CN114399095A - Cloud-side-cooperation-based dynamic vehicle distribution path optimization method and device - Google Patents

Cloud-side-cooperation-based dynamic vehicle distribution path optimization method and device Download PDF

Info

Publication number
CN114399095A
CN114399095A CN202111640629.9A CN202111640629A CN114399095A CN 114399095 A CN114399095 A CN 114399095A CN 202111640629 A CN202111640629 A CN 202111640629A CN 114399095 A CN114399095 A CN 114399095A
Authority
CN
China
Prior art keywords
distribution
vehicle
road
time
cost
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
CN202111640629.9A
Other languages
Chinese (zh)
Other versions
CN114399095B (en
Inventor
李天才
文一凭
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 of Science and Technology
Original Assignee
Hunan University of Science and Technology
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 of Science and Technology filed Critical Hunan University of Science and Technology
Priority to CN202111640629.9A priority Critical patent/CN114399095B/en
Publication of CN114399095A publication Critical patent/CN114399095A/en
Application granted granted Critical
Publication of CN114399095B publication Critical patent/CN114399095B/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
    • G06Q10/04—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
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Development Economics (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Genetics & Genomics (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Biomedical Technology (AREA)
  • Artificial Intelligence (AREA)
  • Physiology (AREA)
  • Game Theory and Decision Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a dynamic vehicle distribution path optimization method and device based on cloud edge-end cooperation, and relates to the field of spare part logistics distribution. According to spare part requests of all demand positions and distribution resources of enterprises, before distribution starts, a genetic algorithm is used, and according to optimization targets such as distribution timeliness and overall consumed cost, corresponding path planning is conducted on all distribution demands. By the edge computing equipment at the edge of the road, the road condition can be monitored in real time, and whether the passing time of the current road section is changed greatly or not can be judged. The method of the invention adopts the edge end to process the road condition change, and adopts a dynamic passing time table and an improved A-x algorithm according to the actual situation to optimize and adjust the distribution path of the remaining distribution tasks of the vehicle in real time.

