CN116127857A - Classification-oriented household garbage collection and transportation path multi-objective optimization method and system - Google Patents
Classification-oriented household garbage collection and transportation path multi-objective optimization method and system Download PDFInfo
- Publication number
- CN116127857A CN116127857A CN202310387128.7A CN202310387128A CN116127857A CN 116127857 A CN116127857 A CN 116127857A CN 202310387128 A CN202310387128 A CN 202310387128A CN 116127857 A CN116127857 A CN 116127857A
- Authority
- CN
- China
- Prior art keywords
- garbage
- transportation
- collection
- vehicle
- optimization
- 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
Links
- 239000010813 municipal solid waste Substances 0.000 title claims abstract description 410
- 238000005457 optimization Methods 0.000 title claims abstract description 169
- 238000000034 method Methods 0.000 title claims abstract description 90
- 230000002068 genetic effect Effects 0.000 claims abstract description 50
- 239000002245 particle Substances 0.000 claims abstract description 45
- 238000012546 transfer Methods 0.000 claims description 104
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims description 67
- 229910052799 carbon Inorganic materials 0.000 claims description 67
- 230000008569 process Effects 0.000 claims description 33
- 239000010806 kitchen waste Substances 0.000 claims description 25
- 239000002699 waste material Substances 0.000 claims description 25
- 239000000446 fuel Substances 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 19
- 239000011159 matrix material Substances 0.000 claims description 17
- 238000004519 manufacturing process Methods 0.000 claims description 8
- 239000010791 domestic waste Substances 0.000 claims description 6
- 230000014759 maintenance of location Effects 0.000 claims description 6
- 238000002485 combustion reaction Methods 0.000 claims description 4
- 238000011160 research Methods 0.000 abstract description 3
- 238000007726 management method Methods 0.000 description 11
- 241000196324 Embryophyta Species 0.000 description 9
- 230000035772 mutation Effects 0.000 description 9
- 230000007613 environmental effect Effects 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 6
- 230000002195 synergetic effect Effects 0.000 description 6
- 238000011161 development Methods 0.000 description 4
- 208000011597 CGF1 Diseases 0.000 description 3
- 238000013459 approach Methods 0.000 description 3
- 238000012804 iterative process Methods 0.000 description 3
- 238000003012 network analysis Methods 0.000 description 3
- 230000004039 social cognition Effects 0.000 description 3
- 238000010835 comparative analysis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003912 environmental pollution Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 239000005431 greenhouse gas Substances 0.000 description 2
- 230000009191 jumping Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000013439 planning Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000010845 search algorithm Methods 0.000 description 2
- 230000008685 targeting Effects 0.000 description 2
- 238000012384 transportation and delivery Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000002922 simulated annealing Methods 0.000 description 1
- 239000002910 solid waste Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- G06N3/12—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
- G06Q10/04—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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Strategic Management (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Marketing (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- General Business, Economics & Management (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Game Theory and Decision Science (AREA)
- Genetics & Genomics (AREA)
- Entrepreneurship & Innovation (AREA)
- Physiology (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Educational Administration (AREA)
Abstract
The invention provides a classification-oriented household garbage collection and transportation path multi-objective optimization method and system, and belongs to the technical field of urban garbage collection and transportation. According to the method, household garbage collection and transportation path optimization models of different collection and transportation modes are respectively established, a mixed particle swarm genetic algorithm is adopted to solve a multi-objective pareto optimal solution set, and multi-objective optimization is carried out at the same time; quantitatively analyzing the coordination or trade-off relation among different optimization targets, setting the decision preference of the path scheme, setting the weight of each target according to the set decision preference, determining the optimization scheme under different preferences, and determining the optimal household garbage collection and transportation path. The invention provides researches on classified combined collection and transportation modes and classified independent collection and transportation modes, realizes the selection of different garbage classified collection and transportation modes according to different preferences and the determination of the optimal collection and transportation path, and can optimize the target more.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a classification-oriented household garbage collection and transportation path multi-objective optimization method and system.
Background
Domestic garbage management is an important issue of worldwide urban management concern, and developing an efficient, economical and environment-friendly operation mode for a domestic garbage management system is a challenge faced by most cities. Garbage collection and transportation are important links in the process of managing household garbage, the process cost accounts for more than 50% of garbage management budget, and negative environmental effects of different degrees can be generated. Optimizing garbage collection and transportation path, saving economic cost, reducing emission of greenhouse gas and pollutant, and promoting sustainable development of city.
Garbage collection route optimization is based on vehicle path problem (VRP) or trip problem (TSP) studies. The essence is to solve the shortest path that the vehicle starts from the starting point, passes through a plurality of garbage generation nodes and returns to the starting point. The simple VRP only solves the shortest distance, and does not consider the vehicle capacity, the garbage generation amount of the garbage generation node and the like. In the vehicle path problem (CVRP) with vehicle capacity constraints, the truck is not completely free-running and needs to be returned to the disposal facility to empty the waste when the capacity limit is reached. The vehicle path problem with time window (VRPTW) is another common extension to VRPs where vehicles need to reach the garbage generating node and complete loading and unloading within a specified time interval.
Algorithms for solving the problem of solid waste collection and transportation can be divided into traditional algorithms, heuristic algorithms and meta-heuristic algorithms. Traditional algorithms include branch-and-bound, linear programming, mixed integer programming, quadratic programming models, and the like. The conventional method has a limitation in that since most path optimization problems are NP (non-polynominal) difficult problems, a large amount of computation time is required in solving a large-scale optimization problem. In order to overcome the limitations of the conventional algorithm, heuristic algorithms and meta-heuristic algorithms have been widely used in the VRP solution process. Khalid et al solve the transport vehicle optimization path based on the traveling salesman problem by adopting a tabu search algorithm (TS) and a simulated annealing algorithm (SA), and compare the solving effects of the two algorithms; seyd et al propose an optimization model targeting cost minimization with the total number of vehicles, formal routes, travel distance, garbage collection as the main decision variables, and solve by employing genetic algorithm; laura et al propose an iterative greedy algorithm incorporating neighborhood searching to solve a shipping optimization model that includes multiple objectives such as cost of travel, length of route, number of routes, etc. In addition, improved Social Engineering Optimizers (SEOs), hybrid heuristics based on the Clarke & Wright algorithm and adaptive neighborhood search algorithm, and the like are also applied to the shipping problem solving.
By integrating the advantages of different heuristic algorithms, a new hybrid heuristic algorithm is provided on the basis of improving the traditional heuristic algorithm, and the method is a mainstream trend for solving the household garbage collection and transportation path optimization problem.
The shipping path optimization typically involves one or more objectives of path length, shipping cost, environmental pollution, workload variance, etc., the number and type of optimization objectives being varied by research issues. Delgado et al propose an optimization model targeting total distance minimization, longest path minimization, path length minimum, path number minimization; the waste collection and transportation optimization model of Lu, pu and Han is optimized by minimizing economic cost and maximizing workload difference; mojtahedi et al minimized fleet size, minimized transportation costs, minimized carbon emissions, maximum workload variance as optimization targets, and elucidated the link between path optimization targets and sustainable development targets. In a multi-objective optimization model, the result of the optimization is a pareto optimal solution set based on the pareto theory, from which it may be difficult for the decision maker to select the final solution. Therefore, the cooperative or trade-off relation between different targets needs to be defined, and a proper decision method is designed to select the optimal scheme. In recent studies, multi-attribute decision methods such as Analytic Hierarchy Process (AHP), multi-objective gray target decision, etc. have been applied to decision processes of multi-objective path optimization final schemes. However, multi-objective path optimization studies that propose the final garbage collection path selection method are still few and lack a comprehensive analysis of the synergistic or trade-off relationships between different path optimization objectives.
The garbage classification treatment can effectively improve the recovery rate of garbage and reduce the landfill amount. Therefore, garbage source classification becomes an important measure widely popularized to promote sustainable development. Meanwhile, with the increasing attention of people on three sustainable posts, a receiver needs to consider not only economic targets but also environmental and social targets when carrying out the receiving and transporting planning. In order to address the above challenges of garbage collection, it is important to study how to implement efficient garbage classification collection, including path optimization under the determination of collection mode and selection of the best collection mode from alternatives. Lu, pu and Han propose an optimized model for the classified joint collection of recoverable, non-recoverable waste and solution using a hybrid whale-genetic algorithm. However, existing shipping path optimization studies still have less focus on the problem of classified shipping, especially lacking comparative analysis of different classified shipping approaches.
