2, constructing a travel time prediction model by using a Wide & Deep model;
as shown in fig. 2, the Wide & Deep model includes a Wide model and a Deep model, the Wide model is a generalized linear model, and the Deep model is a multi-layer perceptron;
the construction steps of the travel time prediction model are as follows:
s21, the feature vector of the congestion information acquired in S1
Figure BDA0003978018590000072
Input to the Wide model, the mathematical formula is as follows:
Figure BDA0003978018590000073
Figure BDA0003978018590000074
phi (b) is the cross characteristic of congestion information for the Wide model coefficient;
Figure BDA0003978018590000075
c ki a boolean variable, 1 if feature i belongs to cross feature d, and 0 if it does not;
s22, inputting the feature vector u of the individual information into a Deep model, wherein the Deep model is a cyclic neural network model, and a hidden layer calculation formula is as follows:
a l+1 ＝σ(W l a l +β l
o＝tanh(W 2 a 2 +β 2 )
l∈[0，1，2]in order to hide the number of layers,
Figure BDA0003978018590000077
for ReLU (activation function), o for output layer, a l ,β l ,W l The activation function, the deviation value and the weight of the hidden layer I are respectively;
s23, merging the Wide model and the Deep model to establish the real driving time e i -s i The minimum loss function with the predicted time f (·) forms a logistic regression problem:
Figure BDA0003978018590000079
Figure BDA00039780185900000710
wherein t is i For road section p i For example, if the speed limit is 120km/h for the whole length of a certain road section, t=1h),
Figure BDA00039780185900000711
the model coefficient is Deep, and beta is a prediction error; thus, a travel time prediction model is obtained, and the travel time is predicted.
S3, inputting individual information data provided by logistics companies and congestion information data issued by railway, shipping and aviation departments into a running time prediction model to predict the running time of each road section; the transport prices (costs) of each road section and the transit prices (costs) of each city are obtained from the quotations of the logistics companies.
S4, constructing a road directed graph by using the time and cost data obtained in the step S3, and initializing various parameters of an algorithm: the method comprises an initial annealing temperature, a minimum annealing temperature, algorithm iteration times at each temperature, a temperature drop rate, a maximum tabu length and a tabu table;
constructing a road directed graph by using traffic data (namely time and cost data obtained in the step S3), wherein the method is to construct transport matrixes with the same quantity according to the quantity of transport modes, wherein matrix rows and columns are nodes corresponding to a map, and whether the nodes are communicated, the transport mode, the path length and the cost are represented as values corresponding to rows and columns in the corresponding matrixes; if the value of the two nodes corresponding to the transport matrix is positive, the two nodes are not communicated in the transport mode;
it is worth pointing out that because of the dynamic nature of the actual situation, connectivity, distance and transportation cost between nodes also change in real time, so the algorithm should be based on the changeability of traffic data, so that each item of data can change along with the actual situation in the actual transportation process, so as to adapt to the continuously changing transportation cost, the cross-node transportation requirement, the actual requirement of manually opening up roads and the like, and the algorithm has more universality.
S5, determining an initial feasible path according to a generation strategy of the initial feasible path, and taking the initial feasible path as a current path;
as shown in fig. 3, the step of generating the initial feasible path specifically includes:
s51, S point is the starting point, t point is the end point, x ij To be a feasible direct point, I i For the i-th layer feasible direct point set, y i Is a search point;
s52, starting from the S point, using the S point as the search point y i Searching all feasible direct points x ij (can reach directly, the distance and the time are all feasible values, i= … m represents the retrieval layer number, j= … n represents the node sequence), and the method is randomly not repeated (ensures that all x is traversed ij ) Selecting a feasible direct point x ik ∈I i As the next retrieval point y i+1 ；
Repeating the step; up to y i+1 =t, at which a viable path is obtained starting from point s and ending at point t; or if all the feasible direct points are searched, outputting that no feasible paths from the s point to the t point exist currently.
S6, arbitrarily selecting two nodes in the current path as modification nodes.
S7, judging whether the selection of the modified node accords with the tabu rule, if not, entering S6, and reducing the tabu length; if yes, randomly searching a new path between the two nodes, and calculating objective function values of the new path and the current path;