Description

Cloud-side-cooperation-based dynamic vehicle distribution path optimization method and device
Technical Field
The invention relates to the field of logistics distribution of industrial spare parts, in particular to a dynamic vehicle distribution path optimization method and device based on cloud edge-end cooperation.
Background
Industrial spare parts are guarantee materials for normal operation of large-scale mechanical equipment, effective and timely supply of the spare parts must be guaranteed, and the efficient spare part logistics distribution method has important significance for improving economic benefits of enterprises.
The quick response to the demand of industrial spare parts is a key factor for ensuring stable production of enterprises, so that the stagnation time of the spare parts in a logistics system is effectively shortened through reasonable planning of logistics distribution paths, the logistics cost is reduced, and the maximization of economic benefits is achieved.
With the rapid development of cloud computing, internet of things and electronic commerce, new technologies such as logistics distribution, cloud computing, internet of things and intelligent transportation are combined, and the method is different from a cloud distribution model of a traditional logistics distribution model. The cloud distribution mode solves the problem that the traditional logistics distribution mode is difficult to adapt to the modern logistics distribution requirements. However, with the rapid increase of the number of various mobile devices and the demand for computing, a traditional cloud computing centralized processing mode taking a cloud data center as a core faces the problems of large network transmission delay, high data transmission cost, computing security, privacy risks and the like. This is an inefficient way to meet the demand for computing services for mobile users, particularly users that require immediate response. Cloud collaboration provides a new approach to solving this problem.
With the increasing demand of people on the timeliness of logistics distribution, how to ensure that various logistics and distribution tasks are completed on time becomes one of the key problems of effectively saving distribution cost, and is also urgently needed to be solved by the current logistics transportation industry. The delivery path of the vehicle is optimized through analysis of each link of vehicle delivery, and meanwhile, the delivery path is dynamically adjusted after the delivery vehicle starts in consideration of the fact that uncertain factors such as road condition changes possibly affect delivery timeliness. The influence of factors such as road conditions and weather on the timeliness of the overall distribution is reduced through optimization, and the method is an important means for improving the logistics distribution efficiency and reducing the overall distribution cost. Due to a plurality of restrictive factors and uncertainty existing in the logistics distribution process of spare parts, it is very important to comprehensively consider the influence of dynamic uncertain factors such as traffic, weather, road conditions and the like on the distribution process.
An edge node contains computing equipment and communication equipment that can handle a range of computing tasks and return results to a particular location. Therefore, it is possible to determine changes in the road condition information using the edge device and to adjust the distribution route in time based on these changes.
The prior art mainly comprises the following steps: the method improves the defect that the convergence time of the traditional genetic algorithm is earlier, the traditional genetic algorithm is insufficient in local searching capability, the searching capability of a high-quality solution is optimized by integrating a hill-climbing algorithm, the road risk is quantified by adopting a fuzzy mathematical evaluation method, and the factors influencing the selection of the transportation path are considered from the aspect of mathematical estimation.
By analyzing the prior art, the following disadvantages are found. Most of the current optimization is to improve the rationality of the planned delivery route by improving the inherent defects of the algorithm or considering the static fixed road conditions in the route planning stage before the delivery starts. However, the factors such as the emergency situation during the period from the start of the delivery to the completion of the delivery are not considered and handled. However, these unexpected sudden road condition changes will affect the distribution effect to different degrees. Therefore, if the delivery route of the vehicle is not dynamically adjusted according to the change in the actual road condition during the delivery, the delivery cost and delivery time increase.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a dynamic vehicle distribution path optimization method and device based on cloud edge side cooperation aiming at the defects of the prior art, so that the risk that the time consumed for completing distribution tasks is greatly increased due to the change of road conditions is reduced.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: a dynamic vehicle distribution path optimization method based on cloud edge-side cooperation comprises the following steps:
s1, carrying out qualitative analysis on the distribution vehicle path planning problem, and determining distribution time and distribution cost as optimization targets of the first-stage path planning;
s2, considering the distribution capacity of a distribution center under the constraint of the cargo demand and the time window proposed according to demand points, and before a vehicle starts, optimizing the whole distribution path in advance at a cloud end by using a genetic algorithm according to an optimization target of first-stage path planning to obtain an initial distribution path planning scheme, wherein the distribution time spent on the whole initial distribution path planning scheme is the least, and the consumed distribution cost is the lowest;
s3, after the delivery vehicle starts, sensing and judging the road condition data, and updating the road network information of the corresponding road section in the initial delivery path planning scheme;
s4, dynamically adjusting the rest distribution paths according to the dynamically updated road network information and the positions of the distribution vehicles until all distribution tasks are finished, and calculating the time spent on distribution;
s5, optimizing and adjusting the route according to the road condition change, and updating the cost consumed by the current updated distribution route;
and S6, taking the distribution center as a starting point of the vehicles, taking the last distribution vehicle as a stop point of completing the distribution task, and dynamically adjusting the distribution vehicle route in the whole distribution area according to the steps S4 and S5 in the whole distribution process.