The prior art has at least the following disadvantages:
the existing method is less in concern of classified collection and transportation problems, and particularly lacks of comparison analysis on different classified collection and transportation modes when multiple targets are realized, so that the garbage collection and transportation path optimization method is not strong in adaptability and cannot meet the multiple target optimization requirements under garbage classification popularization; the existing method lacks concern on the cooperative or trade-off relation among economic, environmental and social targets and the application of a multi-attribute decision-making method, which forms an obstacle for a decision maker to determine a final scheme from the pareto solution set; 3. in the household garbage management practice, garbage types and optimization targets are more, and the existing meta-heuristic algorithm needs to be improved and perfected so as to solve the increasingly complex garbage collection and transportation optimization model.
Disclosure of Invention
The method aims at solving the problem that the garbage collection and transportation multi-target path optimization result is not ideal due to the lack of comparison and analysis of different classification and transportation modes in the prior art. The invention provides a classified household garbage collection and transportation path multi-objective optimization method and system, which respectively establish household garbage collection and transportation path optimization models of different collection and transportation modes, and determine an optimal household garbage collection and transportation path by simultaneously optimizing the following four objectives: economic cost, carbon emissions, vehicle queuing time, and vehicle workload variance; the economic cost comprises driving cost, vehicle depreciation cost, personnel cost and garbage manual classification cost of a garbage transfer station; minimizing economic cost, carbon emissions, vehicle queuing time, and workload variance as optimization objectives; the different collection and transportation modes comprise a classification combined collection and transportation mode and a classification independent collection and transportation mode, wherein the classification combined collection and transportation mode is that a vehicle with a compartment starts from a garbage transfer station and goes to each garbage generation point, different types of garbage are loaded into corresponding compartments, and when any compartment reaches capacity constraint, the garbage is returned to the garbage transfer station; the kitchen garbage treatment plant dispatches vehicles to a garbage transfer station to transport kitchen garbage to the kitchen garbage treatment plant; the separate classification and collection mode is that kitchen waste and other waste are respectively transported by different vehicles, and the kitchen waste collection process does not pass through a waste transfer station; adopting a mixed particle swarm genetic algorithm (HPSO-GA algorithm), solving a multi-objective pareto optimal solution set, and carrying out multi-objective optimization; the mixed particle swarm genetic algorithm is based on a genetic algorithm, and meanwhile, the particle swarm optimization thought is considered; quantitatively analyzing the coordination or trade-off relation among different optimization targets, merging targets with the quantized coordination or trade-off relation exceeding a preset threshold, setting decision preference of a path scheme, setting weights of the targets according to the set decision preference, and determining optimal household garbage collection and transportation paths of different collection and transportation modes under different preferences;
Identifying the optimal household garbage collection and transportation mode and the corresponding optimal household garbage collection and transportation path under each decision preference by comparing the target values of the optimal household garbage collection and transportation paths under different collection and transportation modes; the household garbage collection and transportation path optimization model comprises links from a garbage transfer station to a garbage generation node, from the garbage transfer station to a kitchen garbage treatment plant and from the kitchen garbage treatment plant to the garbage generation node. The invention provides researches on classified combined collection and transportation modes and classified independent collection and transportation modes, realizes the selection of different garbage collection and transportation modes according to different preferences and the determination of the optimal collection and transportation path, and can optimize the target more.
The invention provides a classification-oriented household garbage collection and transportation path multi-objective optimization method, which comprises the following steps:
respectively establishing household garbage collection and transportation path optimization models of different collection and transportation modes, and determining an optimal household garbage collection and transportation path by simultaneously optimizing the following four targets: economic cost, carbon emissions, vehicle queuing time, and vehicle workload variance; minimizing economic cost, carbon emissions, vehicle queuing time, and workload variance as optimization objectives; the different collection and transportation modes comprise a classification combined collection and transportation mode and a classification independent collection and transportation mode, wherein the classification combined collection and transportation mode is that a vehicle with a compartment starts from a garbage transfer station and goes to each garbage generation point, different types of garbage are loaded into corresponding compartments, and when any compartment reaches capacity constraint, the garbage is returned to the garbage transfer station; the kitchen garbage treatment plant dispatches vehicles to a garbage transfer station to transport kitchen garbage to the kitchen garbage treatment plant; the separate classification and collection mode is that kitchen waste and other waste are respectively transported by different vehicles, and the kitchen waste collection process does not pass through a waste transfer station;
Adopting a mixed particle swarm genetic algorithm (HPSO-GA algorithm), solving a multi-objective pareto optimal solution set, and carrying out multi-objective optimization; the mixed particle swarm genetic algorithm is based on a genetic algorithm, and meanwhile, the particle swarm optimization thought is considered; adopting a sequence preference technology TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) method based on ideal solution similarity to quantitatively analyze the coordination or trade-off relation between different optimization targets, merging targets with the quantized coordination or trade-off relation exceeding a preset threshold, setting decision preference of a path scheme, setting weights of the targets according to the set decision preference, and determining optimal household garbage collection and transportation paths of different collection and transportation modes under different preferences;
identifying the optimal household garbage collection and transportation mode and the corresponding optimal household garbage collection and transportation path under each decision preference by comparing the target values of the optimal household garbage collection and transportation paths under different collection and transportation modes;
the household garbage collection and transportation path optimization model comprises links from a garbage transfer station to a garbage generation node, from the garbage transfer station to a kitchen garbage treatment plant and from the kitchen garbage treatment plant to the garbage generation node.
The mixed particle swarm genetic algorithm adopts a particle swarm optimization algorithm, and the optimization is carried out in the variation direction in the iterative process. Particle swarm optimization simulates the process of preying on birds in nature, and in order to find food (to achieve an optimization goal), the motion direction of each individual in the population depends on the motion direction of the individual and the motion direction of the population as a whole. The higher the fitness of the individual is, the easier the original movement direction of the individual is maintained, otherwise, the easier the individual is to follow the overall movement direction of the population.
Preferably, the economic cost includes driving cost, vehicle depreciation cost, personnel cost and garbage manual classification cost of the garbage transfer station;
preferably, the hybrid particle swarm genetic algorithm is specifically based on a genetic algorithm, while combining elite retention strategies and particle swarm optimization ideas. And the elite retention strategy directly puts each generation of better individuals into the next generation, so that the elite individuals are prevented from being eliminated or mutated, and the iteration efficiency can be improved.
Preferably, particle swarm optimization is adopted in the genetic iteration process, and the self-cognition factors are calculated according to the ranking grade
Wherein n is the maximum ranking, and the social cognition factor is +.>
The individuals in the dominated solution are subjected to self-variation by probability r, the bit sequences of two digits in the codes are randomly exchanged, and the generated new individuals enter the next generation population; and (3) cross mutation occurs according to the probability of 1-r, one section of codes is randomly selected, and replaced by elite individual codes, and the rest bit sequences are kept the same as the original codes under the condition that the numbers are not repeated.
Preferably, in the multi-target pareto optimal solution set solved by using a mixed particle swarm genetic algorithm, a good-bad solution distance method is adopted to solve the multi-target pareto optimal solution set.
Preferably, the carbon emission amount is calculated based on the vehicle workload and a carbon emission factor per unit vehicle workload.
Preferably, the vehicle workload is the product of distance and vehicle load capacity.
Preferably, the vehicle workload difference is represented by a coefficient of variation of the vehicle workload.
Preferably, the economic cost minimization is represented by the following objective function:
wherein ,
vencoding a vehicle;
i,jcoding nodes (including garbage generation nodes, transfer stations and kitchen garbage treatment plants);
ECis an economic cost;
N D generating a node set for garbage;
N T the garbage transfer stations are collected;
N K collecting kitchen waste treatment plants;
F v a fixed depreciation cost per day for vehicle v;
H v the human cost per day of vehicle v;
V 1 a vehicle set from a garbage generation point to a garbage transfer station;
V 2 collecting vehicles from a garbage generation point to a kitchen garbage treatment plant;
V 3 the method comprises the steps of collecting vehicles from a garbage transfer station to a kitchen garbage treatment plant;
x ijv decision variables for the vehicle v going from the refuse generating node i to the refuse generating node j;
d ij the distance between garbage generating nodes i and j;
Preferably, the carbon emissions include not only carbon emissions generated during the vehicle operation phase, but also carbon emissions generated during the fuel production and transportation phase, the minimization of which is expressed by the following objective function:
wherein ,
CEis carbon emission;
Preferably, the vehicle queuing time minimization is expressed by the following objective function:
wherein ,
VQ is the total queuing time of the vehicle;
qt iv the sum of queuing time of the vehicle v in the garbage transfer station and the kitchen garbage disposal plant is obtained.