wherein, the tabu rule refers to: adding the modified node into a given tabu list in S4 every time a new solution (path) is adopted, limiting subsequent re-modification, wherein the tabu length is the given maximum tabu length in S4, and reducing the corresponding tabu length and not modifying (not conforming to the tabu rule) if the tabu list exists when the modified node is selected; when the taboo length is 0, deleting the taboo length from the taboo table, and removing the modification restriction.
Randomly selecting two nodes from the current path, and finding a path between the two nodes, which is different from the original path, namely finding a new path; if no other paths are available between the two nodes except the original path, the other two nodes are replaced to continue to try until the nodes are found to meet the requirements; the method for searching the new path specifically comprises the following steps: bringing the two nodes into a generation strategy of the feasible paths in the S5 to find a new path;
in the path { y } 1 ,y 2 ,y 3 ,y 4 Examples are }. Obtaining the transportation cost of each node of the path according to the road directed graph constructed in S4
Figure BDA0003978018590000091
Transfer cost x= { X 1 ,x 2 ,x 3 (ii) and transport time->
Figure BDA0003978018590000092
Figure BDA0003978018590000093
m is { highway, railway, aviation }, the specific objective function is as follows:
Figure BDA0003978018590000094
Figure BDA0003978018590000095
wherein min W is an objective function, a q For the transport decision variables, m q For transportation, w u And w c Respectively time weight coefficientsAnd cost weighting coefficient
S8, judging whether to accept the new path according to the acceptance strategy of the new path; if yes, entering S9, otherwise, entering S10 without changing the current path;
wherein, the acceptance policy of the new path is: if the new path objective function value is better than the current path, accepting the new path; otherwise, calculating a probability function value i according to the current temperature and the objective function value difference, comparing the probability function value i with a random number i 'E (0, 1) generated randomly, and accepting a new path if i is greater than i', otherwise, not accepting the new path;
the probability function is basically as follows:
Figure BDA0003978018590000101
Δ＝f 2 -f 1
f 2 for the objective function value of the new path, f 1 T is the initial annealing temperature given by S4, which is the objective function value of the current path. The objective function value f of the new path is calculated by traversing all transportation modes through depth priority and finding out the optimal transportation mode under the current node sequence 2 。
S9, taking the new path as the current path to carry out subsequent iteration, and entering S10.
S10, judging whether the current temperature reaches the minimum annealing temperature given in S4, if so, entering S12, otherwise, entering S11.
And S11, judging whether the iteration times reach the algorithm iteration times given in the step S4, if so, reducing the current temperature according to the temperature reduction rate given in the step S4, and switching to the step S6 no matter whether the iteration times reach the algorithm iteration times or not.
And S12, ending the iteration, wherein the current path is the current optimal path and is taken as the driving path.
S13, if the next node has a transfer behavior in the driving process, entering into S3, and recalculating an optimal path to ensure scientificity and rationality of the transfer behavior; and if the next node is the end point, the driving process is completed.
It is worth noting that due to practical circumstancesThe magnitude of the objective function value is not determined, and if the same initial temperature and end temperature are used for each path planning, the algorithm may converge too quickly or too slowly, for example, when the difference of the objective function values is 1×10 2 In order of magnitude, if the initial and end temperatures are too high, it is difficult to accept the difference solution just at the beginning of the algorithm, and then a locally optimal solution is trapped. Therefore, in actual circulation, the temperature should be subjected to a suitable treatment so as to be in the same order of magnitude as the objective function value, so as to achieve an optimal convergence effect.
In addition, in the objective function value calculation method according to the present invention, due to the specificity of the multi-modal intermodal transportation, once the transportation mode is changed, a transfer cost, that is, the time and cost consumed by the handling operation during the transfer of the cargo, needs to be added to the target value. The transfer cost is different according to each city and each transportation mode, and is replaced in real time so as to meet the actual requirements.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The solutions in the embodiments of the present application may be implemented in various computer languages, for example, object-oriented programming language Java, and an transliterated scripting language JavaScript, etc.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the spirit or scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims and the equivalents thereof, the present application is intended to cover such modifications and variations.