Aiming at the defects of the prior art, the invention provides a dynamic vehicle distribution path optimization method based on cloud edge-end cooperation, which processes the road condition through the cooperation of the cloud, the edge and the end, thereby dynamically adjusting the vehicle distribution path, reducing the influence of various influencing factors on the whole distribution effect in the distribution process, ensuring the time consumed by distribution and saving the distribution cost. Before distribution begins, according to distribution resources and distribution requirements, a first-stage planning is made on a distribution path by adopting a genetic algorithm at a cloud end. After the delivery is started, the change of the road condition is sensed and judged in real time through the processing capacity of the edge end, and the rest delivery path is dynamically optimized and adjusted according to the change of the road condition and is sent to the vehicle terminal. Through the cooperative cooperation of the cloud side ends, various dynamic influence factors on the road traffic condition after the distribution starts are well processed, the vehicle distribution path is dynamically optimized, the rationality of the whole distribution scheme is improved, the delivery time is guaranteed, and the distribution cost is reduced.
In step S1, the distribution cost is the sum of the vehicle transportation cost TC, the time window penalty cost PC, and the vehicle cost; wherein,
Figure BDA0003442514310000031
Figure BDA0003442514310000032
Figure BDA0003442514310000033
wherein K is the total number of vehicles, N is the total number of demand points to be accessed, cijFor the unit transportation cost between demand point i and demand point j, dijIs the distance, x, between demand point i and demand point jijkThe value of (1) is 0 or 1, the value of 1 represents that the vehicle k goes to the demand point j after leaving from the demand point i, otherwise, the value of 0 is obtained; a, b are distribution time window penalty coefficients, wikFor the waiting time, t, of vehicle k at demand point iikThe time when the vehicle k reaches the delivery point i, liThe latest service time window of the demand point i; x is the number of0jkThe value is 0 or 1, and when the value is 1, the vehicle k departs from the distribution center 0 to the demand point j.
The method comprises the steps of analyzing the planning problem of the vehicle distribution path, constructing a corresponding mathematical model, changing an abstract concept into a concrete model, and measuring and evaluating through numerical indexes. By determining the construction mode of the cost of each part in the distribution process and correspondingly establishing a mathematical calculation formula, a direct basis is provided for the optimization of multiple targets.
The specific implementation process of step S2 includes:
1) constructing a penalty function p (x):
Figure BDA0003442514310000034
Figure BDA0003442514310000035
wherein x represents the number of the corresponding population, T is a positive number, Du_maxRepresenting the maximum driving distance of the u vehicle type; n is the total number of demand points to be accessed;
2) decoding the penalty function p (x) to construct a chromosome corresponding to the penalty function p (x);
3) randomly generating an initial population according to spare part demand data, road network data and distribution resources;
4) and circularly evolving the initial population under the condition of not meeting the termination condition by virtue of selection, intersection and mutation operators until an optimal solution is generated to obtain an initial distribution path planning scheme.
By using genetic algorithm, on one hand, the time of the optimization process is controlled within a reasonable range, and on the other hand, the overall cost of the optimization result (i.e. the distribution path planning scheme) is optimally represented on a plurality of objective functions. And a penalty function is constructed, so that each individual can obtain corresponding genetic probability according to the quality of the individual in the iterative optimization process, and the individual can participate in the stages of selection, crossing and mutation operators with proper probability, thereby ensuring the convergence of the algorithm.
The specific implementation process of step S3 includes:
A) calculating the change of the passing time of the road section to which the edge node belongs according to the road condition change information acquired by each edge node, and updating the current passing time of each road section in the road passing time table according to the calculation result;
B) updating the current position of the delivery vehicle according to the vehicle speed, obtaining the road section which is required to pass through by the traveling route in the initial delivery path planning scheme between the two positions by combining the position of the next demand point to be reached, and completing the updating of the road network information of the corresponding road section in the initial delivery path planning scheme;
C) according to the current positions of all the distribution vehicles and the next distribution position to be moved to, a road traffic schedule is inquired, and whether the changed road condition information can cause bad influence on the current distribution tasks of all the distribution vehicles is judged: if the road passing time after the vehicle is about to go to the road section is greater than N times of the original passing time (N can be set to a specific value according to the actual situation), it is determined that the change of the road passing time causes a bad influence on the distribution task, and the step S4 is entered.
By judging the road condition change, the adverse effect of the road condition change on the distribution result is determined, the setting can be adapted to different conditions by changing the threshold (namely the size of N), and the sensitivity of the proposed method is ensured.
The specific implementation process of step S4 includes:
I) taking the node with the minimum value of f (n) as the next node on the optimal path, wherein f (n) is g (n) + h (n); g (n) is the actual traffic cost from the starting node to the current node, and h (n) is the estimated value of the traffic cost from the current node to the destination;
II) operating the P table and the Q table maintained by the A-star algorithm, which specifically comprises the following steps:
i) setting the P table and the Q table to be null, adding the starting point S into the P table, setting the value of g (n) to be 0, setting the father node to be null, and setting the value of g (n) of other nodes in the road network to be infinity;
ii) if the P table is empty, the algorithm fails, otherwise, a node with the minimum f (n) value in the P table is selected and marked as BT, and the BT is added into the Q table; judging whether BT is an end point T, if yes, turning to step iii);
otherwise, finding each adjacent node NT of the BT according to the topology attribute of the road network and the traffic rule, and executing the following steps:
calculating the heuristic value of NT
f(NT)＝g(NT)+h(NT)；
g(NT)＝g(BT)+cost(BT，NT)；
Wherein cost (BT, NT) is the traffic cost from BT to NT;
if NT is in the P table and the passage cost value calculated by the formula g (NT), g (BT) + cost (BT, NT) is smaller than that of NT, the passage cost value of NT is updated to g (NT), g (BT) + cost (BT, NT), and the parent node of NT is set to BT;
(iii) if NT is in Q table and the passage cost value calculated by g (NT) ═ g (BT) + cost (BT, NT) is smaller than that of NT, update the passage cost value of NT to g (NT) ═ g (BT) + cost (BT, NT), set the parent node of NT as BT, and move NT out to P table;
if the NT is not in the P table or the Q table, setting the father node of the NT as BT and moving the NT to the P table;
returning to the step ii);
iii) backtracking from the end point T, sequentially finding father nodes, and adding the father nodes into the optimized path until the starting point S to obtain the optimized path;
III) calculating the passing time required by the vehicle to pass through the optimized path, wherein the optimized path comprises a plurality of road sections, and the number of the road sections is 1, 2, 3.., k; to [ tk, tk']Representing the transit time T of a vehicle through a section kkThen T isk＝tk’-tk(ii) a Passage time and T 'taken by vehicle to pass through multiple road sections'k1，T′k2，T′k3.., corresponding; f. ofkIndicating a time period corresponding to the moment when the vehicle passes the start of the section k,
Figure BDA0003442514310000051
Figure BDA0003442514310000052
calculating the transit time Tk of the vehicle through the section k by using the following formula:
Figure BDA0003442514310000053
the value of m satisfies the following constraints:
Figure BDA0003442514310000054
Δ T denotes the period length, T0Represents a start time;
IV) calculating the time taken to deal with the road condition change using the following equation: gn(r)＝bn(r)+cn(r)+kn(r)+hn(r)；
Figure BDA0003442514310000061
Figure BDA0003442514310000062
θn，n′Indicating data from vehicle VnTo the vehicle Vn′The number of vehicles passed;
Figure BDA0003442514310000063
Figure BDA0003442514310000064
Figure BDA0003442514310000065
m denotes the number of edge nodes, D denotes the set of edge computing devices, D ═ D1，d2，...，dmV denotes a vehicle set, V ═ V1，v2，...，vn}，λV2VDenotes a data transmission rate, λ, based on the V2V methodV2IIndicating the data transmission rate, ω, based on the V2I methodnRepresenting the amount of data of the computational task to be transferred, vnIndicating a delivery vehicle numbered n, dmEdge computing device denoted m,/nIndicating whether the computing task n is being processed on an edge computing device, lnThe value of 0 or 1, p represents the processing power of each edge computing device, unIs to calculate the number of resources requested by task n;
v) calculating the time required for completing the whole distribution task by using the following formula:
Figure BDA0003442514310000066
k represents the total number of links passed by the route numbered r _ s(ii) a And R represents the total number of the processed road events which cause adverse effects on the road traffic time due to the road condition change until the distribution is completed.
The improved A-algorithm is adopted to re-plan the local path to be adjusted, so that the solving speed of re-optimization is ensured, the calculation time is controlled within a certain range, the re-optimization result conforms to the actual situation, and the adjustment reasonableness is ensured. The contradiction between time consumption and solved quality is reconciled as a whole. By using the V2I and V2V methods, the transmission speed and the transmission efficiency of road condition data and adjustment results among edge nodes and between the edge nodes and vehicles are ensured, and the traffic reliability is ensured.
The specific implementation process of step S5 includes:
a) all the delivery vehicles start delivery according to the initial path planning;
b) dividing demand points needing to be distributed into a plurality of distribution areas according to r x r, numbering each distribution area, and taking r as unit length;
c) the method comprises the steps that edge nodes monitor changes of road conditions, events influencing the road conditions are recorded into event lists according to time lengths, each event in the event lists influences the passing time of roads leading to required points in each distribution area in different sizes, namely, the passing time of the road of a road section in the distribution area where the event occurs is increased to be original e according to the influence degree of the event type corresponding to each eventiMultiple (e)iCan be set according to actual use requirements);
d) changing the road passing time of the road sections in the corresponding distribution area into the value after influence;
e) judging whether the original planned path needs to be adjusted in the current time slice, namely judging whether the changed road condition information can cause adverse effects on the current distribution tasks of the distribution vehicles: if the road passing time of the vehicle about to go to the road section after the change is more than N times of the original passing time (N can be set to a specific numerical value according to the actual condition), namely the change of the road condition information causes bad influence on the distribution task, the adjustment is needed; if the adjustment is needed, re-planning is carried out according to an algorithm for searching the shortest path between two points;
f) and repeating the steps a) to f) for each other time segment until all the distribution tasks are completed, namely all the vehicles traverse all the demand points in the initial planning and return to the distribution center.
By dividing the distribution area, the monitoring range of each edge device is determined, the calculation difficulty of road condition processing is reasonably reduced, each edge device can sense and process the road condition change in the area in which the edge device is responsible, the calculation capacity of the edge device is not exceeded, and the calculation load is reasonably shared.
The invention also provides a computer device, comprising a memory, a processor and a computer program stored on the memory; characterized in that the processor executes the computer program to carry out the steps of the method of the invention.
The present invention also provides a computer program product comprising a computer program/instructions; which when executed by a processor implement the steps of the method of the present invention.
The present invention also provides a computer readable storage medium having stored thereon a computer program/instructions; which when executed by a processor implement the steps of the method of the present invention.
Compared with the prior art, the invention has the beneficial effects that: the method and the system consider a plurality of factors influencing road traffic conditions in the distribution area from the beginning of distribution to the completion of distribution, reduce the risk that the time consumed for completing the distribution task is greatly increased due to the change of road conditions, reduce the distribution cost, bring economic benefits to the logistics distribution of enterprises and reduce the energy consumption.
Drawings
FIG. 1 is a flow chart of a method according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a cloud communication process according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention comprises the following steps:
step 1: and carrying out qualitative analysis on the distribution vehicle path planning problem, and determining distribution time and cost as an optimization target of the first-stage path planning. The distribution cost is composed of vehicle transportation cost TC, time window penalty cost PC and vehicle cost.
Wherein, TC includes transportation fuel consumption and vehicle maintenance cost, etc., the part is proportional to the length of the distribution route, the calculation mode is as follows:
Figure BDA0003442514310000081
cijfor the unit transportation cost between demand point i and demand point j, dijIs the distance, x, between demand point i and demand point jijkThe value of (a) is 0 or 1, the value of 1 indicates that the vehicle k goes to the demand point j after leaving from the demand point i, otherwise, the value of 0. PC represents the penalty cost required to be added when the vehicle exceeds the time window requirement, and the calculation method is as follows:
Figure BDA0003442514310000082
a, b are distribution time window penalty coefficients, wikFor the waiting time, t, of the vehicle k at the delivery point iikThe time when the vehicle k reaches the delivery point i, liThe latest service time window for point i is delivered.
The vehicle cost is proportional to the number of vehicles VN scheduled for delivery, calculated as follows:
Figure BDA0003442514310000083
x0jkthe value is 0 or 1, and when the value is 1, the vehicle k departs from the distribution center 0 to the demand point j.
Step 2: under the constraint of the cargo demand and the time window proposed according to the demand point and in consideration of the distribution capacity of the distribution center, the whole distribution path is optimized in advance at the cloud end by using a genetic algorithm according to the existing data before the vehicle starts.
Step 2.1: formulating a multi-objective optimization function according to the requirements of time window, lowest total distribution cost and the like;
the method adopts a penalty function method to process the constraint condition. The basic idea of the penalty function method is: for an individual who cannot solve in a solution space, if the fitness of the individual is calculated, a penalty function needs to be given to the fitness function of the individual to reduce the fitness, so that the probability of being transmitted to the next generation is reduced, and the individual is eliminated automatically. For the features of the model herein, among others, the vehicle capacity constraint, the maximum distance traveled constraint may be implemented in the form of a penalty function. Is formulated as follows:
Figure BDA0003442514310000091
where p (x) is a penalty function, x represents the number of the corresponding population, where T is a large positive number.
Step 2.2: coding the solution of the problem described in the previous step to construct a chromosome corresponding to the solution;
the method uses natural number symbols to encode the solution to the problem. Natural number encoding is performed by encoding the solution to the problem (i.e., the set of vehicle paths) as an array of natural numbers of length k + m +1, with each chromosome of the encoding representing an initial solution in the problem solution space. Examples are as follows: considering {0, 4, 7, 9, 11, 0, 1, 3, 5, 2, 0, 6, 8, 10, 0} chromosome code, which represents that three vehicles are arranged to complete the distribution tasks of 11 spare part demand points, the distribution sub-paths of the three vehicles are respectively: route 1: 0- > 4- > 7- > 9- > 11- > 0; the path is 2: 0- > 1- > 3- > 5- > 2- > 0; the path is 3: 0- > 6- > 8- > 10- > 0; wherein, the number 0 represents the distribution center, 1-11 represent the numbers of 11 demand points, and each vehicle starts from the distribution center, and returns to the distribution center after completing the distribution task in turn.
Step 2.3: randomly generating an initial population according to spare part demand data, road network data and distribution resources;
step 2.4: carrying out cyclic evolution on the initial population under the condition of not meeting a termination condition through selection, intersection and mutation operators until an optimal solution is generated;
when the method is applied to solve the problem of planning the spare part vehicle distribution path, the parameters are set as follows: the population size Popsize is 80, the maximum iteration generation Maxgen is 500, the crossing rate Pc is 0.88, and the variation rate Pm is 0.05. The termination conditions were set as: if the iteration times reach 500 generations; or if the fitness value of a certain generation of chromosome reaches 0.9 times of the initial optimal chromosome fitness value, terminating the search and outputting an optimal solution.
Step 2.5: and decoding to obtain an initial distribution path planning scheme.
And step 3: after the delivery vehicle starts, the relevant data of the road condition change are collected through edge equipment: the road condition data are sensed and judged through the edge equipment, the road network data of the corresponding road section are updated, and a basis is provided for dynamically adjusting the local distribution path.
Step 3.1: processing the road condition change information acquired by the edge equipment, and calculating the change of the passing time of the road section to which the edge equipment belongs;
step 3.2: and dynamically updating the road network data and the road passing schedule. Updating the current position of the distributed vehicle according to the vehicle speed by taking 0.5h as unit time delta T, and updating the current passing time of all road sections in the road passing schedule according to the data acquired and processed from the edge nodes;
step 3.3: and judging the road condition change information and the current positions and the next distribution positions of all the distribution vehicles through the edge equipment, and providing a basis for the next adjustment.
And 4, step 4: dynamically adjusting the predetermined path based on the improved a-algorithm: and (4) dynamically adjusting the rest distribution paths until all distribution tasks are finished according to the dynamically updated road network information provided by the edge equipment and the positions of the distribution vehicles by running an improved A-algorithm on the edge equipment.
Step 4.1: and under the condition that the driving path needs to be changed from the current distribution point to the next distribution point (in the original path planning), searching for the shortest path between the two points to realize the purpose. And (4) replanning the path of the current demand point to be adjusted and the next demand point by using an A-x algorithm, namely searching the shortest path between the two points. The concrete implementation is as follows:
the a-algorithm is a typical heuristic search algorithm, which is based on the Diikstra algorithm and is widely used for finding the shortest path between two points.
The algorithm A is mainly characterized in that a heuristic evaluation function is maintained:
f(n)＝g(n)+h(n)
wherein, f (n) is the corresponding heuristic function of the algorithm when each node is searched. The method comprises two parts, wherein the first part g (n) is the actual traffic cost from the starting node to the current node, and the second part h (n) is the estimated value of the traffic cost from the current node to the destination. And (3) selecting the node with the minimum value of f (n) as the next node on the optimal path each time the algorithm expands.
And 4.2, operating the P table and the Q table maintained by the A-star algorithm. The a algorithm maintains two sets: p table and Q table. The P table stores nodes which are searched but not added to the optimal path tree; the Q table maintains those nodes that have joined the optimal path tree.
(1) And (4) setting the P table and the Q table to be null, adding the starting point S into the P table, setting the g value to be 0, setting the father node to be null, and setting the g values of other nodes in the road network to be infinity.
(2) If the P table is empty, the algorithm fails. Otherwise, selecting the node with the minimum f value in the P table, marking as BT, and adding the BT into the Q table. Judging whether BT is an end point T, if yes, turning to the step (3); otherwise, finding each adjacent node NT of the BT according to the topology attribute of the road network and the traffic rule, and performing the following steps:
calculating the heuristic value of NT
f(NT)＝g(NT)+h(NT)
g(NT)＝g(BT)+cos t(BT，NT)
Where cost (BT, NT) is the traffic cost from BT to NT.
② if NT is in the P table, and g value calculated by g (NT) ═ g (BT) + cos t (BT, NT) is smaller than the original g value of NT, updating the g value of NT to the result of equation (3), and setting the parent node of NT as BT.
③ if NT is in Q table and g value calculated by g (NT) ═ g (BT) + cos t (BT, NT) is smaller than the original g value of NT, then the g value of NT is updated to g (NT) ═ g (BT) + cos t (BT, NT) result, the parent node of NT is set as BT, and NT is moved out to P table.
And fourthly, if the NT is not in the P table or the Q table, setting the parent node of the NT as BT and moving the NT to the P table.
And fifthly, turning to the step (2) to continue execution.
(3) And backtracking from the end point T, sequentially finding father nodes, adding the father nodes into the optimized path until the starting point S, and obtaining the optimized path.
Step 4.3: the dynamic variation of the transit time is calculated and processed. In order to calculate the passing time of a certain road section under the real-time condition, a structure of a road passing time table is adopted, as shown in table 1. The table stores the current passing time of the road and the predicted values of the passing time at a plurality of future times.
With t0The starting time is shown, a period of time in the future is divided into a plurality of periods, the length of one period is shown by delta T, and the time when the system starts to work belongs to the first period. These time periods are then numbered, e.g., 1, 2, 3, 4. Similarly, each road segment is also numbered 1, 2, 3, 4. By TijRepresenting the transit time for segment i during time period j. Thus, the passing time of different road sections at different moments can be obtained.
TABLE 1 dynamic road passage time table
Figure BDA0003442514310000111
When the road condition changes, namely different road events occur, the re-planning system judges whether the current distribution route needs to be changed, if the changed road passing time is more than N times of the original passing time, the re-planning system needs to perform route re-planning. Optimization roadThe path may contain a plurality of road segments, which are numbered 1, 2, 3. By [ t ]k，tk′]Representing the transit time T of a vehicle through a section kkThen T isk＝tk′-tk. The vehicle may spend a number of slots through segment k, with these slots being related to transit time T'k1，T′k2，T′k3,..
Firstly, calculating a time period f corresponding to the moment when the vehicle passes through the starting point of the road section kk：
Figure BDA0003442514310000112
Then the corresponding can be found:
Figure BDA0003442514310000121
according to different values of the time interval length delta T, the road length L and the road passing speed, the situation that a vehicle can pass through a road section only in one time interval or can pass through a plurality of time intervals possibly occurs. The specific formula for the vehicle to pass through the section k can thus be found as follows:
Figure BDA0003442514310000122
wherein, the value of m satisfies the following constraint:
Figure BDA0003442514310000123
and 5: the computing edge device processes the road information changes and transmits the results to the time of the destination node through the communication network between vehicles in the area.
Step 5.1: b, calculating the time required for transmitting the road condition change information to the edge server as bn(i) The calculation method is as follows:
Figure BDA0003442514310000124
Figure BDA0003442514310000125
where N represents the number of vehicles on the current road segment, M represents the number of edge computing devices erected, D represents the set of edge computing devices, and D ═ D1，d2，...，dmV denotes a vehicle set, V ═ V1，v2，...，vn}，λV2VDenotes the data transmission rate, λ, based on the V2V techniqueV2IIndicating the data transfer rate, ω, based on the V2I techniquenRepresenting the amount of data of the computational task to be transferred, vnIndicating a delivery vehicle numbered n, dmRepresenting an edge computing device numbered m;
step 5.2: the time when the computing task is sent to the edge server is cn(i) The calculation method is as follows:
Figure BDA0003442514310000126
step 5.3: calculating the time k for performing path re-planning at the edge servern(i) The calculation method is as follows:
Figure BDA0003442514310000127
step 5.4: calculating the time h for sending the re-planned path back to the vehiclen(i) The calculation method is as follows:
Figure BDA0003442514310000128
step 5.5: the overall elapsed time is determined, and the calculation method is as follows:
gn(i)＝bn(i)+cn(i)+kn(i)+hn(i)
step 6: and optimizing and adjusting the route according to the road condition change, and updating the cost consumed by the updated distribution route.
And 7: starting from the distribution center, the vehicles are not delivered by the last distribution vehicle, and the distribution vehicle route in the whole distribution area is dynamically adjusted according to the methods of the steps 3, 4 and 5 in the whole distribution process. For processing convenience, before the operations of the following steps are implemented, we propose the following two assumptions:
(1) the logistics distribution center and the customer nodes are accessible through road combinations in a road network;
(2) a change in road conditions occurs in a grid area, and the transit time of a road to a demand point in the area is affected to different degrees in the time period (depending on the type of the event).
Step 7.1: all the delivery vehicles start delivery according to the initial path planning;
step 7.2: dividing the demand points to be distributed into regions according to r x r, numbering, wherein r is unit length, and determining specific numerical values according to the computing capacity of the edge equipment;
step 7.3: monitoring the change of the road condition by the edge node, recording events influencing the road condition into an events list according to a segment with the time length of 0.5h, wherein each event has different influences on the passing time of a road leading to a required point in the area: according to the influence degree of the event type corresponding to each event, the road passing time of the road section in the area where the event occurs is increased to the original eiMultiple (e)iSpecific values can be set according to actual conditions), the specific considered road events are listed as shown in table 2;
TABLE 2 road event Table
Figure BDA0003442514310000131
Step 7.4: changing the road passing time of the road section in the corresponding area into the value after influence;
step 7.5: judging whether the original planned path needs to be adjusted in the time segment, namely judging whether the changed road condition information can cause adverse effects on the current distribution tasks of the distribution vehicles: if the road passing time after the vehicle is about to go to the road section is more than N times of the original passing time, namely the change of the road condition information is considered to cause bad influence on the distribution task, the adjustment is needed;
step 7.6: if the adjustment is needed, re-planning is carried out according to an A-x algorithm for searching the shortest path between two points; no operation is performed without adjustment;
step 7.7: and repeating the steps according to the time segment of 0.5h until all distribution tasks are completed, namely all vehicles traverse all distribution points in the initial planning and return to the distribution center.