Preferably, the vehicle workload difference minimization is represented by the following objective function:
wherein ,
Preferably, constraints of the optimization model are as follows:
Each trash producing node is accessed once and only once by the same vehicle:
vehicles entering the debris generating node must leave the debris generating node:
in the classified and independent collection and transportation mode, after a single car reaches a garbage generation node, the garbage type (kitchen garbage/other garbage) responsible for collection and transportation needs to be emptied:
when the vehicle v starts from a garbage transfer station or a kitchen garbage station, the initial load capacity is 0:
the vehicle capacity constraints are:
vehicle V 1 Starting from the waste transfer station and finally needing to return to the waste transfer station:
the time when the vehicle arrives at the kitchen garbage disposal plant and the garbage transfer station is required to be within the range of a time window which allows the kitchen garbage disposal plant and the garbage transfer station to operate:
the queuing time of the vehicles in the garbage transfer station is as follows, wherein the number v is the arrangement of the sequence of the vehicles arriving at the garbage transfer station:
in the classified intermodal mode, V 3 The vehicle needs to transport all kitchen waste of the waste transfer station to a kitchen waste treatment plant:
vehicle V 3 After being transported to a garbage transfer station from a kitchen garbage disposal plant, the garbage is required to return to the kitchen garbage disposal plant:
vehicle V 3 Is defined by the capacity constraints of:
preferably, the algorithmic contrast analysis is performed by error rate, supersvolume, and computational efficiency.
Error rate is defined as the ratio of the number of dominant individuals in a population to the total number of individuals. It reflects the goodness of the model output result. The higher the error rate, the more suboptimal number of individuals in the population; conversely, the lower the error rate, the greater the number of pareto individuals in the population, indicating that the pareto solution set has diversity, and at the same time, indicating that the algorithm has higher reliability.
The supersolume index measures the volume of the target space governed by the pareto solution set of the algorithm with a set of preset reference points r distributed in the target space as boundaries. The convergence and diversity of the solution set can be comprehensively measured by the hyper-volume index (HV), and the larger the HV value is, the closer the pareto solution set obtained by the algorithm is to the real pareto front, and the solution set has better convergence and diversity.
The computational efficiency is specifically the algorithm run time, which is the run time required by the algorithm to achieve a particular quality solution set or to complete a particular number of iterations. The shorter the run time, the higher the solution efficiency of the algorithm. It is an important index for measuring the effectiveness of the algorithm.
Preferably, in a multi-objective pareto optimal solution set obtained by solving by using a mixed particle swarm genetic algorithm, a good-bad solution distance method is adopted, the score of each scheme in the pareto solution set is calculated according to the preference weight, and the scheme with the highest score is selected as a final path optimization scheme.
Preferably, the score of each scheme in the pareto solution set is calculated according to the preference weight, and the scheme with the highest score is selected as the final path optimization scheme calculation step:
carrying out standardized processing on the data, eliminating the data dimension, and obtaining a standardized matrix:
in the formula ,
is the schemeiTarget objectjTarget value of->
Is a standardized schemeiTarget objectjIs set to be a target value of (c),NMis a standardized matrix.
Multiplying the normalized target value by the target weight according to the decision preference to obtain a weighted normalized matrix:
in the formula ,
normalized scheme for weightingiTarget objectjIs set to be a target value of (c),WNMfor weighting the normalization matrix +.>
Is the object ofjWeights of (when the target isjWhen the target is a preference target, the value is 0.7; when the object isjFor a non-preference target, the value is 0.1).
Calculating the distance between the solution and the optimal target value and the worst target value:
wherein ,
、/>
respectively as targetsjOptimal and worst values of +.>
、/>
Respectively is the schemeiDistance from the optimal value and the worst value.
Calculating a score for the solution:
wherein ,
is the schemeiThe scheme with the highest score is selected as the final path optimization scheme.
Preferably, the synergy or trade-off relationship between different optimization objectives is quantitatively analyzed by calculating the spearman correlation coefficients of the different optimization objectives in the pareto solution set.
Preferably, the calculation formula of the spearman correlation coefficient is:
in the formula ,ris the correlation coefficient between the two objects,
is the rank after the first target rank, +. >
Is the rank after the second target ordering,nis the number of individuals in the pareto solution set.
Preferably, the different preferences include: "economic cost preference", "environmental benefit preference", "time optimization preference", "vehicle management preference", "balanced decision preference" and collaborative preference: when the correlation coefficient of the two targets is more than or equal to 0.8, the two targets are considered to have a strong synergistic relationship, the two target preferences can be combined, the weights of the two targets are respectively set to 0.4, and the weights of other targets are set to 0.1
Preferably, the decision preference of setting the path scheme according to the synergistic or trade-off relationship between different optimization objectives specifically includes: in the pareto optimal solution set, according to correlation coefficient calculation results of different optimization targets, determining a synergistic or trade-off relation of the two optimization targets, wherein the correlation is greater than a threshold value, and the two optimization targets are combined and set to be the same preference; for example, when the correlation coefficient of two targets is equal to or greater than 0.8 (the threshold is 0.8, and of course, other values can be used), the two targets are considered to have a strong synergistic or trade-off relationship, and when the optimization preference is set, the two targets are combined into one preference, such as: and if the correlation coefficient of the economic cost target and the carbon emission target is more than or equal to 0.8, combining the economic cost and the environmental benefit into the economic-environmental preference.
Preferably, the code of the vehicle path takes a positive real number as the number of the garbage generation nodes, and the sequence of the garbage generation nodes with the length being the same as the number of the garbage generation nodes represents the running sequence of the vehicle among the nodes to be received; in view of vehicle capacity constraints, the decoding process of the vehicle path is divided into three steps:
dividing a garbage generation node sequence according to the garbage generation amount and the vehicle capacity of each garbage generation node, and distributing the garbage generation node sequence to different vehicles;
representing a garbage transfer station by a value of 0, inserting the garbage transfer station into an original garbage generation node sequence, and acquiring the actual running sequence of a vehicle between the garbage transfer station and a node to be received;
calculating the economic cost, the vehicle queuing time, the vehicle workload difference and the value of each objective function of the carbon emission according to the distance matrix between the nodes to be received and the vehicle parameters, the transportation cost parameters and the carbon emission factors, wherein the vehicle parameters comprise the vehicle capacity, the average running speed, the carbon emission of the unit workload, the time required for loading garbage and the time required for unloading garbage;
the transportation cost parameters include a daily fixed depreciation cost of the vehicle, a daily labor cost of the vehicle, and a unit distance transportation cost.
Preferably, a particle swarm optimization idea is adopted in the genetic iteration process, a self-cognition factor r= (n-rank)/n is calculated according to a ranking level, wherein n is the maximum ranking level, a social cognition factor is 1-r, individuals in a dominated solution set are subjected to self-mutation according to probability r, bit sequences of two digits in a code are randomly exchanged, and a new individual generated enters a next generation population; and (3) cross mutation occurs according to the probability of 1-r, one section of codes is randomly selected, and replaced by elite individual codes, and the rest bit sequences are kept the same as the original codes under the condition that the numbers are not repeated.
On the one hand, the iterative process based on the particle swarm optimization idea ensures the randomness of variation, so that the algorithm has the capability of jumping out of a local optimal solution; on the other hand, the mutation directivity is ensured, the mutation of the dominant individual to the non-dominant individual approaches, and the probability of generating better individuals in the mutation process is improved.
Preferably, the optimization process comprises the steps of:
collecting the coordinate position of the garbage generation node, the garbage generation amount of the garbage generation node and the road network vector data, calculating a road network distance matrix between the nodes by using an ArcGIS network analysis tool, and taking the garbage generation amount and the distance matrix as input data of an optimization process;
respectively establishing a classification intermodal transportation mode and a household garbage collection and transportation path optimization model of a classification independent collection and transportation mode;
for each garbage collection mode, a mixed particle swarm genetic algorithm (HPSO-GA algorithm) and a rapid non-dominant sorting genetic algorithm (NSGA-II algorithm) are respectively used for solving a multi-objective pareto optimal solution set so as to carry out comparison analysis on algorithm performances;
verifying the effectiveness of an optimization algorithm through error rate indexes, super-volume indexes and calculation efficiency evaluation algorithm performance;
calculating spearman correlation coefficients of different optimization targets in the pareto solution set, analyzing synergy or trade-off relation among the different targets, setting decision preference of a path scheme, and determining specific optimization schemes under different preference by a TOPSIS method;
And determining optimal household garbage collection and transportation paths in different collection and transportation modes according to the optimization schemes in different preferences, and selecting a final collection and transportation mode and the optimal household garbage collection and transportation path corresponding to the final collection and transportation mode according to the optimal household garbage collection and transportation paths in different collection and transportation modes.