Claims (10)

1. An intelligent multi-mode intermodal route planning method based on an improved simulated annealing algorithm is characterized by comprising the following steps:
s1, acquiring real running time, individual information and congestion information;
s2, constructing a travel time prediction model by using a Wide & Deep model according to the real travel time, the individual information and the congestion information;
s3, inputting individual information data provided by a logistics company and congestion information data provided by a traffic department into a running time prediction model, predicting running time of each road section, and obtaining transportation cost of each road section and transportation cost of each city according to quotation of the logistics company;
s4, constructing a road directed graph by using the driving time of each road section, the transportation cost of each road section and the transportation cost of each city obtained in the S3, and initializing various parameters of an algorithm, wherein the method comprises the following steps: initial annealing temperature, minimum annealing temperature, algorithm iteration number, temperature drop rate, maximum tabu length, tabu table;
s5, determining an initial feasible path according to a generation strategy of the initial feasible path, and taking the initial feasible path as a current path;
s6, arbitrarily selecting two nodes in the current path as modification nodes;
s7, judging whether the selection of the modified node accords with the tabu rule, if not, entering S6, and reducing the tabu length; if yes, randomly searching a new path between the two nodes, and calculating objective function values of the new path and the current path;
s8, judging whether to accept the new path according to the acceptance strategy of the new path; if yes, entering S9, otherwise, entering S10 without changing the current path;
s9, taking the new path as the current path to carry out subsequent iteration;
s10, judging whether the current temperature reaches the minimum annealing temperature given in S4, if so, entering S12, otherwise, entering S11;
s11, judging whether the iteration times reach the algorithm iteration times given in S4, if so, reducing the current temperature according to the temperature reduction rate given in S4, and switching to S6;
s12, ending iteration, wherein the current path is the current optimal path and is taken as the driving path;
and S13, in the running process, if the next node has a transfer behavior, entering into S3, recalculating the optimal path, and if the next node is the end point, finishing the running process.
2. The intelligent multi-modal transportation path planning method based on the improved simulated annealing algorithm according to claim 1, wherein the individual information in S1 includes: load capacity, driving age, mileage, and travel record; the congestion information in S1 includes: road class, date of travel, weather conditions, and historical traffic.
3. The intelligent multi-modal transportation path planning method based on the improved simulated annealing algorithm as claimed in claim 1, wherein the step of constructing the travel time prediction model comprises:
s21, the eigenvector b= [ b ] of the congestion information acquired in S1 1 ,b 2 ,…,b d1 ]Input to the Wide model, the mathematical formula is as follows:
Figure FDA0003978018580000021
Figure FDA0003978018580000022
phi (b) is the cross characteristic of congestion information for the Wide model coefficient;
Figure FDA0003978018580000023
c ki a boolean variable, 1 if feature i belongs to cross feature d, and 0 if it does not;
s22, inputting the feature vector u of the individual information into a Deep model, wherein the Deep model is a cyclic neural network model, and a hidden layer calculation formula is as follows:
a l+1 ＝σ(W l a l +β l )
o＝tanh(W 2 a 2 +β 2 )
l∈[0,1,2]in order to hide the number of layers,
Figure FDA0003978018580000024
for the activation function, o is the output layer, a l ,β l ,W l The activation function, the deviation value and the weight of the hidden layer I are respectively;
s23, merging the Wide model and the Deep model to establish the real driving time e i -s i The minimum loss function with the predicted time f (·) forms a logistic regression problem:
Figure FDA0003978018580000031
Figure FDA0003978018580000032
wherein t is i For road section p i Ideal travel time, w 2 ＝[w 1 ,w 2 ,…,w d2 ]And beta is a prediction error.
4. The intelligent multi-modal transportation path planning method based on the improved simulated annealing algorithm of claim 1, wherein in S5, the step of generating an initial feasible path specifically includes:
s51, S point is the starting point, t point is the end point, x ij To be a feasible direct point, I i For the i-th layer feasible direct point set, y i Is a search point;
s52, starting from the S point, using the S point as the search point y i Searching all feasible direct points x ij (can reach directly, the distance and the time are all feasible values, i= … m represents the retrieval layer number, j= … n represents the node sequence), and the method is randomly not repeated (ensures that all x is traversed ij ) Selecting a feasible direct point x ik ∈I i As the next retrieval point y i+1 ；
Repeating the step; up to y i+1 =t, at which a viable path is obtained starting from point s and ending at point t; or if all the feasible direct points are searched, outputting that the current direct points do not exist from the point s to the point tFeasible paths ending at points.
5. The intelligent multi-modal transportation path planning method based on the improved simulated annealing algorithm as claimed in claim 4, wherein in S7, the method for searching the new path is as follows: the two nodes are brought into the generation strategy of the feasible paths in S5.
6. The intelligent multi-modal transportation path planning method based on the improved simulated annealing algorithm as set forth in claim 5, wherein the objective function value solving step of the new path and the current path is as follows:
1) Obtaining the transportation cost of each node of the path according to the road directed graph constructed in S4
Figure FDA0003978018580000033
Figure FDA0003978018580000034
Transfer cost x= { X 1 ,x 2 ,x 3 (ii) and transport time->
Figure FDA0003978018580000035
m∈/>
Figure FDA0003978018580000041
2) The specific objective function is as follows:
Figure FDA0003978018580000042
Figure FDA0003978018580000043
wherein min W is an objective function, a q For the transport decision variables, m q For transportation, w u And w c Respectively time weightsA weight coefficient and a cost weight coefficient.
7. The intelligent multi-modal transportation path planning method based on the improved simulated annealing algorithm of claim 1, wherein in S8, the acceptance policy of the new path is: if the new path objective function value is better than the current path, accepting the new path; otherwise, calculating a probability function value i according to the current temperature and the objective function value difference, comparing the probability function value i with a random number i 'epsilon (0, 1) generated randomly, and accepting a new path if i > i', otherwise, not accepting the new path.
8. The intelligent multi-modal transportation path planning method based on the improved simulated annealing algorithm as claimed in claim 7, wherein the solving formula of the probability function value i is:
Figure FDA0003978018580000044
/>
Δ＝f 2 -f 1
wherein f 2 For the objective function value of the new path, f 1 T is the initial annealing temperature given by S4, which is the objective function value of the current path.
9. A computer storage medium storing a readable program, characterized in that the method according to any one of claims 1-8 is performed when the program is run.
10. An apparatus, comprising: one or more processors, memory for storing one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of claims 1-8.
CN202211541760.4A 2022-12-02 2022-12-02 Intelligent multi-mode intermodal route planning method based on improved simulated annealing algorithm Pending CN116151499A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211541760.4A CN116151499A (en) 2022-12-02 2022-12-02 Intelligent multi-mode intermodal route planning method based on improved simulated annealing algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211541760.4A CN116151499A (en) 2022-12-02 2022-12-02 Intelligent multi-mode intermodal route planning method based on improved simulated annealing algorithm