Claims (9)

1. A dynamic vehicle distribution path optimization method based on cloud edge-end cooperation is characterized by comprising the following steps:
s1, carrying out qualitative analysis on the distribution vehicle path planning problem, and determining distribution time and distribution cost as optimization targets of the first-stage path planning;
s2, considering the distribution capacity of a distribution center under the constraint of the cargo demand and the time window proposed according to demand points, and before a vehicle starts, optimizing the whole distribution path in advance at a cloud end by using a genetic algorithm according to an optimization target of first-stage path planning to obtain an initial distribution path planning scheme, wherein the distribution time spent on the whole initial distribution path planning scheme is the least, and the consumed distribution cost is the lowest;
s3, after the delivery vehicle starts, sensing and judging the road condition data, and updating the road network information of the corresponding road section in the initial delivery path planning scheme;
s4, dynamically adjusting the rest distribution paths according to the dynamically updated road network information and the positions of the distribution vehicles until all distribution tasks are finished, and calculating the time spent on distribution;
s5, optimizing and adjusting the route according to the road condition change, and updating the cost consumed by the current updated distribution route;
and S6, taking the distribution center as a starting point of the vehicles, taking the last distribution vehicle as a stop point of completing the distribution task, and dynamically adjusting the distribution vehicle route in the whole distribution area according to the steps S4 and S5 in the whole distribution process.
2. The dynamic vehicle distribution path optimization method based on cloud edge-side coordination according to claim 1, wherein in step S1, the distribution cost is the sum of vehicle transportation cost TC, time window penalty cost PC and vehicle cost; wherein,
Figure FDA0003442514300000011
Figure FDA0003442514300000012
Figure FDA0003442514300000013
wherein K is the total number of vehicles, N is the total number of demand points to be accessed, cijFor the unit transportation cost between demand point i and demand point j, dijIs the distance, x, between demand point i and demand point jijkThe value of (1) is 0 or 1, the value of 1 represents that the vehicle k goes to the demand point j after leaving from the demand point i, otherwise, the value of 0 is obtained; a, b are distribution time window penalty coefficients, wikFor the waiting time, t, of vehicle k at demand point iikThe time when the vehicle k reaches the delivery point i, liThe latest service time window of the demand point i; x is the number of0jkThe value is 0 or 1, and when the value is 1, the vehicle k is shown before starting from the distribution center 0To the demand point j.
3. The cloud-edge-cooperation-based dynamic vehicle distribution path optimization method according to claim 1, wherein the specific implementation process of the step S2 includes:
1) constructing a penalty function p (x):
Figure FDA0003442514300000021
wherein x represents the number of the corresponding population, T is a positive number, Du_maxRepresenting the maximum driving distance of the u vehicle type; n is the total number of demand points to be accessed;
2) decoding the penalty function p (x) to construct a chromosome corresponding to the penalty function p (x);
3) randomly generating an initial population according to spare part demand data, road network data and distribution resources;
4) and circularly evolving the initial population under the condition of not meeting the termination condition by virtue of selection, intersection and mutation operators until an optimal solution is generated to obtain an initial distribution path planning scheme.
4. The cloud-edge-cooperation-based dynamic vehicle distribution path optimization method according to claim 1, wherein the specific implementation process of the step S3 includes:
A) calculating the change of the passing time of the road section to which the edge node belongs according to the road condition change information acquired by each edge node, and updating the current passing time of each road section in the road passing time table according to the calculation result;
B) updating the current position of the delivery vehicle according to the vehicle speed, obtaining the road section which is required to pass through by the traveling route in the initial delivery path planning scheme between the two positions by combining the position of the next demand point to be reached, and completing the updating of the road network information of the corresponding road section in the initial delivery path planning scheme;
C) according to the current positions of all the distribution vehicles and the next distribution position to be moved to, a road traffic schedule is inquired, and whether the changed road condition information can cause bad influence on the current distribution tasks of all the distribution vehicles is judged: if the road passing time after the vehicle is about to go to the road section is greater than N times of the original passing time, it is determined that the change of the road condition information has a bad influence on the distribution task, and the step S4 is entered.
5. The cloud-edge-cooperation-based dynamic vehicle distribution path optimization method according to claim 1, wherein the specific implementation process of the step S4 includes:
I) taking the node with the minimum value of f (n) as the next node on the optimal path, wherein f (n) is g (n) + h (n), g (n) is the actual passing cost from the starting node to the current node, and h (n) is the estimated value of the passing cost from the current node to the destination;
II) operating the P table and the Q table maintained by the A-star algorithm, which specifically comprises the following steps:
i) setting the P table and the Q table to be null, adding the starting point S into the P table, setting the value of g (n) to be 0, setting the father node to be null, and setting the value of g (n) of other nodes in the road network to be infinity;
ii) if the P table is empty, the algorithm fails, otherwise, a node with the minimum f (n) value in the P table is selected and marked as BT, and the BT is added into the Q table; judging whether BT is an end point T, if yes, turning to step iii); otherwise, finding each adjacent node NT of the BT according to the topology attribute of the road network and the traffic rule, and executing the following steps:
calculating the heuristic value of NT
f(NT)＝g(NT)+h(NT)；
g(NT)＝g(BT)+cost(BT，NT)；
Wherein cost (BT, NT) is the traffic cost from BT to NT;
② if NT is in the P table and the passage cost value calculated by the formula g (NT) g (BT) + cost (BT, NT) is less than that of NT, then
Updating the transit cost value of NT to g (NT) ═ g (BT) + (BT, NT), and setting the parent node of NT as BT;
(iii) if NT is in Q table and the passage cost value calculated by g (NT) ═ g (BT) + cost (BT, NT) is smaller than that of NT, update the passage cost value of NT to g (NT) ═ g (BT) + cost (BT, NT), set the parent node of NT as BT, and move NT out to P table;
if the NT is not in the P table or the Q table, setting the father node of the NT as BT and moving the NT to the P table;
returning to the step ii);
iii) backtracking from the end point T, sequentially finding father nodes, and adding the father nodes into the optimized path until the starting point S to obtain the optimized path;
III) calculating the passing time required by the vehicle to pass through the optimized path, wherein the optimized path comprises a plurality of road sections, and the number of the road sections is 1, 2, 3.., k; by [ t ]k，tk′]Representing the transit time T of a vehicle through a section kkThen T isk＝tk′-tk(ii) a Passage time and T 'taken by vehicle to pass through multiple road sections'k1，T′k2，T′k3.., corresponding; f. ofkIndicating a time period corresponding to the moment when the vehicle passes the start of the section k,
Figure FDA0003442514300000041
calculating the transit time T of a vehicle through a section k using the following equationk：
Figure FDA0003442514300000042
The value of m satisfies the following constraints:
Figure FDA0003442514300000043
Δ T denotes the period length, T0Represents a start time;
IV) calculating the time taken to deal with the road condition change using the following equation: gn(r)＝bn(r)+cn(r)+kn(r)+hn(r)；
Figure FDA0003442514300000044
Figure FDA0003442514300000045
θn，n′Indicating data from vehicle VnTo the vehicle Vn′The number of vehicles passed;
Figure FDA0003442514300000046
Figure FDA0003442514300000047
Figure FDA0003442514300000048
m denotes the number of edge nodes, D denotes the set of edge computing devices, D ═ D1，d2，...，dmV denotes a vehicle set, V ═ V1，v2，...，vn}，λV2VDenotes a data transmission rate, λ, based on the V2V methodV2IIndicating the data transmission rate, ω, based on the V2I methodnRepresenting the amount of data of the computational task to be transferred, vnIndicating a delivery vehicle numbered n, dmEdge computing device denoted m,/nIndicating whether the computing task n is being processed on an edge computing device, lnThe value of 0 or 1, p represents the processing power of each edge computing device, unIs the amount of resources requested by the computing task n.
V) calculating the time required for completing the whole distribution task by using the following formula:
Figure FDA0003442514300000049
k represents the total number of road sections passed by the route with the number r _ s; r represents the processed road condition until the distribution is completedThe total number of road events that vary and adversely affect the road travel time.
6. The cloud-edge-cooperation-based dynamic vehicle distribution path optimization method according to claim 1, wherein the specific implementation process of the step S5 includes:
a) all the delivery vehicles start delivery according to the initial path planning;
b) dividing demand points needing to be distributed into a plurality of distribution areas according to r x r, numbering each distribution area, and taking r as unit length;
c) monitoring the change of road conditions by the edge nodes, recording events influencing the road conditions into event lists according to time lengths, wherein each event in the event lists has different influences on the passing time of roads leading to the demand points in each distribution area;
d) changing the road passing time of the road sections in the corresponding distribution area into the value after influence;
e) judging whether the original planned path needs to be adjusted in the current time slice; if the adjustment is needed, re-planning is carried out according to an A-x algorithm for searching the shortest path between two points;
f) and repeating the steps a) to f) for each other time segment until all the distribution tasks are completed, namely all the vehicles traverse all the demand points in the initial planning and return to the distribution center.
7. A computer apparatus comprising a memory, a processor and a computer program stored on the memory; characterized in that the processor executes the computer program to carry out the steps of the method according to one of claims 1 to 6.
8. A computer program product comprising a computer program/instructions; characterized in that the computer program/instructions, when executed by a processor, performs the steps of the method according to one of claims 1 to 6.
9. A computer readable storage medium having stored thereon a computer program/instructions; characterized in that the computer program/instructions, when executed by a processor, implement the steps of the method of one of claims 1 to 6.
CN202111640629.9A 2021-12-29 2021-12-29 Cloud edge end cooperation-based dynamic vehicle distribution path optimization method and device Active CN114399095B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111640629.9A CN114399095B (en) 2021-12-29 2021-12-29 Cloud edge end cooperation-based dynamic vehicle distribution path optimization method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111640629.9A CN114399095B (en) 2021-12-29 2021-12-29 Cloud edge end cooperation-based dynamic vehicle distribution path optimization method and device

Publications (2)

Publication Number Publication Date
CN114399095A true CN114399095A (en) 2022-04-26
CN114399095B CN114399095B (en) 2024-07-26

Family

ID=81229660

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111640629.9A Active CN114399095B (en) 2021-12-29 2021-12-29 Cloud edge end cooperation-based dynamic vehicle distribution path optimization method and device

Country Status (1)

Country Link
CN (1) CN114399095B (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114565195A (en) * 2022-04-27 2022-05-31 苏州美集供应链管理股份有限公司 Local feedback and cloud global optimization combined allocation system and method
CN115578879A (en) * 2022-09-14 2023-01-06 江苏天一航空工业股份有限公司 Vehicle optimal path navigation control method and system based on Internet of things
CN116307332A (en) * 2023-05-16 2023-06-23 武汉大学人民医院（湖北省人民医院） Real-time medicine distribution method and device
CN116739467A (en) * 2023-06-16 2023-09-12 暨南大学 Circulating delivery path planning method, device and medium based on cloud edge cooperation
CN117010670A (en) * 2023-10-07 2023-11-07 普迪智能装备有限公司 Intelligent logistics distribution system and method
CN117894201A (en) * 2024-03-11 2024-04-16 湘江实验室 Road traffic guiding method and device based on global traffic information and related equipment

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130054050A1 (en) * 2011-08-24 2013-02-28 Dimitar Petrov Filev Methods and apparatus for a vehicle to cloud to vehicle control system
CN111191813A (en) * 2019-10-14 2020-05-22 重庆邮电大学 Vehicle distribution path optimization method based on cargo load and soft time window limitation
CN111882099A (en) * 2020-05-11 2020-11-03 武汉理工大学 Logistics distribution path planning method based on variable neighborhood parallel annealing algorithm
CN113469431A (en) * 2021-06-29 2021-10-01 深圳市点购电子商务控股股份有限公司 Method, device and computer readable storage medium for recommending logistics distribution path

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130054050A1 (en) * 2011-08-24 2013-02-28 Dimitar Petrov Filev Methods and apparatus for a vehicle to cloud to vehicle control system
CN111191813A (en) * 2019-10-14 2020-05-22 重庆邮电大学 Vehicle distribution path optimization method based on cargo load and soft time window limitation
CN111882099A (en) * 2020-05-11 2020-11-03 武汉理工大学 Logistics distribution path planning method based on variable neighborhood parallel annealing algorithm
CN113469431A (en) * 2021-06-29 2021-10-01 深圳市点购电子商务控股股份有限公司 Method, device and computer readable storage medium for recommending logistics distribution path

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114565195A (en) * 2022-04-27 2022-05-31 苏州美集供应链管理股份有限公司 Local feedback and cloud global optimization combined allocation system and method
CN114565195B (en) * 2022-04-27 2022-08-19 苏州美集供应链管理股份有限公司 Local feedback and cloud global optimization combined allocation system and method
CN115578879A (en) * 2022-09-14 2023-01-06 江苏天一航空工业股份有限公司 Vehicle optimal path navigation control method and system based on Internet of things
CN115578879B (en) * 2022-09-14 2023-12-08 江苏天一航空工业股份有限公司 Vehicle optimal path navigation control method and system based on Internet of things
CN116307332A (en) * 2023-05-16 2023-06-23 武汉大学人民医院（湖北省人民医院） Real-time medicine distribution method and device
CN116307332B (en) * 2023-05-16 2023-08-04 武汉大学人民医院（湖北省人民医院） Real-time medicine distribution method and device
CN116739467A (en) * 2023-06-16 2023-09-12 暨南大学 Circulating delivery path planning method, device and medium based on cloud edge cooperation
CN116739467B (en) * 2023-06-16 2024-02-02 暨南大学 Circulating delivery path planning method, device and medium based on cloud edge cooperation
CN117010670A (en) * 2023-10-07 2023-11-07 普迪智能装备有限公司 Intelligent logistics distribution system and method
CN117010670B (en) * 2023-10-07 2024-06-07 普迪智能装备有限公司 Intelligent logistics distribution system and method
CN117894201A (en) * 2024-03-11 2024-04-16 湘江实验室 Road traffic guiding method and device based on global traffic information and related equipment

Also Published As

Publication number Publication date
CN114399095B (en) 2024-07-26

Similar Documents

Publication Publication Date Title
CN114399095A (en) Cloud-side-cooperation-based dynamic vehicle distribution path optimization method and device
CN110119839B (en) Urban road traffic accident emergency rescue path planning method
Qin et al. Reinforcement learning for ridesharing: An extended survey
CN112418497B (en) Material distribution path optimization method for manufacturing Internet of things
Hu et al. A two-stage dynamic capacity planning approach for agricultural machinery maintenance service with demand uncertainty
He et al. Vehicle scheduling under stochastic trip times: An approximate dynamic programming approach
Li et al. Learning to optimize industry-scale dynamic pickup and delivery problems
CN112884241A (en) Cloud edge collaborative manufacturing task scheduling method based on intelligent Agent
Chen et al. PPtaxi: Non-stop package delivery via multi-hop ridesharing
CN113359702B (en) Intelligent warehouse AGV operation optimization scheduling method based on water wave optimization-tabu search
CN101789096A (en) Material flow simulation optimization method in steel making-continuous casting production based on target guiding
Zhou et al. A scalable vehicle assignment and routing strategy for real-time on-demand ridesharing considering endogenous congestion
CN106447266A (en) Intelligent optimized scheduling method for multiple vehicle models and multi-time dispensing in logistics field
CN115511213A (en) Path optimization method, device, medium and system for transporting hydroelectric equipment
CN104537446A (en) Bilevel vehicle routing optimization method with fuzzy random time window
Xie et al. A shared parking optimization framework based on dynamic resource allocation and path planning
Hu et al. Optimal design for urban mass transit network based on evolutionary algorithms
CN116681197B (en) Logistics line searching method and system based on big data
Wang et al. Good or mediocre? A deep reinforcement learning approach for taxi revenue efficiency optimization
Pillac et al. A fast re-optimization approach for dynamic vehicle routing
Zhang et al. Dynamic joint decision of matching parameters and relocation strategies in ride-sourcing systems interacting with traffic congestion
Li et al. Dynamic repositioning in dock-less bike-sharing system: A multi-agent reinforcement learning approach
Nahum et al. A framework for solving real-time multi-objective VRP
CN115099569A (en) Passenger and taxi matching method based on fuzzy logic controller
Li et al. Handling dynamic capacitated vehicle routing problems based on adaptive genetic algorithm with elastic strategy

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