The invention provides a classified household garbage collection and transportation path multi-target optimization system, which comprises a path optimization module, wherein the path optimization module performs household garbage collection and transportation path multi-target optimization by using any classified household garbage collection and transportation path multi-target optimization method, and determines an optimal household garbage collection and transportation path.
Compared with the prior art, the invention has the following beneficial effects:
(1) In the urban household garbage classified transportation route optimization model provided by the invention, the economic cost, carbon emission, vehicle queuing time and vehicle workload difference targets under two garbage classified collection and transportation modes are comprehensively considered, and compared with the existing route optimization technology taking the shortest route as a main target, the urban household garbage classified transportation route optimization model can better adapt to the requirements of garbage transportation management on economic and environmental benefits, and reduce economic cost, environmental pollution and management burden.
(2) In the genetic iteration process, the elite retention strategy is adopted, after the non-dominant sorting of the original population is completed, individuals in the non-dominant solution set are retained, and the individuals directly enter the next generation population, so that the non-dominant individuals can be prevented from generating negative variation, the non-dominant solutions are prevented from being lost in the genetic process, the algorithm iteration efficiency is improved, the optimal solution set with higher convergence and diversity is obtained in a shorter time, and the better target can be obtained more quickly.
(3) In the genetic iteration process, the particle swarm optimization idea is adopted, so that the variation direction of an individual in a genetic algorithm can be controlled, the individual can keep certain development inertia, and the individual can be randomly varied on the basis of self-coding; the optimization variation trend can be kept to be close to non-dominant elite individuals, the algorithm iteration efficiency can be improved, an optimization solution set with higher convergence and diversity can be obtained in a shorter time, and a better target can be obtained more quickly.
(4) According to the invention, the relationship between the targets is analyzed and optimized by calculating the spearman correlation coefficient, the dimensions of the multiple targets are reduced according to the target synergy or trade-off relationship, the decision preference is set, and finally, the optimal optimization scheme under different decision preferences is determined by the TOPSIS method, so that the optimal garbage collection and transportation path can be obtained.
Drawings
FIG. 1 is a flow chart of a classification-oriented domestic waste collection path multi-objective optimization method according to one embodiment of the invention;
FIG. 2 is a schematic diagram of a vehicle path codec process according to one embodiment of the present invention;
FIG. 3 a is an algorithmic representation of HPSO-GA in a categorical intermodal mode of an embodiment of the invention; b is an algorithmic representation of the replacement of HPSO-GA with NSGA II (fast non-dominant ranking genetic algorithm) in the categorical intermodal model of one embodiment of the invention;
FIG. 4 a is an algorithmic representation of HPSO-GA in a sort by shipping mode in accordance with one embodiment of the present invention; b is an algorithmic representation of the replacement of HPSO-GA with NSGA II (fast non-dominant ordering genetic algorithm) in a classified single shipping mode of an embodiment of the invention;
FIG. 5 is a schematic diagram of a technical framework of a process of solving a final path according to one embodiment of the invention.
Detailed Description
The following describes in detail the embodiments of the present invention with reference to fig. 1 to 5.
The invention provides a classification-oriented household garbage collection and transportation path multi-objective optimization method, which comprises the following steps:
respectively establishing household garbage collection and transportation path optimization models of different collection and transportation modes, and determining an optimal household garbage collection and transportation path by simultaneously optimizing the following four targets: economic cost, carbon emissions, vehicle queuing time, and vehicle workload variance; minimizing economic cost, carbon emissions, vehicle queuing time, and workload variance as optimization objectives; the different collection and transportation modes comprise a classification combined collection and transportation mode and a classification independent collection and transportation mode, wherein the classification combined collection and transportation mode is that a vehicle with a compartment starts from a garbage transfer station and goes to each garbage generation point, different types of garbage are loaded into corresponding compartments, and when any compartment reaches capacity constraint, the garbage is returned to the garbage transfer station; the kitchen garbage treatment plant dispatches vehicles to a garbage transfer station to transport kitchen garbage to the kitchen garbage treatment plant; the separate classification and collection mode is that kitchen waste and other waste are respectively transported by different vehicles, and the kitchen waste collection process does not pass through a waste transfer station;
Adopting a mixed particle swarm genetic algorithm (HPSO-GA algorithm), solving a multi-objective pareto optimal solution set, and carrying out multi-objective optimization; the mixed particle swarm genetic algorithm is based on a genetic algorithm, and meanwhile, the particle swarm optimization thought is considered;
quantitatively analyzing the coordination or trade-off relation among different optimization targets, merging targets with the quantized coordination or trade-off relation exceeding a preset threshold, setting decision preference of a path scheme, setting weights of the targets according to the set decision preference, and determining optimal household garbage collection and transportation paths of different collection and transportation modes under different preferences;
identifying the optimal household garbage collection and transportation mode and the corresponding optimal household garbage collection and transportation path under each decision preference by comparing the target values of the optimal household garbage collection and transportation paths under different collection and transportation modes;
the household garbage collection and transportation path optimization model comprises links from a garbage transfer station to a garbage generation node, from the garbage transfer station to a kitchen garbage treatment plant and from the kitchen garbage treatment plant to the garbage generation node.
According to a specific embodiment of the invention, the economic cost comprises driving cost, vehicle depreciation cost, personnel cost and garbage manual classification cost of the garbage transfer station;
According to a specific embodiment of the invention, the hybrid particle swarm genetic algorithm is specifically based on a genetic algorithm, combined with elite retention strategy and particle swarm optimization ideas. After the non-dominant ordering of the original population is completed, the individuals in the non-dominant solution set are reserved, so that the individuals directly enter the next generation population, the non-dominant individuals can be prevented from generating negative variation, and the non-dominant solution is prevented from being lost in the genetic process.
According to a specific embodiment of the invention, a particle swarm optimization idea is adopted in a genetic iteration process, a self-cognition factor r= (n-rank)/n is calculated according to a ranking grade, wherein n is a maximum ranking grade, a social cognition factor is 1-r, individuals in a dominated solution are subjected to self-variation according to probability r, the bit sequences of two digits in coding are randomly exchanged, and a new individual generated enters a next generation population; and (3) cross mutation occurs according to the probability of 1-r, one section of codes is randomly selected, and replaced by elite individual codes, and the rest bit sequences are kept the same as the original codes under the condition that the numbers are not repeated.
On the one hand, the iterative process based on the particle swarm optimization idea ensures the randomness of variation, so that the algorithm has the capability of jumping out of a local optimal solution; on the other hand, the mutation directivity is ensured, the mutation of the dominant individual to the non-dominant individual approaches, and the probability of generating better individuals in the mutation process is improved.
According to a specific embodiment of the invention, in the process of solving the multi-target pareto optimal solution set by using a mixed particle swarm genetic algorithm, a good-bad solution distance method is adopted to solve the multi-target pareto optimal solution set.
According to one embodiment of the present invention, the carbon emission amount is calculated based on the vehicle workload and the carbon emission factor per unit of vehicle workload.
According to one embodiment of the invention, the vehicle workload is the product of distance and vehicle load capacity.
According to one embodiment of the invention, the vehicle workload difference is represented by a coefficient of variation of the vehicle workload.
According to one embodiment of the invention, the economic cost minimization is represented by the following objective function:
wherein ,
vencoding a vehicle;
i,jcoding nodes (including garbage generation nodes, transfer stations and kitchen garbage treatment plants);
ECis an economic cost;
N D generating a node set for garbage;
N T the garbage transfer stations are collected;
N K collecting kitchen waste treatment plants;
F v a fixed depreciation cost per day for vehicle v;
H v the human cost per day of vehicle v;
V 1 a vehicle set from a garbage generation point to a garbage transfer station;
V 2 collecting vehicles from a garbage generation point to a kitchen garbage treatment plant;
V 3 The method comprises the steps of collecting vehicles from a garbage transfer station to a kitchen garbage treatment plant;
x ijv decision variables for the vehicle v going from the refuse generating node i to the refuse generating node j;
d ij the distance between garbage generating nodes i and j;
According to one embodiment of the invention, the carbon emissions include not only carbon emissions produced during the vehicle operation phase, but also carbon emissions produced during the fuel production and transportation phase, the minimization of which is expressed by the following objective function:
wherein ,
CEis carbon emission;
According to one embodiment of the invention, vehicle queuing time minimization is represented by the following objective function:
wherein ,
VQ is the total queuing time of the vehicle;
qt iv the sum of queuing time of the vehicle v in the garbage transfer station and the kitchen garbage disposal plant is obtained.
According to one embodiment of the invention, the minimization of the vehicle working load difference is represented by the following objective function:
wherein ,
According to one embodiment of the invention, constraints of the optimization model are as follows:
each trash producing node is accessed once and only once by the same vehicle:
vehicles entering the debris generating node must leave the debris generating node:
in the classified intermodal mode, all of the refuse needs to be emptied after the multi-car vehicle reaches the refuse generating node:
in the classified and independent collection and transportation mode, after a single car reaches a garbage generation node, the garbage type (kitchen garbage/other garbage) responsible for collection and transportation needs to be emptied:
when the vehicle v starts from a garbage transfer station or a kitchen garbage station, the initial load capacity is 0:
the vehicle capacity constraints are:
vehicle V 1 Starting from the waste transfer station and finally needing to return to the waste transfer station:
the time when the vehicle arrives at the kitchen garbage disposal plant and the garbage transfer station is required to be within the range of a time window which allows the kitchen garbage disposal plant and the garbage transfer station to operate:
the queuing time of the vehicles in the garbage transfer station is as follows, wherein the number v is the arrangement of the sequence of the vehicles arriving at the garbage transfer station:
in the classified intermodal mode, V 3 The vehicle needs to transport all kitchen waste of the waste transfer station to a kitchen waste treatment plant:
Vehicle V 3 After the kitchen waste treatment plant goes to a waste transfer station for shipment, the kitchen waste is required to be returnedWaste treatment plant:
vehicle V 3 Is defined by the capacity constraints of:
according to one embodiment of the invention, algorithmic contrast analysis is performed by error rate, supersvolume, and computational efficiency.
Error rate is defined as the ratio of the number of dominant individuals in a population to the total number of individuals. It reflects the goodness of the model output result. The higher the error rate, the more suboptimal number of individuals in the population; conversely, the lower the error rate, the greater the number of pareto individuals in the population, indicating that the pareto solution set has diversity, and at the same time, indicating that the algorithm has higher reliability.
The supersolume index measures the volume of the target space governed by the pareto solution set of the algorithm with a set of preset reference points r distributed in the target space as boundaries. The convergence and diversity of the solution set can be comprehensively measured by the hyper-volume index (HV), and the larger the HV value is, the closer the pareto solution set obtained by the algorithm is to the real pareto front, and the solution set has better convergence and diversity.
The computational efficiency is specifically the algorithm run time, which is the run time required by the algorithm to achieve a particular quality solution set or to complete a particular number of iterations. The shorter the run time, the higher the solution efficiency of the algorithm. It is an important index for measuring the effectiveness of the algorithm.
According to a specific embodiment of the invention, in a multi-objective pareto optimal solution set obtained by solving by using a mixed particle swarm genetic algorithm, a good-bad solution distance method is adopted, the score of each scheme in the pareto solution set is calculated according to the preference weight, and the scheme with the highest score is selected as a final path optimization scheme.
According to a specific embodiment of the present invention, the scores of the schemes in the pareto solution set are calculated according to the preference weights, and the scheme with the highest score is selected as the final path optimization scheme calculation step:
carrying out standardized processing on the data, eliminating the data dimension, and obtaining a standardized matrix:
in the formula ,
is the schemeiTarget objectjTarget value of->
Is a standardized schemeiTarget objectjIs set to be a target value of (c),NMis a standardized matrix.
Multiplying the normalized target value by the target weight according to the decision preference to obtain a weighted normalized matrix:
in the formula ,
normalized scheme for weightingiTarget objectjIs set to be a target value of (c),WNMfor weighting the normalization matrix +.>
Is the object ofjWeights of (when the target isjWhen the target is a preference target, the value is 0.7; when the object isjFor a non-preference target, the value is 0.1).
Calculating the distance between the solution and the optimal target value and the worst target value:
wherein ,
、/>
respectively as targetsjOptimal and worst values of +.>
、/>
Respectively is the schemeiDistance from the optimal value and the worst value.
Calculating a score for the solution:
wherein ,
is the schemeiThe scheme with the highest score is selected as the final path optimization scheme.
According to a specific embodiment of the invention, the synergy or trade-off relationship between different optimization objectives is quantitatively analyzed by calculating the spearman correlation coefficients of the different optimization objectives in the pareto solution set.
According to one embodiment of the invention, the calculation formula of the spearman correlation coefficient is:
in the formula ,ris the correlation coefficient between the two objects,
is the rank after the first target rank, +.>
Is the rank after the second target ordering,nis the number of individuals in the pareto solution set.
According to a specific embodiment of the invention, different optimization schemes are determined for different preferences as follows:
for "economic cost preference": taking economic cost minimization as a main aim and taking other aims into consideration; the weight of the economic target is set to 0.7, the weights of other targets are set to 0.1, and the top solution is selected from the pareto optimal solution set by using the TOPSIS method to obtain the final solution.
For "environmental benefit preference": taking carbon emission minimization as a main aim and taking other aims into consideration; the weight of the carbon emission target is set to 0.7, the weights of other targets are set to 0.1, and the top solution is selected from the pareto optimal solution set by using the TOPSIS method to obtain the final solution.
For "time optimization preference": taking the vehicle queuing time minimization as a main target and taking other targets into consideration; the weight of the queuing time target is set to be 0.7, the weights of other targets are set to be 0.1, and the top solution is selected from the pareto optimal solution set by using the TOPSIS method to be used as a final solution.
For "vehicle management preference": taking the vehicle working quantity difference minimization as a main target and taking other targets into consideration; the weight of the different targets of the vehicle working quantity is set to be 0.7, the weights of other targets are set to be 0.1, and the top optimization scheme with the highest score is selected from the pareto optimal solution set by using the TOPSIS method to serve as a final scheme.
For "balanced decision preference": the weights of the four targets are set to be 0.25, and the top solution is selected from the pareto optimal solution set by using the TOPSIS method to obtain the final solution.
For collaborative preferences: when the correlation coefficient of the two targets is more than or equal to 0.8, the two targets are considered to have a strong synergistic relationship, the two target preferences can be combined, the weights of the two targets are respectively set to 0.4, and the weights of other targets are set to 0.1.
According to a specific embodiment of the invention, the code of the vehicle path takes a positive real number as the number of the garbage generation nodes, and the sequence of the garbage generation nodes with the length being the same as the number of the garbage generation nodes represents the running sequence of the vehicle among the nodes to be received; in view of vehicle capacity constraints, the decoding process of the vehicle path is divided into three steps:
dividing a garbage generation node sequence according to the garbage generation amount and the vehicle capacity of each garbage generation node, and distributing the garbage generation node sequence to different vehicles;
representing a garbage transfer station by a value of 0, inserting the garbage transfer station into an original garbage generation node sequence, and acquiring the actual running sequence of a vehicle between the garbage transfer station and a node to be received;
calculating the economic cost, the vehicle queuing time, the vehicle workload difference and the value of each objective function of the carbon emission according to the distance matrix between the nodes to be received and the vehicle parameters, the transportation cost parameters and the carbon emission factors, wherein the vehicle parameters comprise the vehicle capacity, the average running speed, the carbon emission of the unit workload, the time required for loading garbage and the time required for unloading garbage;
the transportation cost parameters include a daily fixed depreciation cost of the vehicle, a daily labor cost of the vehicle, and a unit distance transportation cost.
According to a specific embodiment of the invention, the optimization process comprises the following steps:
collecting the coordinate position of the garbage generation node, the garbage generation amount of the garbage generation node and the road network vector data, calculating a road network distance matrix between the nodes by using an ArcGIS network analysis tool, and taking the garbage generation amount and the distance matrix as input data of an optimization process;
respectively establishing a classification intermodal transportation mode and a household garbage collection and transportation path optimization model of a classification independent collection and transportation mode;
for each garbage collection mode, a mixed particle swarm genetic algorithm (HPSO-GA algorithm) and a rapid non-dominant sorting genetic algorithm (NSGA-II algorithm) are respectively used for solving a multi-objective pareto optimal solution set so as to carry out comparison analysis on algorithm performances;
verifying the effectiveness of an optimization algorithm through error rate indexes, super-volume indexes and calculation efficiency evaluation algorithm performance;
calculating spearman correlation coefficients of different optimization targets in the pareto solution set, quantitatively analyzing synergy or trade-off relation among the different targets, setting decision preference of a path scheme, and determining specific optimization schemes under different preference by a TOPSIS method;
and determining optimal household garbage collection and transportation paths in different collection and transportation modes according to the optimization schemes in different preferences, and selecting a final collection and transportation mode and the optimal household garbage collection and transportation path corresponding to the final collection and transportation mode according to the optimal household garbage collection and transportation paths in different collection and transportation modes.
The invention provides a classified household garbage collection and transportation path multi-target optimization system, which comprises a path optimization module, wherein the path optimization module performs household garbage collection and transportation path multi-target optimization by using any classified household garbage collection and transportation path multi-target optimization method, and determines an optimal household garbage collection and transportation path.
Example 1
According to a specific embodiment of the invention, the classification-oriented domestic garbage collection and transportation path multi-objective optimization method is described in detail with reference to the accompanying drawings.
The invention provides a classification-oriented household garbage collection and transportation path multi-objective optimization method, which comprises the following steps:
respectively establishing household garbage collection and transportation path optimization models of different collection and transportation modes, and determining an optimal household garbage collection and transportation path by simultaneously optimizing the following four targets: economic cost, carbon emissions, vehicle queuing time, and vehicle workload variance; minimizing economic cost, carbon emissions, vehicle queuing time, and workload variance as optimization objectives; the different collection and transportation modes comprise a classification combined collection and transportation mode and a classification independent collection and transportation mode, wherein the classification combined collection and transportation mode is that a vehicle with a compartment starts from a garbage transfer station and goes to each garbage generation point, different types of garbage are loaded into corresponding compartments, and when any compartment reaches capacity constraint, the garbage is returned to the garbage transfer station; the kitchen waste treatment dispatch vehicle goes to a waste transfer station and conveys the kitchen waste to a kitchen waste treatment plant; the separate classification and collection mode is that kitchen waste and other waste are respectively transported by different vehicles, and the kitchen waste collection process does not pass through a waste transfer station;
Adopting a mixed particle swarm genetic algorithm (HPSO-GA algorithm), solving a multi-objective pareto optimal solution set, and carrying out multi-objective optimization; the mixed particle swarm genetic algorithm is based on a genetic algorithm, and meanwhile, the particle swarm optimization thought is considered;
quantitatively analyzing the coordination or trade-off relation among different optimization targets, merging targets with the quantized coordination or trade-off relation exceeding a preset threshold, setting decision preference of a path scheme, setting weights of the targets according to the set decision preference, and determining optimal household garbage collection and transportation paths of different collection and transportation modes under different preferences;
identifying the optimal household garbage collection and transportation mode and the corresponding optimal household garbage collection and transportation path under each decision preference by comparing the target values of the optimal household garbage collection and transportation paths under different collection and transportation modes;
the household garbage collection and transportation path optimization model comprises links from a garbage transfer station to a garbage generation node, from the garbage transfer station to a kitchen garbage treatment plant and from the kitchen garbage treatment plant to the garbage generation node.
Example 2
According to a specific embodiment of the invention, the classification-oriented domestic garbage collection and transportation path multi-objective optimization method is described in detail with reference to the accompanying drawings.
The invention provides a classification-oriented household garbage collection and transportation path multi-objective optimization method, which comprises the following steps:
respectively establishing household garbage collection and transportation path optimization models of different collection and transportation modes, and determining an optimal household garbage collection and transportation path by simultaneously optimizing the following four targets: economic cost, carbon emissions, vehicle queuing time, and vehicle workload variance; minimizing economic cost, carbon emissions, vehicle queuing time, and workload variance as optimization objectives; the different collection and transportation modes comprise a classification combined collection and transportation mode and a classification independent collection and transportation mode, wherein the classification combined collection and transportation mode is that a vehicle with a compartment starts from a garbage transfer station and goes to each garbage generation point, different types of garbage are loaded into corresponding compartments, and when any compartment reaches capacity constraint, the garbage is returned to the garbage transfer station; the kitchen garbage treatment plant dispatches vehicles to a garbage transfer station to transport kitchen garbage to the kitchen garbage treatment plant; the separate classification and collection mode is that kitchen waste and other waste are respectively transported by different vehicles, and the kitchen waste collection process does not pass through a waste transfer station;
adopting a mixed particle swarm genetic algorithm (HPSO-GA algorithm), solving a multi-objective pareto optimal solution set, and carrying out multi-objective optimization; the mixed particle swarm genetic algorithm is based on a genetic algorithm, and meanwhile, the particle swarm optimization thought is considered;
Quantitatively analyzing the coordination or trade-off relation among different optimization targets, merging targets with the quantized coordination or trade-off relation exceeding a preset threshold, setting decision preference of a path scheme, setting weights of the targets according to the set decision preference, and determining optimal household garbage collection and transportation paths of different collection and transportation modes under different preferences;
identifying the optimal household garbage collection and transportation mode and the corresponding optimal household garbage collection and transportation path under each decision preference by comparing the target values of the optimal household garbage collection and transportation paths under different collection and transportation modes;
the household garbage collection and transportation path optimization model comprises links from a garbage transfer station to a garbage generation node, from the garbage transfer station to a kitchen garbage treatment plant and from the kitchen garbage treatment plant to the garbage generation node.
The mixed particle swarm genetic algorithm is specifically based on a genetic algorithm, and combines an elite retention strategy and a particle swarm optimization idea.
And solving the multi-target pareto optimal solution set by adopting a good-bad solution distance method in the multi-target pareto optimal solution set solved by using a mixed particle swarm genetic algorithm.
Wherein the carbon emission amount is calculated based on the vehicle workload and a carbon emission factor per unit vehicle workload.
Wherein, the vehicle workload is the product of the distance and the vehicle load capacity.
Wherein the vehicle workload difference is represented by a coefficient of variation of the vehicle workload.
Wherein the economic cost minimization is represented by the following objective function:
wherein ,
vencoding a vehicle;
i,jcoding nodes (including garbage generation nodes, transfer stations and kitchen garbage treatment plants);
ECis an economic cost;
N D generating a node set for garbage;
N T the garbage transfer stations are collected;
N K collecting kitchen waste treatment plants;
F v a fixed depreciation cost per day for vehicle v;
H v the human cost per day of vehicle v;
V 1 a vehicle set from a garbage generation point to a garbage transfer station;
V 2 collecting vehicles from a garbage generation point to a kitchen garbage treatment plant;
V 3 in garbageThe station is transferred to a kitchen waste treatment plant vehicle collection;
x ijv decision variables for the vehicle v going from the refuse generating node i to the refuse generating node j;
d ij the distance between garbage generating nodes i and j;
Wherein the carbon emissions include not only carbon emissions generated during the vehicle operation phase, but also carbon emissions generated during the fuel production and transportation phase, the minimization of which is expressed by the following objective function:
wherein ,
CEIs carbon emission;
Wherein the vehicle queuing time minimization is represented by the following objective function:
wherein ,
VQ is the total queuing time of the vehicle;
qt iv the sum of queuing time of the vehicle v in the garbage transfer station and the kitchen garbage disposal plant is obtained.
Wherein the vehicle workload minimization is represented by the following objective function:
wherein ,
Wherein the comparative analysis is performed by error rate, supersvolume, and computational efficiency.
And in a multi-target pareto optimal solution set obtained by solving by using a mixed particle swarm genetic algorithm, calculating the scores of all schemes in the pareto optimal solution set according to the preference weights by adopting a good-bad solution distance method, and selecting the scheme with the highest score as a final path optimization scheme.
And quantitatively analyzing the synergy or trade-off relation between different optimization targets by calculating the spearman correlation coefficients of the different optimization targets in the pareto solution set.
The method comprises the steps of taking a positive real number as a garbage generation node number in the vehicle path coding, and representing the running sequence of vehicles among nodes to be received by using a garbage generation node sequence with the same length as the number of the garbage generation nodes; in view of vehicle capacity constraints, the decoding process of the vehicle path is divided into three steps:
dividing a garbage generation node sequence according to the garbage generation amount and the vehicle capacity of each garbage generation node, and distributing the garbage generation node sequence to different vehicles;
representing a garbage transfer station by a value of 0, inserting the garbage transfer station into an original garbage generation node sequence, and acquiring the actual running sequence of a vehicle between the garbage transfer station and a node to be received;
calculating the values of objective functions of economic cost, vehicle queuing time, vehicle workload difference and carbon emission of a garbage collection and transportation scheme according to the distance matrix between nodes to be collected and transported, vehicle parameters, transportation cost parameters and carbon emission factors; vehicle parameters include vehicle capacity, average travel speed, carbon emissions per unit of work, time required to load garbage, and time required to unload garbage;
The transportation cost parameters include a daily fixed depreciation cost of the vehicle, a daily labor cost of the vehicle, and a unit distance transportation cost.
Wherein, when the rapid non-dominant sorting is adopted, the method comprises the following steps:
firstly, comparing objective function values of individuals in an original population with other individuals so as to determine a non-dominant solution set, and marking the ranking grade as rank=1;
then separating the non-dominant solution set from the original population, re-ordering the rest individuals, determining the non-dominant solution set in the rest individuals, and marking the ordering grade as rank=2;
and so on until the non-dominant ordering of all individuals is completed.
Example 3
The collection and transportation paths of the citizen garbage transfer stations and the corresponding garbage generation nodes in a certain place are optimized.
The number of the garbage generation nodes served by the garbage transfer station is 29 cells, and the garbage amount of each garbage generation node is an average value calculated based on actual historical data of 12 months and 30 days in 2020. The inter-node distance matrix is obtained based on an actual road network and GIS network analysis method. The main parameter settings and data sources of the model are shown in Table 1.
Wherein the carbon emission factor setting between the refuse transfer station and the refuse-producing source vehicle is referenced by Sharafi A, bashiri M. Green Vehicle Routing Problem with Safety and Social Concerns [ J ]. Journal of Optimization in Industrial Engineering, 2016,10 (21): 93-100;
Carbon emission factor setting reference Chinese Academy of Environmental Planning, beijing Normal University, sun Yat-Sen University, china City Greenhouse Gas Working group, china Products Carbon Footprint Factors Database (2022) [ R ]. Beijin, 2022;
vehicle loading time and classified individual collection and delivery vehicle unloading time setting reference Zhang Duoyu. Garbage collection and delivery route optimization [ J ] introducing loading and unloading time. Logistics engineering and management 2016,38 (07): 125-127;
classified intermodal vehicle unloading time settings reference Wilson B G, vincent J K Estimating Waste transfer station delays using GPS [ J ]. Waste Management, 2008, 28 (10): 1742-1750
Table 1: optimizing parameters of case model in certain place
The HPSO-GA method and NSGA II are used for solving respectively, and the comparison of the optimization results is used for verifying that the HPSO-GA adopted by the invention is superior to NSGA II when solving the multi-objective path optimization problem.
When the rapid non-dominant sorting is adopted, the method comprises the following steps:
firstly, comparing objective function values of individuals in an original population with other individuals so as to determine a non-dominant solution set, and marking the ranking grade as rank=1;
Then separating the non-dominant solution set from the original population, re-ordering the rest individuals, determining the non-dominant solution set in the rest individuals, and marking the ordering grade as rank=2;
and so on until the non-dominant ordering of all individuals is completed.
Algorithms were encoded using the python language and run on a personal computer (Intel Core i5-11260H@2.60GHz CPU,Ubuntu64 bit virtual machine operating system, 4GB memory). And when the iteration times are given, the time required by the HPSO-GA to complete the case optimization problem is smaller than NSGA II, and the algorithm has higher operation efficiency. Furthermore, the HPSO-GA solution error rate is lower in different transportation scenarios. Under the classified intermodal scenario, the HV indexes of HPSO-GA and NSGA II solution sets are relatively close and stable at about 0.9. Under the situation of separate transportation by classification, as the iteration times are increased, the HV index of the HPSO-GA solution set is stabilized at about 0.95, the HV index of the NSGA II solution set is stabilized at about 0.85, and the convergence and diversity of the solution set are better.
By combining the results, in the model optimization process of a certain place, the HPSO-GA provided by the invention is an effective multi-objective path optimization problem solving algorithm; compared with NSGA II, HPSO-GA has advantages in solving efficiency, solution convergence, diversity and the like.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.
Claims (10)
1. A multi-objective optimization method for classified household garbage collection and transportation paths is characterized by comprising the following steps:
respectively establishing household garbage collection and transportation path optimization models of different collection and transportation modes, and determining an optimal household garbage collection and transportation path by simultaneously optimizing the following four targets; minimizing economic cost, carbon emissions, vehicle queuing time, and workload variance as optimization objectives; the different collection and transportation modes comprise a classification combined collection and transportation mode and a classification independent collection and transportation mode, wherein the classification combined collection and transportation mode is that a vehicle with a compartment starts from a garbage transfer station and goes to each garbage generation point, different types of garbage are loaded into corresponding compartments, and when any compartment reaches capacity constraint, the garbage is returned to the garbage transfer station; the kitchen garbage treatment plant dispatches vehicles to a garbage transfer station to transport kitchen garbage to the kitchen garbage treatment plant; the separate classification and collection mode is that kitchen waste and other waste are respectively transported by different vehicles, and the kitchen waste collection process does not pass through a waste transfer station;
Adopting a mixed particle swarm genetic algorithm to solve a multi-target pareto optimal solution set, and carrying out multi-target optimization; the mixed particle swarm genetic algorithm is based on a genetic algorithm, and meanwhile, the particle swarm optimization thought is considered;
quantitatively analyzing the coordination or trade-off relation among different optimization targets, merging targets with the quantized coordination or trade-off relation exceeding a preset threshold, setting decision preference of a path scheme, setting weights of the targets according to the set decision preference, and determining optimal household garbage collection and transportation paths of different collection and transportation modes under different preferences;
identifying the optimal household garbage collection and transportation mode and the corresponding optimal household garbage collection and transportation path under each decision preference by comparing the target values of the optimal household garbage collection and transportation paths under different collection and transportation modes;
the household garbage collection and transportation path optimization model comprises links from a garbage transfer station to a garbage generation node, from the garbage transfer station to a kitchen garbage treatment plant and from the kitchen garbage treatment plant to the garbage generation node.
2. The classification-oriented domestic waste collection and transportation path multi-objective optimization method according to claim 1, wherein the mixed particle swarm genetic algorithm is specifically based on a genetic algorithm, and an elite retention strategy and a particle swarm optimization idea are combined.
3. The classification-oriented domestic waste collection path multi-objective optimization method according to claim 1, wherein the carbon emission amount is calculated based on the vehicle workload and a carbon emission factor per unit vehicle workload.
4. The classification-oriented domestic waste collection path multi-objective optimization method according to claim 1, wherein the vehicle workload difference is represented by a coefficient of variation of the vehicle workload.
5. The multi-objective optimization method for classified domestic waste collection and transportation paths according to claim 1, wherein the vehicle workload is a product of distance and vehicle load capacity.
6. The classification-oriented domestic waste collection path multi-objective optimization method of claim 1, wherein the carbon emissions include not only carbon emissions generated during the vehicle operation phase, but also carbon emissions generated during the fuel production and transportation phase, the minimization of which is expressed by an objective function of:
wherein ,
CEis carbon emission;
7. The multi-objective optimization method for classified domestic garbage collection and transportation paths according to claim 1, wherein in a multi-objective pareto optimal solution set obtained by solving by using a mixed particle swarm genetic algorithm, a good-bad solution distance method is adopted, the scores of all schemes in the pareto solution set are calculated according to preference weights, and the scheme with the highest score is selected as a final path optimization scheme.
8. The multi-objective optimization method for classified domestic garbage collection and transportation paths according to claim 7, wherein the synergy or trade-off relationship between different optimization objectives is quantitatively analyzed by calculating spearman correlation coefficients of the different optimization objectives in the pareto solution set.
9. The multi-objective optimization method for classified domestic garbage collection and transportation paths according to any one of claims 1 to 8, wherein the codes of the vehicle paths are numbered by taking positive real numbers as garbage generation nodes, and the sequences of the garbage generation nodes with the same length as the number of the garbage generation nodes represent the driving sequence of vehicles among the nodes to be collected and transported; in view of vehicle capacity constraints, the decoding process of the vehicle path is divided into three steps:
Dividing a garbage generation node sequence according to the garbage generation amount and the vehicle capacity of each garbage generation node, and distributing the garbage generation node sequence to different vehicles;
representing a garbage transfer station by a value of 0, inserting the garbage transfer station into an original garbage generation node sequence, and acquiring the actual running sequence of a vehicle between the garbage transfer station and a node to be received;
calculating the values of objective functions of economic cost, vehicle queuing time, vehicle workload difference and carbon emission of a garbage collection and transportation scheme according to the distance matrix between nodes to be collected and transported, vehicle parameters, transportation cost parameters and carbon emission factors; vehicle parameters include vehicle capacity, average travel speed, carbon emissions per unit of work, time required to load garbage, and time required to unload garbage;
the transportation cost parameters include a daily fixed depreciation cost of the vehicle, a daily labor cost of the vehicle, and a unit distance transportation cost.
10. A classification-oriented household garbage collection and transportation path multi-objective optimization system, which is characterized by comprising a path optimization module, wherein the path optimization module performs household garbage collection and transportation path multi-objective optimization by using the classification-oriented household garbage collection and transportation path multi-objective optimization method according to any one of claims 1-9, and determines an optimal household garbage collection and transportation path.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310387128.7A CN116127857B (en) | 2023-04-12 | 2023-04-12 | Classification-oriented household garbage collection and transportation path multi-objective optimization method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310387128.7A CN116127857B (en) | 2023-04-12 | 2023-04-12 | Classification-oriented household garbage collection and transportation path multi-objective optimization method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116127857A true CN116127857A (en) | 2023-05-16 |
| CN116127857B CN116127857B (en) | 2023-09-22 |
Family
ID=86306564
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310387128.7A Active CN116127857B (en) | 2023-04-12 | 2023-04-12 | Classification-oriented household garbage collection and transportation path multi-objective optimization method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116127857B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116384611A (en) * | 2023-06-01 | 2023-07-04 | 中诚华隆计算机技术有限公司 | Garbage transportation route optimization method and device based on quantum genetic algorithm |
| CN116562486A (en) * | 2023-06-01 | 2023-08-08 | 中诚华隆计算机技术有限公司 | Optimization method and device for urban household garbage transportation route |
| CN116974241A (en) * | 2023-07-10 | 2023-10-31 | 清华大学 | Geometric optimization method and device for numerical control machine tool for green low-carbon manufacturing |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109284875A (en) * | 2018-11-12 | 2019-01-29 | 国网冀北综合能源服务有限公司 | Energy internet Multipurpose Optimal Method |
| EP3546890A1 (en) * | 2018-03-27 | 2019-10-02 | Fundacíon Tecnalia Research & Innovation | Method, system and computer program product for computing routes |
| CN111815048A (en) * | 2020-07-07 | 2020-10-23 | 江南大学 | Low-carbon garbage classification receiving and transporting method and system based on improved brainstorming algorithm |
| CN114330874A (en) * | 2021-12-28 | 2022-04-12 | 扬州大学 | Vehicle collecting and transporting scheduling method and system based on urban household garbage classification |
| CN114386792A (en) * | 2021-12-28 | 2022-04-22 | 扬州青商欣环境科技有限公司 | Urban domestic waste collection and transportation vehicle scheduling method based on genetic algorithm |
| CN115759396A (en) * | 2022-11-15 | 2023-03-07 | 昆明理工大学 | Multi-dimensional ant colony optimization method and system for energy-saving scheduling in garbage classification recycling process |
| CN115907254A (en) * | 2022-11-23 | 2023-04-04 | 东北大学 | Vehicle route planning method based on evolutionary multi-objective reinforcement learning |
-
2023
- 2023-04-12 CN CN202310387128.7A patent/CN116127857B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3546890A1 (en) * | 2018-03-27 | 2019-10-02 | Fundacíon Tecnalia Research & Innovation | Method, system and computer program product for computing routes |
| CN109284875A (en) * | 2018-11-12 | 2019-01-29 | 国网冀北综合能源服务有限公司 | Energy internet Multipurpose Optimal Method |
| CN111815048A (en) * | 2020-07-07 | 2020-10-23 | 江南大学 | Low-carbon garbage classification receiving and transporting method and system based on improved brainstorming algorithm |
| CN114330874A (en) * | 2021-12-28 | 2022-04-12 | 扬州大学 | Vehicle collecting and transporting scheduling method and system based on urban household garbage classification |
| CN114386792A (en) * | 2021-12-28 | 2022-04-22 | 扬州青商欣环境科技有限公司 | Urban domestic waste collection and transportation vehicle scheduling method based on genetic algorithm |
| CN115759396A (en) * | 2022-11-15 | 2023-03-07 | 昆明理工大学 | Multi-dimensional ant colony optimization method and system for energy-saving scheduling in garbage classification recycling process |
| CN115907254A (en) * | 2022-11-23 | 2023-04-04 | 东北大学 | Vehicle route planning method based on evolutionary multi-objective reinforcement learning |
Non-Patent Citations (2)
| Title |
|---|
| MOHAMMAD MOJTAHEDI ET AL.: "Sustainable vehicle routing problem for coordinated solid waste management", 《JOURNAL OF INDUSTRIAL INFORMATION INTEGRATION》, pages 2 - 21 * |
| 付文静: "城市生活垃圾分类收运系统的电动车辆调度及路径优化", 《中国优秀博硕士学位论文全文数据库(电子期刊) 工程科技Ⅱ辑》, pages 25 - 84 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116384611A (en) * | 2023-06-01 | 2023-07-04 | 中诚华隆计算机技术有限公司 | Garbage transportation route optimization method and device based on quantum genetic algorithm |
| CN116562486A (en) * | 2023-06-01 | 2023-08-08 | 中诚华隆计算机技术有限公司 | Optimization method and device for urban household garbage transportation route |
| CN116384611B (en) * | 2023-06-01 | 2023-08-11 | 中诚华隆计算机技术有限公司 | Garbage transportation route optimization method and device based on quantum genetic algorithm |
| CN116562486B (en) * | 2023-06-01 | 2023-11-28 | 中诚华隆计算机技术有限公司 | Optimization method and device for urban household garbage transportation route |
| CN116974241A (en) * | 2023-07-10 | 2023-10-31 | 清华大学 | Geometric optimization method and device for numerical control machine tool for green low-carbon manufacturing |
| CN116974241B (en) * | 2023-07-10 | 2024-02-06 | 清华大学 | Geometric optimization method and device for numerical control machine tool for green low-carbon manufacturing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116127857B (en) | 2023-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116127857B (en) | Classification-oriented household garbage collection and transportation path multi-objective optimization method and system | |
| Xu et al. | A model for capacitated green vehicle routing problem with the time-varying vehicle speed and soft time windows | |
| Liao et al. | Multi-objective green meal delivery routing problem based on a two-stage solution strategy | |
| Ma et al. | A multi-objective robust optimization model for customized bus routes | |
| Goli et al. | RETRACTED ARTICLE: Two-echelon electric vehicle routing problem with a developed moth-flame meta-heuristic algorithm | |
| Yin | Multiobjective optimization for vehicle routing optimization problem in low-carbon intelligent transportation | |
| Hao et al. | Research on e-commerce distribution optimization of rice agricultural products based on consumer satisfaction | |
| Shen et al. | Energy-efficient multi-trip routing for municipal solid waste collection by contribution-based adaptive particle swarm optimization | |
| Tiwari et al. | An optimization model for vehicle routing problem in last-mile delivery | |
| CN116882883A (en) | Power grid material distribution path planning method utilizing big data | |
| Normasari et al. | Heterogeneous Fleet Green Vehicle Routing Problem: A Literature | |
| Akkad et al. | ENERGY EFFICIENCY OPTIMIZATION OF LAST MILE SUPPLY SYSTEM WITH REVERSE LOGISTICS CONSIDERATION. | |
| Qiao et al. | Analysis of logistics linkage by digital twins technology and lightweight deep learning | |
| Zhang et al. | Distribution path optimization method of gas cylinder based on Genetic-Tabu hybrid algorithm | |
| Zhang et al. | Shared bikes scheduling under users’ travel uncertainty | |
| CN114611864A (en) | Garbage vehicle low-carbon scheduling method and system | |
| Li et al. | Freshness-driven vehicle routing problem: Modeling and application to the fresh agricultural product pick-storage-transportation | |
| Hu et al. | Municipal solid waste collection and transportation routing optimization based on IAC-SFLA | |
| Yang et al. | Location Optimization for Community Smart Parcel Lockers Based on Bilevel Programming | |
| Li | Research on open vehicle routing problem with time windows based on improved genetic algorithm | |
| Zhu | Mobile battery replacement service routing problem for shared electric bikes based on electric logistics vehicles | |
| Pendyala et al. | An Overview of Carbon Footprint Mitigation Strategies. Machine Learning for Societal Improvement, Modernization, and Progress | |
| Peng et al. | Effect of carbon emission policy on fresh aquatic product carriers in augmented reality respective: a parallel simulation research based on genetic algorithm | |
| Shi-yue | Research on Open Vehicle Routing Problem based on Multi-Objective and Multi-Constrain Genetic Algorithm | |
| Shangguan | The Location-Routing Problem of Electric Vehicles in City Logistics |
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 |