Publications (1)

Publication Number Publication Date
CN116151499A true CN116151499A (en) 2023-05-23

Family

ID=86337978

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211541760.4A Pending CN116151499A (en) 2022-12-02 2022-12-02 Intelligent multi-mode intermodal route planning method based on improved simulated annealing algorithm

Country Status (1)

Country Link
CN (1) CN116151499A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116468352A (en) * 2023-06-16 2023-07-21 跨越速运集团有限公司 Logistics departure time calculation method, device, equipment and storage medium
CN117077883A (en) * 2023-10-18 2023-11-17 南通钢安机械制造有限公司 Scheduling optimization method and system for cast steel production process

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116468352A (en) * 2023-06-16 2023-07-21 跨越速运集团有限公司 Logistics departure time calculation method, device, equipment and storage medium
CN116468352B (en) * 2023-06-16 2023-09-22 跨越速运集团有限公司 Logistics departure time calculation method, device, equipment and storage medium
CN117077883A (en) * 2023-10-18 2023-11-17 南通钢安机械制造有限公司 Scheduling optimization method and system for cast steel production process
CN117077883B (en) * 2023-10-18 2023-12-22 南通钢安机械制造有限公司 Scheduling optimization method and system for cast steel production process

Similar Documents

Publication Publication Date Title
CN116151499A (en) Intelligent multi-mode intermodal route planning method based on improved simulated annealing algorithm
WO2022262469A1 (en) Industrial park logistics scheduling method and system based on game theory
CN111860754A (en) AGV scheduling method based on ant colony and genetic algorithm
Harbaoui Dridi et al. Optimisation of the multi-depots pick-up and delivery problems with time windows and multi-vehicles using PSO algorithm
CN113537580B (en) Public transportation passenger flow prediction method and system based on self-adaptive graph learning
D’Souza et al. Pickup and delivery problem using metaheuristics techniques
Ting et al. Multi-vehicle selective pickup and delivery using metaheuristic algorithms
Kallestad et al. A general deep reinforcement learning hyperheuristic framework for solving combinatorial optimization problems
Xu et al. A dynamic scheduling method for logistics tasks oriented to intelligent manufacturing workshop
CN108596385A (en) The queuing strategy and device of vehicle, readable medium, logistics system
Sedehzadeh et al. Optimization of a multi-modal tree hub location network with transportation energy consumption: A fuzzy approach
Liu et al. Physical Internet-enabled E-grocery delivery Network: A load-dependent two-echelon vehicle routing problem with mixed vehicles
CN114358233A (en) Multi-AGV path planning problem optimization method and system based on double-hybrid particle swarm
Wang et al. Optimization of ride-sharing with passenger transfer via deep reinforcement learning
CN114444809A (en) Data-driven multi-target strip mine card path optimization method
Alaia et al. Optimization of the multi-depot & multi-vehicle pickup and delivery problem with time windows using genetic algorithm
Jun et al. Evolutionary neural network for learning of scalable heuristics for pickup and delivery problems with time windows
CN116629479A (en) AGV multi-target traffic control method and system for single-way collision-free AGV
Yang Research on Hybrid Quantum Genetic Algorithm Based on Cross-Docking Delivery Vehicle Scheduling
Chen et al. Multiconstrained network intensive vehicle routing adaptive ant colony algorithm in the context of neural network analysis
Wen et al. Hierarchical sarsa learning based route guidance algorithm
Bagherian et al. A swarm based method for solving transit network design problem
Gao et al. The stochastic share-a-ride problem with electric vehicles and customer priorities
Lin et al. Multi-task assignment of logistics distribution based on modified ant colony optimization
Guo et al. [Retracted] Transportation Path Optimization of Modern Logistics Distribution considering Hybrid Tabu Search Algorithm

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination