Shared bicycle dispatching method, device, equipment and readable storage medium

Abstract

The present invention provides a shared bicycle dispatching method, device, equipment and readable storage medium, which relate to the field of shared bicycle technology. The method comprises: determining a dispatching center in a target area; constructing a shared bicycle demand prediction model for each placement point in the target area; and determining the number of shared bicycles dispatched by the dispatching center to each placement point according to the shared bicycle demand prediction model. The solution of the present invention can predict the demand quantity of shared bicycles at each placement point, and the dispatching center can perform the dispatching to avoid uneven delivery.

Description

Technical Field

[0001] The present invention belongs to the technical field of shared bicycles, and in particular relates to a shared bicycle dispatching method, device and electronic equipment. Background Art

[0002] As a green travel mode, dockless shared bicycles have solved the "last mile" problem, which not only facilitates users' travel, but also alleviates urban congestion to a certain extent. However, the explosive development of shared bicycles has also brought certain negative effects. The following problems need to be solved in the operation and management process:

[0003] (1) Major bike-sharing manufacturers are deploying bikes in an unordered manner

[0004] As shared bicycles quickly became popular, major manufacturers noticed the huge economic benefits brought by this industry. Therefore, shared bicycle companies that were unwilling to lag behind increased their deployment in order to occupy a larger share of the market. After this large-scale, disorderly deployment, shared bicycles did not play the role they should have in alleviating traffic pressure, but instead aggravated the congestion of urban roads. At present, due to the irregular parking of shared bicycles, bicycles are parked everywhere on motorway lanes, sidewalks, and green belts on the streets, seriously affecting public order. In particular, these shared bicycles are parked randomly around subway stations, making the areas around subway stations, which already have a large traffic volume, even more congested, causing a lot of inconvenience to normal travel.

[0005] (2) Imbalance in demand for shared bicycles

[0006] Shared bicycle travel has different time and space characteristics from other modes of transportation. It is also because of these unique time and space characteristics that the demand for shared bicycles is unbalanced. From a temporal perspective, the demand for shared bicycles to be borrowed and returned at each station is constantly changing throughout the day. During peak hours, the frequency of bicycle use is high and the number of bicycles borrowed and returned is large, while other periods are relatively low. From a spatial perspective, the travel intensity in different areas of the city is different, resulting in differences in the demand for shared bicycles to be borrowed and returned at stations with different land uses during the same period. Due to the imbalance in the demand for shared bicycles, there are often no bicycles available in hot spots during peak hours, which is also one of the main problems facing the shared bicycle industry.

[0007] (3) Damage and repair issues of shared bicycles

[0008] It is inevitable that shared bicycles will gradually wear out after being put on the market. However, malicious vandalism such as lost handlebars, seats, cracked tires, and tampered QR codes occur frequently, resulting in a high failure rate for shared bicycles, which not only brings huge economic losses to shared bicycle companies, but also reduces user satisfaction. Summary of the invention

[0009] The purpose of the embodiments of the present invention is to provide a shared bicycle scheduling method, device, equipment and readable storage medium, so as to solve the problem of improper scheduling of shared bicycles in the prior art, resulting in uneven deployment.

[0010] In a first aspect, an embodiment of the present invention provides a shared bicycle scheduling method, comprising:

[0011] Identify dispatch centers in target areas;

[0012] Constructing a shared bicycle demand prediction model for each placement point in the target area;

[0013] The number of shared bicycles to be dispatched by the dispatch center to each placement point is determined according to the shared bicycle demand prediction model.

[0014] Optionally, determining the dispatch center in the target area includes:

[0015] Obtaining coordinate information of each placement point in the target area;

[0016] Constructing a dispatch center operation cost function according to the coordinate information of each placement point;

[0017] Based on the dispatch center operation cost function, obtaining target coordinate information corresponding to the minimum value of the dispatch center operation cost;

[0018] The dispatch center is determined according to the target coordinate information.

[0019] Optionally, constructing a dispatch center operation cost function according to the coordinate information of each placement point includes:

[0020] Constructing the operation cost function of the dispatch center according to the coordinate information of each placement point, the coordinate information of the candidate dispatch center, the unit distance cost and the personnel cost;

[0021] Among them, the candidate dispatch center is a location point in the target area other than the placement point.

[0022] Optionally, obtaining target coordinate information corresponding to a minimum value of the dispatch center operation cost based on the dispatch center operation cost function includes:

[0023] Acquire multiple first candidate dispatch centers, and encode the coordinate information of the multiple first candidate dispatch centers respectively to acquire the encoded coordinates of the multiple first candidate dispatch centers;

[0024] Calculating the fitness of the plurality of first candidate dispatching centers respectively according to the dispatching center operation cost function and the coordinate information of the plurality of first candidate dispatching centers;

[0025] Based on the encoded coordinates and fitness of the multiple first candidate dispatching centers, genetic iteration is performed to obtain a target dispatching center corresponding to the minimum dispatching center operation cost;

[0026] Obtain target coordinate information of the target dispatch center;

[0027] The first candidate dispatch center is a location point in the target area other than the placement point.

[0028] Optionally, performing genetic iteration based on the coded coordinates and fitness of the plurality of first candidate dispatch centers to obtain a target dispatch center corresponding to a minimum dispatch center operation cost includes:

[0029] Selecting a plurality of second candidate dispatching centers from the plurality of first candidate dispatching centers according to the fitness corresponding to each first candidate dispatching center;

[0030] Processing the coded coordinates of the plurality of second candidate dispatch centers by gene recombination and / or gene mutation, selecting a plurality of third candidate dispatch centers from the plurality of second candidate dispatch centers, and obtaining the coded coordinates and fitness of each third candidate dispatch center;

[0031] Repeat the steps of obtaining the plurality of second candidate dispatch centers and the plurality of third candidate dispatch centers until a preset number of iterations is reached, obtain the encoded coordinates of the plurality of fourth candidate dispatch centers, and obtain the fitness of each fourth candidate dispatch center;

[0032] Among the multiple fourth candidate dispatching centers, a fourth candidate dispatching center with the greatest fitness is determined as the target dispatching center.

[0033] Optionally, constructing a shared bicycle demand prediction model for each placement point in the target area includes:

[0034] Obtain multiple potential factors that affect shared bicycle travel;

[0035] Obtaining travel data of the shared bicycles at each placement point;

[0036] Determining, based on the travel data, a plurality of target factors among the plurality of potential factors, and determining an influencing factor function of each target factor;

[0037] Fitting the influencing factor function of each target factor to construct the shared bicycle demand prediction model;

[0038] The target factor includes at least one of the following:

[0039] Age factors;

[0040] Gender factors;

[0041] Weather factors;

[0042] Travel distance factor.

[0043] Optionally, the obtaining of the travel data of the shared bicycles at each placement point includes:

[0044] Acquire the travel data of the shared bicycles at each placement point whose reliability coefficient is within a preset range.

[0045] Optionally, determining a plurality of target factors among the plurality of potential factors according to the travel data includes:

[0046] Obtaining a correlation between each potential factor and the travel data;

[0047] The multiple target factors are determined according to the correlation.

[0048] Optionally, the dispatching center dispatches shared bicycles to multiple placement points in the target area according to the shared bicycle demand prediction model, including:

[0049] Based on the shared bicycle demand prediction model, obtaining the demand for shared bicycles at each placement point;

[0050] Based on the demand, the number of shared bicycles to be dispatched by the dispatch center to each placement point is determined.

[0051] The second invention, an embodiment of the present invention further provides a shared bicycle dispatching device, comprising:

[0052] A first determination module is used to determine a dispatch center in a target area;

[0053] A first construction module is used to construct a shared bicycle demand prediction model for each placement point in the target area;

[0054] The second determination module is used to determine the number of shared bicycles to be dispatched by the dispatch center to each placement point according to the shared bicycle demand prediction model.

[0055] In a third aspect, an embodiment of the present invention further provides a shared bicycle scheduling device, comprising: a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor; the processor is used to read the program in the memory to implement the steps in the shared bicycle scheduling method described in the first aspect above.

[0056] In a fourth aspect, an embodiment of the present invention further provides a readable storage medium, comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the program is executed by the processor, the steps in the shared bicycle scheduling method described in the first aspect above are implemented.

[0057] The above technical solution of the present invention has at least the following beneficial effects:

[0058] In the above scheme, by determining the dispatching center in the target area, a shared bicycle demand prediction model for each placement point in the target area is constructed, and based on the shared bicycle demand prediction model, the number of shared bicycles to be dispatched by the dispatching center to each placement point is determined. The number of shared bicycles at each placement point can be predicted, and the dispatching center can dispatch shared bicycles to achieve on-demand deployment, avoid uneven deployment, and reduce operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0060] Figure 1 A flow chart of a shared bicycle scheduling method provided by an embodiment of the present invention;

[0061] Figure 2 is a schematic diagram of a target area provided by an embodiment of the present invention;

[0062] Figure 3 is a structural block diagram of a shared bicycle dispatching device provided by an embodiment of the present invention;

[0063] Figure 4 It is a schematic diagram of the hardware structure of the shared bicycle dispatching device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] In order to enable those skilled in the art to better understand the embodiments of the present invention, the genetic algorithm is first described as follows.

[0066] Scientific definition of genetic algorithm:

[0067] The concept of genetic algorithm originally came from a computer science simulation method based on the study of human trace animal life cycle system inheritance. Its main purpose is to imitate the evolution mechanism of human biological system in nature by modern humans to establish the randomness of its development for global biological search and gene optimization, drawing on Darwin's theory of human biological system evolution and genetic evolution mechanism and Mendel's theory of biological genetic evolution. In essence, it is an efficient, parallel and global related search information technology and management method, which can help search users automatically acquire and automatically analyze the relevant basic knowledge of the relevant network search information space that has been accumulated during the whole process of related search, and users can automatically adjust and automatically control the whole process of related search in order to continuously seek the best user understanding.

[0068] Each chromosome in the genetic algorithm corresponds to a solution of the genetic algorithm. Generally, we use the formula of fitness function to judge and measure the efficiency and quality of the solution. Therefore, a mapping is constructed from a gene group to its solution fitness. We can regard the process of genetic algorithm as a process of seeking the best solution inside a multivariate function. There are countless "peaks" in this multidimensional surface, and these peaks correspond to local optimal solutions. Among them, there may be only one "peak" with the highest altitude, which is itself a global optimal understanding. One of the main tasks of the genetic algorithm is to try to climb to its highest peak, rather than stumbling down or falling into some smaller peaks. In addition, it is worth noting that the genetic algorithm does not have to find the "highest peak". If the evaluation of the best fitness of the problem is the smaller and the best, then the global optimal solution is the lowest value of the function. Correspondingly, what the genetic algorithm needs to find is the "deepest valley".

[0069] Related basic concepts of genetic algorithm:

[0070] Genotype: Genotype is a natural trait with a stronger dominant internal chromosome structure;

[0071] Phenotype: the external manifestation of the traits determined by chromosomes, or in other words, the external manifestation of an individual formed according to the genotype;

[0072] Population evolution: With the continuous evolution of modern humans, the population has gradually adapted to the natural living environment in which it lives, and its quality has been continuously developed and greatly improved. The evolution of the main biological populations of humans and human evolution are mainly based on certain biological population relationships.

[0073] Fitness: A measure of how well a species adapts to its living environment.

[0074] Selection: Selecting a number of individuals from a population with a certain probability. Generally, the selection process is a process of survival of the fittest based on fitness.

[0075] DNA re-replication: When an aging cell cannot undergo benign division, the DNA replication in the genetic material will transfer its genes to new aging cells through re-replication. The new aging cells will re-inherit the genetic genes of the remaining aging granulocytes from there.

[0076] Crossover: DNA is cut at the same position of two chromosomes, and the front and back strands cross and combine to form two new chromosomes. Also called gene recombination or hybridization;

[0077] Mutation: There is a certain probability of mutation throughout the entire process of genetic replication. Mutation will directly produce a new chromosome, which means it creates a new genetic characteristic for us, that is, a new trait.

[0078] Coding: The genetic information in DNA is encoded by arranging the longer points on the chain in a certain order. The genetic code can therefore be widely considered to be a mapping from phenotype to genotype.

[0079] Decoding: A genotype-to-phenotype decoding mapping.

[0080] Individual: An individual is a general term for an entity with distinct chromosome characteristics.

[0081] Population: A collection of individuals. The number of individuals in this collection is called the population size.

[0082] See also Figure 1 , Figure 1 is a flow chart of a shared bicycle scheduling method provided by an embodiment of the present invention, such as Figure 1 As shown, the following steps are included:

[0083] Step 101, determining a dispatch center in a target area;

[0084] In this step, taking the campus as an example, the campus is regionalized and divided into several areas, and a dispatch center is established for each area. Here, at least one area is taken as the target area, and a site is selected in the target area to determine the dispatch center.

[0085] Step 102, constructing a shared bicycle demand prediction model for each placement point in the target area;

[0086] In this step, since each placement point in the target area faces different user demands, it is necessary to build a shared bicycle demand prediction model for each placement point.

[0087] It should be pointed out that the time series forecasting method or the influencing factor analysis method can be used to determine the shared bicycle demand forecasting model.

[0088] Step 103: Determine the number of shared bicycles to be dispatched by the dispatch center to each placement point based on the shared bicycle demand prediction model.

[0089] That is to say, based on the shared bicycle demand prediction model, the number of shared bicycles to be dispatched by the dispatch center to each placement point can be determined, and the dispatch center in the target area can uniformly dispatch shared bicycles to each placement point to achieve on-demand delivery, improve the management level of shared bicycles, and improve the appearance of the city.

[0090] In this embodiment, by determining the dispatching center in the target area, a shared bicycle demand prediction model for each placement point in the target area is constructed, and based on the shared bicycle demand prediction model, the number of shared bicycles to be dispatched to each placement point by the dispatching center is determined. The number of shared bicycles at each placement point can be predicted, and the dispatching center dispatches shared bicycles to achieve on-demand delivery, avoid disorderly and uneven delivery, and reduce operating costs. At the same time, the dispatching center centrally manages shared bicycles and repairs damaged shared bicycles before deploying them to various placement points in the target area.

[0091] In a specific embodiment, step 101 includes:

[0092] Obtaining coordinate information of each placement point in the target area;

[0093] Constructing a dispatch center operation cost function according to the coordinate information of each placement point;

[0094] Based on the dispatch center operation cost function, obtaining target coordinate information corresponding to the minimum value of the dispatch center operation cost;

[0095] The dispatch center is determined according to the target coordinate information.

[0096] In order to select a dispatch center in the target area, that is, to solve the problem of the location of the dispatch center, it is first necessary to represent the placement points of each shared bicycle in the target area on the map using coordinates.

[0097] Among them, the location of the shared bicycle placement point in the target area can be determined based on the field survey results, and the coordinate information of the placement point can be obtained. Here, the coordinate information is the center coordinate.

[0098] For example, according to the field survey results, Figure 2 As shown, it can be found that there is one placement point each in Student Dormitory Building No. 11, between Student Dormitory Buildings No. 5 and 6, between Student Dormitory Buildings No. 13 and 14, between Student Dormitory Buildings No. 7 and 8, between Student Dormitory Building No. 12 and Office Building No. 10, and on the east side of the Sports and Cultural Center, for a total of six placement points.

[0099] Taking the corner of Student Dormitory Building No. 16 and Student Dormitory Building No. 5 as the coordinate origin, construct the rectangular coordinate system of the target area, and obtain the coordinates of the following 6 placement points:

[0100] (1.7, 1.1); (1.7, 3.4); (1.7, 4.8); (5.1, 5.2); (7.2, 3.1); (6.0, 1.2).

[0101] Since the location of the dispatching center is selected within the target area, the cost optimization problem is selected, that is, the site selection when solving the optimal cost value, which is transformed into a cost minimization problem. The model is constructed using operations research linear programming, and the following dispatching center operating cost function is constructed. Based on the dispatching center operating cost function, the target coordinate information with the minimum distance corresponding to the minimum cost is obtained.

[0102] Furthermore, according to the above fitness function, the target coordinate information corresponding to the minimum value is obtained, so as to determine the address of the dispatching center.

[0103] In one embodiment, constructing a dispatch center operation cost function according to the coordinate information of each placement point includes:

[0104] Constructing the operation cost function of the dispatch center according to the coordinate information of each placement point, the coordinate information of the candidate dispatch center, the unit distance cost and the personnel cost;

[0105] Among them, the candidate dispatch center is a location point in the target area other than the placement point.

[0106] It should be noted that the following dispatch center operation cost function is constructed:

[0107]

[0108] Where C represents the total operating cost of the dispatch center;

[0109] represents the objective function, which represents the cost incurred by the dispatch center due to the dispatch distance during actual use;

[0110] A represents the employment cost of hiring relevant dispatching personnel, which can be regarded as a constant.

[0111] The fitness function to be solved is the minimum value of the distance-related cost, which is a minimum value problem. Here, it is necessary to convert the objective function and the fitness function, and let the fitness function be the opposite of the objective function, as follows:

[0112]

[0113] In one embodiment, obtaining target coordinate information corresponding to the minimum value of the dispatch center operation cost based on the dispatch center operation cost function includes:

[0114] Acquire multiple first candidate dispatch centers, and encode the coordinate information of the multiple first candidate dispatch centers respectively to acquire the encoded coordinates of the multiple first candidate dispatch centers;

[0115] Calculating the fitness of the plurality of first candidate dispatching centers respectively according to the dispatching center operation cost function and the coordinate information of the plurality of first candidate dispatching centers;

[0116] Based on the encoded coordinates and fitness of the multiple first candidate dispatching centers, genetic iteration is performed to obtain a target dispatching center corresponding to the minimum dispatching center operation cost;

[0117] Obtain target coordinate information of the target dispatch center;

[0118] The first candidate dispatch center is a location point in the target area other than the placement point.

[0119] It should be noted that the genetic algorithm is used to obtain the target dispatch center corresponding to the minimum value of the dispatch center operation cost. The implementation process of the genetic algorithm is as follows:

[0120] Since the implementation process of the genetic algorithm is like the biological evolution process in nature. First, it is necessary to find and obtain a potential text solution to this problem, that is, to perform digital text encoding. Then, a random population is initialized with random numbers, and the individual data in this random population is the digital data encoding of these populations. Next, after appropriate gene decoding, a gene fitness evaluation function is used to perform the same gene fitness function evaluation on each gene. Use a function with selectable values ​​to select the optimal value of the specified parameter according to certain specific conditions. Start to mutate the offspring genome, and then let it produce an offspring. Among them, the operation steps of the genetic algorithm are defined as follows:

[0121] Step 1: Evaluate the fitness of the individual corresponding to each chromosome.

[0122] Step 2: Following the principle that the higher the degree of adaptability, the greater the probability of selection, two individuals are selected from a population as the father and mother.

[0123] Step 3: Extract two different male chromosomes from different ancestors and biological parents through the method, perform offspring exchange, and produce a new offspring.

[0124] Step 4: Mutate the chromosomes of the offspring.

[0125] Step 5: Repeat steps 2, 3 and 4 until a new population is generated.

[0126] Step 6: End the loop.

[0127] Several important concepts in the application of genetic algorithms:

[0128] (1) Binary encoding

[0129] Inspired by the profound influence and research of biologists on the basic structure of human chromosomes, we can assume that there are only two bases, "0" and "1", and use a simple dye chain to directly connect them in order. Because each basic unit must have a basic amount of information that can accurately display 1 bit for its performance, so as long as the length of the chromosome chain is long enough, it is enough to accurately outline and display all the characteristics of an individual. This is actually a binary-based dye coding method. The general structure of chromosomes is as follows: 010010011011011110111110

[0131] (3) Selection method in genetic algorithm - roulette

[0132] For example, there are 5 chromosomes, and the fitness scores corresponding to the 5 chromosomes are 5; 7; 10; 13 and 15 respectively.

[0133] So the cumulative total fitness is:

[0134]

[0135] So the probability of each individual being selected is:

[0136]

[0137]

[0138]

[0139]

[0140]

[0141] Spin a new roulette wheel in the air, and when it stops, the pointer may randomly point to a position or area represented by an individual, which means that the individual is directly selected.

[0142] (2) Genetic Recombination in Genetic Algorithms

[0143] Gene recombination and gene mutation are the fundamental factors that lead to the difference between offspring and their parents. For the genetic operations of these two genes, there are great differences in the signal processing methods between binary coding and floating-point coding. Among them, the gene exchange process of binary coding is very similar to the process of homologous chromosomes associating with each other in high school biology, that is, several codes at the same time or position are randomly exchanged to produce new individuals.

[0144] (3) Genetic Algorithm Mutation

[0145] The formation process of gene mutation: The so-called gene mutation mainly refers to a change in the gene structure at a specific gene site on a chromosome. Gene allele mutation can usually directly transform one allele into another allele, and often the mutation will directly lead to certain changes in the gene appearance and phenotype. In other words, the operation process of the genetic gene operation sequence encoded by binary genes is very similar to the various genetic gene operation processes in the genetic biology process. The "0" or "1" sequence on the gene string will be transformed into the completely opposite "1" or "0" with a certain high probability.

[0146] The roulette wheel method is used to obtain the first candidate dispatch center in the target area. Here, the process of obtaining the first candidate dispatch center can be regarded as the process of selecting a population in a genetic algorithm, and the specific steps are as follows:

[0147] Perform the best retention operation, sort the NP individuals in the contemporary population from large to small according to fitness, and record the individual with the largest fitness as the optimal individual elite, which is ranked first, and perform the hill climbing operation without participating in the crossover and mutation process.

[0148] Perform a roulette wheel selection operation and select NP-1 offspring.

[0149] First, calculate the probability of each chromosome being selected in the current population using the probability calculation formula:

[0150]

[0151] Then, the cumulative probability of each individual is calculated:

[0152]

[0153] Wherein, i=1, 2, ..., NP.

[0154] Then select individuals from the current group. If the selected individual is not the optimal individual, repeat the selection until the scale of the selected individuals meets the requirement of NP-1.

[0155] In one embodiment, performing genetic iteration based on the encoded coordinates and fitness of the plurality of first candidate dispatch centers to obtain a target dispatch center corresponding to the minimum dispatch center operation cost includes:

[0156] Selecting a plurality of second candidate dispatching centers from the plurality of first candidate dispatching centers according to the fitness corresponding to each first candidate dispatching center;

[0157] Processing the coded coordinates of the plurality of second candidate dispatch centers by gene recombination and / or gene mutation, selecting a plurality of third candidate dispatch centers from the plurality of second candidate dispatch centers, and obtaining the coded coordinates and fitness of each third candidate dispatch center;

[0158] Repeat the steps of obtaining the plurality of second candidate dispatch centers and the plurality of third candidate dispatch centers until a preset number of iterations is reached, obtain the encoded coordinates of the plurality of fourth candidate dispatch centers, and obtain the fitness of each fourth candidate dispatch center;

[0159] Among the multiple fourth candidate dispatching centers, a fourth candidate dispatching center with the greatest fitness is determined as the target dispatching center.

[0160] It should be noted that the second candidate dispatch center with the largest fitness is selected from the first candidate dispatch center, and then multiple second candidate dispatch centers are subjected to gene recombination and / or gene mutation. Gene recombination is similar to the biological association process, which is an array exchange between two binary codes, while gene mutation is a group of two-dimensional codes in which one or several "0"s are changed to "1", or "1" is changed to "0". Each time a gene recombination or gene mutation is carried out, the binary code will be changed, thereby changing the corresponding coordinate value. In other words, each time a gene recombination or gene mutation is carried out, a new individual will be generated, that is, the second candidate dispatch center will be obtained.

[0161] In the actual use and development of genetic algorithms, each time a new individual, that is, a new population, is iterated, it is necessary to calculate the parent data with a high degree of adaptability, so as to select a parent data from the most adaptable population for the next iteration. The specific operation method is as follows:

[0162] Find the elite individual with the best fitness function in the parent, use the elite individual to perform gene recombination and / or gene mutation to obtain a new elite individual, and calculate the fitness of elite-1. If the fitness function of elite-1 is greater than elite, then elite-1 is regarded as the best individual in the new generation population. Repeat the above operation until no better elite-1 individual can be found.

[0163] During the genetic iteration process, the number of iterations is pre-set as the termination condition of the iteration, and the iteration is continued until the number of iterations is reached to complete the genetic iteration.

[0164] Among them, an exemplary procedure of the genetic algorithm is as follows.

[0165]

[0166]

[0167]

[0168]

[0169]

[0170] In a specific embodiment, step 102 includes:

[0171] Obtain multiple potential factors that affect shared bicycle travel;

[0172] Obtaining travel data of the shared bicycles at each placement point;

[0173] Determining, based on the travel data, a plurality of target factors among the plurality of potential factors, and determining an influencing factor function of each target factor;

[0174] Fitting the influencing factor function of each target factor to construct the shared bicycle demand prediction model;

[0175] The target factor includes at least one of the following:

[0176] Age factors;

[0177] Gender factors;

[0178] Weather factors;

[0179] Travel distance factor.

[0180] It should be pointed out that, since the embodiment of the present invention takes the area in the campus as the target area as an example, an influencing factor analysis method can be used here to determine the shared bicycle demand prediction model.

[0181] Among them, potential factors are determined for the target area, and the potential factors may include at least one of gender factors, age factors, occupation factors, income factors, travel habit factors, road factors and weather factors.

[0182] In order to obtain travel data, field survey methods and questionnaire survey methods are used to collect shared bicycle travel data in the target area. First, the field survey results are as follows:

[0183] (1) The main distribution points of shared bicycles in the target area are dormitory buildings and some teaching buildings. The number of shared bicycles at each placement point is basically between 30 and 50.

[0184] (2) The shared bicycles in the dormitory area are generally borrowed by students before 9 a.m. and are then distributed to various areas on campus.

[0185] (3) Through on-site investigation of the cafeteria, it was found that the peak flow of shared bicycles in the cafeteria is from 11:30 to 12:00 noon, while the flow of shared bicycles decreases during dinner time.

[0186] Then, in order to understand the degree of dependence of students and teachers in the target area on shared bicycles and their satisfaction with the management of shared bicycles in the current target area, it is necessary to conduct a subjective survey of students and teachers. The most direct and simple method is to use a questionnaire survey. Among them, since a questionnaire survey tends to be subjective, it is necessary to reasonably set up counter-selected questions during the questionnaire design stage to improve the effectiveness of the questionnaire and the degree of concentration during the questionnaire survey.

[0187] Furthermore, based on the travel data, the influencing factors are analyzed to determine the target factors including: gender factor, age factor, weather factor and distance factor, and the influencing factor function of each target factor is determined.

[0188] Specifically, the influencing factor function of each target factor is as follows:

[0189] (1) Gender influence factor function:

[0190] Since the use of shared bicycles has a linear correlation with the target factor of gender, it is assumed here that for every 100 males, 100a shared bicycles are needed, a is the impact factor function corresponding to the male gender, and k×a is the impact factor function corresponding to the female gender, where k is the linear coefficient between male and female users.

[0191] According to the results of the questionnaire survey, among male users, high-frequency users of shared bicycles account for 22%, medium-frequency users account for 47%, and low-frequency users account for 31%. By taking the weighted average of the three frequencies, we can get the influencing factor function a.

[0192] Here, combined with the actual situation and comprehensive consideration of function fitting, the weight coefficient ratio corresponding to the frequency can be set to 6:3:1, thus obtaining:

[0193] The impact factor function corresponding to male gender:

[0194] a=0.22×0.6+0.47×0.3+0.31×0.1=0.304

[0195] According to the questionnaire survey and the usage report of the shared bicycle industry, the linear coefficient k = 0.716 can be obtained. Therefore, the impact factor function corresponding to female gender is:

[0196] k×a=0.716×0.304=0.218

[0197] (2) Age influence factor function:

[0198] In the research environment within the target area, the age difference is basically reflected in the difference between teachers and students. Therefore, the questionnaire is processed separately by age group, and it can be found that students' dependence on shared bicycles is much higher than that of teachers, and it conforms to a linear correlation. Here, it is assumed that for every 100 students, 100b shared bicycles are required, b is the student influence factor function, and n×b is the teacher influence factor function, where n is the linear coefficient between the two age groups of students and teachers.

[0199] According to the results of the questionnaire survey, among students, 29% are high-frequency users of shared bicycles, 52% are medium-frequency users, and 19% are low-frequency users. By taking the weighted average of the three frequencies, we can get the value of the influencing factor b.

[0200] In this study, combined with the actual situation and comprehensive consideration of function fitting, the weight coefficient ratio is set to 6:3:1, so the final result is:

[0201] Student impact factor function:

[0202] b=0.29×0.6+0.52×0.3+0.19×0.1=0.349

[0203] According to the questionnaire survey, the linear coefficient n=0.193 can be obtained.

[0204] Therefore, the teacher influence factor function is:

[0205] n×b=0.193×0.349=0.061

[0206] (3) Weather influence factor function: The use of shared bicycles in the target area is bound to be affected by the weather. Assume that in bad weather conditions such as rainy days, no one in the target area will use shared bicycles for travel. Therefore, the weather conditions are divided into hot days, sunny days, cloudy days, cold days and windy days. At the same time, it can be judged based on common sense. The number of people using shared bicycles on sunny days must be the largest, followed by hot days and cloudy days, and the least on cold days and windy days. Therefore, it is assumed that the impact of weather on the use of shared bicycles in the target area conforms to the following quadratic function:

[0207] y=px 2 +qx+c

[0208] Then establish the corresponding relationship between weather and function independent variable x as follows:

[0209] When the weather is sunny, let x=1; when the weather is hot, let x=2; when the weather is cloudy, let x=-1; when the weather is cold, let x=-2; when the weather is windy, let x=-2.5.

[0210] And when x=1, let y=1, y is the weather influencing factor. According to the question in the questionnaire "Suppose under certain circumstances on a sunny day, you would choose to ride a shared bicycle, and now replace the weather with the following conditions, what is the weather when you would still choose to ride a shared bicycle", we can get the following weather influencing factor function:

[0211] y=-0.238x 2 +0.114x+1.124

[0212] (4) Distance influence factor function:

[0213] In the target area, whether students and teachers use shared bicycles is related to the distance they travel. If the distance is far, it will be difficult for teachers and students to travel and they will prefer to ride bicycles. If the distance is short, most teachers and students will still choose to walk. By fitting the data collected from the questionnaire, it can be found that the influencing factors of distance on the use of shared bicycles in the target area basically conform to the functional characteristics of the exponential function.

[0214] Through data collection and analysis of the question "How far do you travel on campus before you choose to ride a shared bike?" in the questionnaire, the four options in the questionnaire, "arrive within five minutes on foot", "arrive within five to ten minutes on foot", "arrive within ten to twenty minutes on foot" and "arrive within twenty minutes on foot", are associated with the variable t:

[0215] When the travel distance is "arrived within five minutes' walk", let t=1; when the travel distance is "arrived within five to ten minutes' walk", let t=2; when the travel distance is "arrived within ten to twenty minutes' walk", let t=3; when the travel distance is "arrived within more than twenty minutes' walk", let t=5.

[0216] Let the distance influence factor be T. Through the questionnaire data analysis, we found that when t=5, the distance influence factor function is T=0.961, when t=1, the distance influence factor function is T=0.107, and when t=2, the distance influence factor function is T=0.403.

[0217] Therefore, we can get the distance impact factor function as an exponential function relationship:

[0218] T=0.0705e 0.539t

[0219] That is to say, by solving the influencing factors according to the above four influencing factors respectively, four cumulative functions, namely the influencing factor functions, can be obtained. When predicting the demand for the site now, the four influencing factor functions need to be summarized and fitted to obtain the shared bicycle demand prediction model.

[0220] It is not difficult to infer that the shared bicycle demand forecasting model is a composite function of one or some combination of the four influencing factor functions. Therefore, it is necessary to consider which of the four influencing factor functions should be placed as the basis function in the first layer of the composite function, and so on to finally obtain the shared bicycle demand forecasting model.

[0221] Since the impact of gender and age on shared bicycle travel within the target area has a linear characteristic, and the influencing factor functions of the two are also the most basic linear functions, these two influencing factor functions can be used as the bottom-level functions of the composite function for functional compounding. In addition, these two influencing factors are factors possessed by shared bicycle users themselves, and it is also very reasonable to regard them as the most basic influencing factors.

[0222] As external influencing factors, distance and weather have independent impacts on teachers and students' use of shared bicycles and will not affect each other. Therefore, both weather influencing factors and distance influencing factors can be called environmental influencing factors.

[0223] As for the gender and age of the user's own influencing factors, since these two factors affect the use of shared bicycles at the same time, they cannot be directly multiplied. It is necessary to consider constructing new factors of the user's own influencing factors for demand forecasting.

[0224] Where a is the gender influence factor function, which is defined as follows:

[0225] Male gender influence factor function a1=0.304;

[0226] Female gender influence factor function a2 = 0.218;

[0227] Let b be the age impact factor function, which is defined as follows:

[0228] Student impact factor function b1 = 0.349;

[0229] Teacher influence factor function b2=0.061.

[0230] To construct a new user's own influencing factor function for demand forecasting, it is necessary to combine the gender influencing factor function and the age influencing factor function. Here, we consider using the geometric mean to construct a new influencing factor function, and finally we can get:

[0231] The impact factor function of male students is f1 = 0.326;

[0232] The influence factor function of male teachers is f2 = 0.136;

[0233] The impact factor function of female students f3 = 0.275;

[0234] The impact factor function of female teachers is f4=0.115.

[0235] Since the influence factor function of the subject is fixed and the travel distance is almost unchanged in the short term, the travel distance of students and teachers will only change within a fixed range in one semester. Therefore, the geometric mean of the distance influence factor function and the user's own influence factor function can be used as an inherent influence factor function, and the weather is multiplied with the former as a constant influence factor function:

[0236] Finally, the shared bicycle demand forecasting model is:

[0237]

[0238] In one embodiment, the step of obtaining the travel data of the shared bicycles at each placement point includes:

[0239] Acquire the travel data of the shared bicycles at each placement point whose reliability coefficient is within a preset range.

[0240] It should be noted that reliability analysis is a method used to measure samples, that is, to measure whether the subjects answer questions truthfully. Use Cronbach's reliability coefficient as a reliability criterion. If the Cronbach's reliability coefficient is above 0.8, it can generally be directly confirmed that the reliability of the reliability scale or test is very good; if the Cronbach's reliability coefficient is above 0.7, it is generally believed that the reliability is widely accepted; if the Cronbach's reliability coefficient is within the international reliability coefficient range of 0.6 to 0.7, it is generally believed that the reliability of the reliability scale or test has certain international academic research value, but it may need to be revised more. If the Cronbach's reliability coefficient is lower than 0.6, the test needs to be re-set to ensure the scientific nature of the test.

[0241] Using SPSSAU (a data science analysis platform system) for analysis, it was found that the Cronbach reliability coefficient of the questionnaire was 0.941. Therefore, the reliability of the questionnaire is very good, and the collected results can fully support the subsequent analysis.

[0242] In addition, the questionnaire also needs to be analyzed for validity. Validity refers to the effectiveness of the survey. Validity can be used to measure the correctness and reliability of the questionnaire items (including quantitative data) in design. It can be verified through methods such as factor analysis (exploratory factor analysis). Due to the corresponding relationship between the preset variables and items, after factor analysis, the relationship between the factor (that is, a variable, collectively referred to as a factor when using factor analysis) and other items corresponds to each other. If the expectations of the two are basically the same, it can be said that they have good working validity.

[0243] Using SPSSAU for analysis, we can get the validity analysis KMO value of this questionnaire survey to be 0.778. When using the KMO value as the most direct validity judgment standard, a KMO value of 0.6 is in line with the standard, and 0.7 is relatively suitable. Therefore, the validity of this questionnaire fully meets the standard and can be used for subsequent analysis.

[0244] In one embodiment, determining multiple target factors from the multiple potential factors based on the travel data includes:

[0245] Obtaining a correlation between each potential factor and the travel data;

[0246] The multiple target factors are determined according to the correlation.

[0247] It should be noted that this questionnaire survey analyzes the inherent laws of various potential factors and the use of shared bicycles in the target area. However, before the research results are determined, it cannot be said with certainty that there is a strong correlation between the two. Therefore, after collecting enough answers, it is necessary to do a simple correlation analysis to determine the correlation between various potential factors and the use of shared bicycles, that is, travel data. Here, cross analysis and correlation analysis can be used to obtain the correlation between each potential factor and the travel data.

[0248] Among them, the cross analysis method refers to using the existing collected questionnaires, selecting a few questions in the personality test as the dependent variable, and then selecting a question in the frequency of use as the independent variable, analyzing the two, and then making relevant charts to visually observe whether there is an objective correlation between the two. In order to make the cross analysis more convenient, the reverse selection question is not considered as the dependent variable here, so the question "What is your gender" is selected as the dependent variable, and the question "How many times do you use shared bicycles per month" is selected as the dependent variable to analyze the correlation between the two.

[0249] The cross-analysis results are shown in Table 1 below.

[0250] Number of shared bikes used per month Male gender female 5 times or less 2(4.17％) 24(50％) 5 to 10 times 20(18.18％) 54(49.09％) 10 to 20 times 19(27.94％) 30(44.12％) More than 20 times 1(11.11％) 4(44.44％)

[0251] Table 1

[0252] In addition, from Table 1 above, we can also know that there are obvious differences in the use of shared bicycles between male and female subjects in the target area, and there may be a proportional relationship in the number of male and female subjects. Therefore, after doubling the number of female subjects, it can be obtained that the score of male subjects in the monthly use of shared bicycles is 1.4 times that of female subjects. Therefore, it can be determined that gender factors have a certain correlation with the travel data of shared bicycles.

[0253] The correlation analysis method refers to selecting one question from each of the questionnaire's influencing factors and the questionnaire's use of shared bicycles, and using the principles of mathematical statistics to calculate the correlation parameters between the two, in order to prove that the potential factors have a definite correlation with the travel data of shared bicycles on campus. The following uses two different correlation coefficients to judge the correlation between the two questionnaires:

[0254] Here, we select the following questions for analysis: "The distance you travel each time (to class, eat, or pick up a package, etc.)" and "The number of times you use a shared bike each month."

[0255] Table 2 below shows the correlation analysis using Kendalls correlation coefficient.

[0256] Questions in the survey The distance you travel each time (to attend classes, eat, or pick up a package, etc.) How many times do you use shared bikes per month? 0.487

[0257] Table 2

[0258] From Table 2 above, we can see that the correlation analysis method is used to analyze the correlation between "the distance you travel each time (to attend classes, eat, or pick up parcels, etc.)" and "the number of times you use shared bicycles each month", and the Kendall correlation coefficient is used to indicate the strength of the correlation. The specific analysis shows that the correlation value between "the distance you travel each time (to attend classes, eat, or pick up parcels, etc.)" and "the number of times you use shared bicycles each month" is 0.487, and it shows a significance level of 0.01.

[0259] Therefore, it can be judged that there is a significant positive correlation between "the distance you travel each time (to class, eat, or pick up a package, etc.)" and "the number of times you use a shared bicycle each month."

[0260] The Spearman correlation coefficient was used for correlation analysis as shown in Table 3 below.

[0261] Questions in the survey The distance you travel each time (to attend classes, eat, or pick up a package, etc.) How many times do you use shared bikes per month? 0.574

[0262] Table 3

[0263] As shown in Table 3 above, the correlation analysis method is used to analyze the correlation between "the distance you travel each time (to attend classes, eat, or pick up parcels, etc.)" and "the number of times you use shared bicycles each month", and the Spearman correlation coefficient is used to indicate the strength of the correlation. The specific analysis shows that the correlation coefficient between "the distance you travel each time (to attend classes, eat, or pick up parcels, etc.)" and "the number of times you use shared bicycles each month" is 0.574, and it is significant at the 0.01 level.

[0264] Therefore, it can be determined that there is a significant positive correlation between "the distance you travel each time (to attend classes, eat, or pick up a package, etc.)" and "the number of times you use a shared bike each month."

[0265] In a specific embodiment, step 103 includes:

[0266] Based on the shared bicycle demand prediction model, obtaining the demand for shared bicycles at each placement point;

[0267] Based on the demand, the number of shared bicycles to be dispatched by the dispatch center to each placement point is determined.

[0268] It should be noted that after the shared bicycle demand forecasting model is built, it needs to be applied to an example to verify its effectiveness. Here, a student dormitory in the target area with 400 male students is taken as an example. To ensure that the demand is met, sunny weather should be selected as the research object, so the weather impact factor function y = 1. The walking distance from the student dormitory to the library, information teaching building and central teaching building is about 10 to 15 minutes, so t = 3 in the distance impact factor function here. Substituting it into the distance impact factor function formula, the distance impact factor is calculated:

[0269] T=0.0705e 0.539t =0.355

[0270] According to the shared bicycle demand prediction model, the impact factor F=0.340 is obtained.

[0271] It can be known that the demand for shared bicycles in the student dormitory is 400×0.340=136.

[0272] Furthermore, after learning that the demand for shared bicycles in the student dormitory on sunny days is 136, it does not mean that the dispatch center needs to dispatch the shared bicycles of this demand to the placement point of the student dormitory. The dispatch center needs to dispatch the number of shared bicycles to the placement point based on this demand and the reuse rate and operating cost of shared bicycles. In other words, the shared bicycle demand prediction model is to provide the dispatch center with demand data so that the dispatch center can consider how to dispatch and how to operate.

[0273] In summary, the embodiments of the present invention determine the dispatching center in the target area through genetic iteration, and construct a shared bicycle demand prediction model for the placement points, and determine the number of shared bicycles dispatched by the dispatching center to each placement point, thereby solving the problem of uneven placement due to disorderly placement and reducing operating costs. In addition, the dispatching center has unified management of shared bicycles, and damaged vehicles can be repaired before being dispatched to avoid affecting their use.

[0274] like Figure 3 As shown, an embodiment of the present invention further provides a shared bicycle dispatching device, comprising:

[0275] A first determination module 301 is used to determine a dispatch center in a target area;

[0276] The first construction module 302 is used to construct a shared bicycle demand prediction model for each placement point in the target area;

[0277] The second determination module 303 is used to determine the number of shared bicycles to be dispatched by the dispatch center to each placement point according to the shared bicycle demand prediction model.

[0278] Optionally, the first determining module 301 includes:

[0279] A first acquisition unit, used to acquire coordinate information of each placement point in the target area;

[0280] A first construction unit is used to construct an operation cost function of the dispatching center according to the coordinate information of each placement point;

[0281] A first obtaining unit, configured to obtain target coordinate information corresponding to a minimum value of the operation cost of the dispatching center based on the operation cost function of the dispatching center;

[0282] The first determining unit is used to determine the dispatching center according to the target coordinate information.

[0283] Optionally, the first construction unit is specifically used to:

[0284] Constructing the operation cost function of the dispatch center according to the coordinate information of each placement point, the coordinate information of the candidate dispatch center, the unit distance cost and the personnel cost;

[0285] Among them, the candidate dispatch center is a location point in the target area other than the placement point.

[0286] Optionally, the first obtaining unit includes:

[0287] A first acquisition subunit is used to acquire a plurality of first candidate dispatch centers, and encode the coordinate information of the plurality of first candidate dispatch centers respectively to acquire the encoded coordinates of the plurality of first candidate dispatch centers;

[0288] A calculation subunit, configured to calculate the fitness of the plurality of first candidate dispatching centers respectively according to the dispatching center operation cost function and the coordinate information of the plurality of first candidate dispatching centers;

[0289] An obtaining subunit is used to perform genetic iteration based on the encoding coordinates and fitness of the plurality of first candidate dispatching centers to obtain a target dispatching center corresponding to the minimum dispatching center operation cost;

[0290] The second acquisition subunit is used to acquire target coordinate information of the target dispatch center;

[0291] The first candidate dispatch center is a location point in the target area other than the placement point.

[0292] Optionally, the obtaining subunit is specifically used for:

[0293] Selecting a plurality of second candidate dispatching centers from the plurality of first candidate dispatching centers according to the fitness corresponding to each first candidate dispatching center;

[0294] Processing the coded coordinates of the plurality of second candidate dispatch centers by gene recombination and / or gene mutation, selecting a plurality of third candidate dispatch centers from the plurality of second candidate dispatch centers, and obtaining the coded coordinates and fitness of each third candidate dispatch center;

[0295] Repeat the steps of obtaining the plurality of second candidate dispatch centers and the plurality of third candidate dispatch centers until a preset number of iterations is reached, obtain the encoded coordinates of the plurality of fourth candidate dispatch centers, and obtain the fitness of each fourth candidate dispatch center;

[0296] Among the multiple fourth candidate dispatching centers, a fourth candidate dispatching center with the greatest fitness is determined as the target dispatching center.

[0297] Optionally, the first building module 302 includes:

[0298] The second acquisition unit is used to acquire multiple potential factors that affect shared bicycle travel;

[0299] A third acquisition unit is used to acquire travel data of the shared bicycles at each placement point;

[0300] A second determining unit, configured to determine, based on the travel data, a plurality of target factors among the plurality of potential factors, and determine an influencing factor function of each target factor;

[0301] A second construction unit is used to fit the influencing factor function of each target factor to construct the shared bicycle demand prediction model;

[0302] The target factor includes at least one of the following:

[0303] Age factors;

[0304] Gender factors;

[0305] Weather factors;

[0306] Travel distance factor.

[0307] Optionally, the third acquiring unit is specifically configured to:

[0308] Acquire the travel data of the shared bicycles at each placement point whose reliability coefficient is within a preset range.

[0309] Optionally, the second determining unit is specifically configured to:

[0310] Obtaining a correlation between each potential factor and the travel data;

[0311] The multiple target factors are determined according to the correlation.

[0312] Optionally, the second determining module 303 is specifically configured to:

[0313] Based on the shared bicycle demand prediction model, obtaining the demand for shared bicycles at each placement point;

[0314] Based on the demand, the number of shared bicycles to be dispatched by the dispatch center to each placement point is determined.

[0315] The device provided in the embodiment of the present invention can execute the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0316] like Figure 4 As shown, the shared bicycle dispatching device of an embodiment of the present invention includes: a processor 400; and a memory 420 connected to the processor 400 through a bus interface, wherein the memory 420 is used to store programs and data used by the processor 400 when performing operations, and the processor 400 calls and executes the programs and data stored in the memory 420.

[0317] The processor 400 is used to read the program in the memory 420 and execute the following process:

[0318] Identify dispatch centers in target areas;

[0319] Constructing a shared bicycle demand prediction model for each placement point in the target area;

[0320] The number of shared bicycles to be dispatched by the dispatch center to each placement point is determined according to the shared bicycle demand prediction model.

[0321] The transceiver 410 is configured to receive and send data under the control of the processor 400 .

[0322] Among them, Figure 3 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by processor 400 and various circuits of memory represented by memory 420 are linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The transceiver 410 may be a plurality of components, namely, a transmitter and a transceiver, providing a unit for communicating with various other devices on a transmission medium. For different user devices, the user interface 430 may also be an interface capable of externally and internally connecting required devices, and the connected devices include but are not limited to a keypad, a display, a speaker, a microphone, a joystick, and the like.

[0323] The processor 400 is responsible for managing the bus architecture and general processing, and the memory 420 can store data used by the processor 400 when performing operations.

[0324] Optionally, the processor 400 is further configured to read the computer program and execute the following steps:

[0325] Obtaining coordinate information of each placement point in the target area;

[0326] Constructing a dispatch center operation cost function according to the coordinate information of each placement point;

[0327] Based on the dispatch center operation cost function, obtaining target coordinate information corresponding to the minimum value of the dispatch center operation cost;

[0328] The dispatch center is determined according to the target coordinate information.

[0329] Optionally, the processor 400 is further configured to read the computer program and execute the following steps:

[0330] Constructing the operation cost function of the dispatch center according to the coordinate information of each placement point, the coordinate information of the candidate dispatch center, the unit distance cost and the personnel cost;

[0331] Among them, the candidate dispatch center is a location point in the target area other than the placement point.

[0332] Optionally, the processor 400 is further configured to read the computer program and execute the following steps:

[0333] Acquire multiple first candidate dispatch centers, and encode the coordinate information of the multiple first candidate dispatch centers respectively to acquire the encoded coordinates of the multiple first candidate dispatch centers;

[0334] Calculating the fitness of the plurality of first candidate dispatching centers respectively according to the dispatching center operation cost function and the coordinate information of the plurality of first candidate dispatching centers;

[0335] Based on the encoded coordinates and fitness of the multiple first candidate dispatching centers, genetic iteration is performed to obtain a target dispatching center corresponding to the minimum dispatching center operation cost;

[0336] Obtain target coordinate information of the target dispatch center;

[0337] The first candidate dispatch center is a location point in the target area other than the placement point.

[0338] Optionally, the processor 400 is further configured to read the computer program and execute the following steps:

[0339] Selecting a plurality of second candidate dispatching centers from the plurality of first candidate dispatching centers according to the fitness corresponding to each first candidate dispatching center;

[0340] Processing the coded coordinates of the plurality of second candidate dispatch centers by gene recombination and / or gene mutation, selecting a plurality of third candidate dispatch centers from the plurality of second candidate dispatch centers, and obtaining the coded coordinates and fitness of each third candidate dispatch center;

[0341] Repeat the steps of obtaining the plurality of second candidate dispatch centers and the plurality of third candidate dispatch centers until a preset number of iterations is reached, obtain the encoded coordinates of the plurality of fourth candidate dispatch centers, and obtain the fitness of each fourth candidate dispatch center;

[0342] Among the multiple fourth candidate dispatching centers, a fourth candidate dispatching center with the greatest fitness is determined as the target dispatching center.

[0343] Optionally, the processor 400 is further configured to read the computer program and execute the following steps:

[0344] Obtain multiple potential factors that affect shared bicycle travel;

[0345] Obtaining travel data of the shared bicycles at each placement point;

[0346] Determining, based on the travel data, a plurality of target factors among the plurality of potential factors, and determining an influencing factor function of each target factor;

[0347] Fitting the influencing factor function of each target factor to construct the shared bicycle demand prediction model;

[0348] The target factor includes at least one of the following:

[0349] Age factors;

[0350] Gender factors;

[0351] Weather factors;

[0352] Travel distance factor.

[0353] Optionally, the processor 400 is further configured to read the computer program and execute the following steps:

[0354] Acquire the travel data of the shared bicycles at each placement point whose reliability coefficient is within a preset range.

[0355] Optionally, the processor 400 is further configured to read the computer program and execute the following steps:

[0356] Obtaining a correlation between each potential factor and the travel data;

[0357] The multiple target factors are determined according to the correlation.

[0358] Optionally, the processor 400 is further configured to read the computer program and execute the following steps:

[0359] Based on the shared bicycle demand prediction model, obtaining the demand for shared bicycles at each placement point;

[0360] Based on the demand, the number of shared bicycles to be dispatched by the dispatch center to each placement point is determined.

[0361] The device provided in the embodiment of the present invention can execute the above method embodiment, and its implementation principle and technical effect are similar, which will not be repeated in this embodiment.

[0362] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or may be accomplished by instructing relevant hardware through a computer program, wherein the computer program includes instructions for executing part or all of the steps of the above method; and the computer program may be stored in a readable storage medium, and the storage medium may be any form of storage medium.

[0363] In addition, a specific embodiment of the present invention further provides a readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the above-mentioned shared bicycle scheduling method are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0364] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0365] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may be physically included separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0366] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform some steps of the sending and receiving method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0367] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A shared bicycle dispatching method, characterized in that: include: Identify dispatch centers in target areas; Using a time series prediction method or an influencing factor analysis method, a shared bicycle demand prediction model is constructed for each placement point in the target area; Determining the number of shared bicycles to be dispatched by the dispatch center to each placement point according to the shared bicycle demand prediction model; Wherein, determining the dispatch center in the target area includes: Obtaining coordinate information of each placement point in the target area; Constructing a dispatch center operation cost function according to the coordinate information of each placement point; Based on the dispatch center operation cost function, obtaining target coordinate information corresponding to the minimum value of the dispatch center operation cost; Determining the dispatch center according to the target coordinate information; The obtaining, based on the dispatch center operation cost function, target coordinate information corresponding to the minimum value of the dispatch center operation cost includes: Acquire multiple first candidate dispatch centers, and encode the coordinate information of the multiple first candidate dispatch centers respectively to acquire the encoded coordinates of the multiple first candidate dispatch centers; Calculating the fitness of the plurality of first candidate dispatching centers respectively according to the dispatching center operation cost function and the coordinate information of the plurality of first candidate dispatching centers; Based on the encoded coordinates and fitness of the multiple first candidate dispatching centers, genetic iteration is performed to obtain a target dispatching center corresponding to the minimum dispatching center operation cost; Obtain target coordinate information of the target dispatch center; The first candidate dispatch center is a location point in the target area other than the placement point.

2. The method according to claim 1, characterized in that The step of constructing a dispatch center operation cost function according to the coordinate information of each placement point includes: Constructing the operation cost function of the dispatch center according to the coordinate information of each placement point, the coordinate information of the candidate dispatch center, the unit distance cost and the personnel cost; Among them, the candidate dispatch center is a location point in the target area other than the placement point.

3. The method according to claim 1, characterized in that The step of performing genetic iteration based on the encoded coordinates and fitness of the plurality of first candidate dispatch centers to obtain a target dispatch center corresponding to the minimum dispatch center operation cost includes: Selecting a plurality of second candidate dispatching centers from the plurality of first candidate dispatching centers according to the fitness corresponding to each first candidate dispatching center; Processing the coded coordinates of the plurality of second candidate dispatch centers by gene recombination and / or gene mutation, selecting a plurality of third candidate dispatch centers from the plurality of second candidate dispatch centers, and obtaining the coded coordinates and fitness of each third candidate dispatch center; Repeat the steps of obtaining the plurality of second candidate dispatch centers and the plurality of third candidate dispatch centers until a preset number of iterations is reached, obtain the encoded coordinates of the plurality of fourth candidate dispatch centers, and obtain the fitness of each fourth candidate dispatch center; Among the multiple fourth candidate dispatching centers, a fourth candidate dispatching center with the greatest fitness is determined as the target dispatching center.

4. The method according to claim 1, characterized in that: The step of constructing a shared bicycle demand prediction model for each placement point in the target area includes: Obtain multiple potential factors that affect shared bicycle travel; Obtaining travel data of the shared bicycles at each placement point; Determining, based on the travel data, a plurality of target factors among the plurality of potential factors, and determining an influencing factor function of each target factor; Fitting the influencing factor function of each target factor to construct the shared bicycle demand prediction model; The target factor includes at least one of the following: Age factors; Gender factors; Weather factors; Travel distance factor.

5. The method according to claim 4, characterized in that The obtaining of the travel data of the shared bicycles at each placement point includes: Acquire the travel data of the shared bicycles at each placement point whose reliability coefficient is within a preset range.

6. The method according to claim 4, characterized in that Determining multiple target factors from the multiple potential factors according to the travel data includes: Obtaining a correlation between each potential factor and the travel data; The multiple target factors are determined according to the correlation.

7. The method according to claim 1, characterized in that The step of determining the number of shared bicycles to be dispatched by the dispatch center to each placement point according to the shared bicycle demand prediction model includes: Based on the shared bicycle demand prediction model, obtaining the demand for shared bicycles at each placement point; Based on the demand, the number of shared bicycles to be dispatched by the dispatch center to each placement point is determined.

8. A shared bicycle dispatching device, characterized in that: include: A first determination module is used to determine a dispatch center in a target area; The first construction module is used to construct a shared bicycle demand prediction model for each placement point in the target area by using a time series prediction method or an influencing factor analysis method; A second determination module is used to determine the number of shared bicycles to be dispatched by the dispatch center to each placement point according to the shared bicycle demand prediction model; Wherein, the first determining module includes: A first acquisition unit, used to acquire coordinate information of each placement point in the target area; A first construction unit is used to construct an operation cost function of the dispatching center according to the coordinate information of each placement point; A first obtaining unit, configured to obtain target coordinate information corresponding to a minimum value of the operation cost of the dispatching center based on the operation cost function of the dispatching center; A first determining unit, configured to determine the dispatching center according to the target coordinate information; The first obtaining unit comprises: A first acquisition subunit is used to acquire a plurality of first candidate dispatch centers, and encode the coordinate information of the plurality of first candidate dispatch centers respectively to acquire the encoded coordinates of the plurality of first candidate dispatch centers; A calculation subunit, configured to calculate the fitness of the plurality of first candidate dispatching centers respectively according to the dispatching center operation cost function and the coordinate information of the plurality of first candidate dispatching centers; An obtaining subunit is used to perform genetic iteration based on the encoding coordinates and fitness of the plurality of first candidate dispatching centers to obtain a target dispatching center corresponding to the minimum dispatching center operation cost; The second acquisition subunit is used to acquire target coordinate information of the target dispatch center; The first candidate dispatch center is a location point in the target area other than the placement point.

9. A shared bicycle dispatching device, comprising: A transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor; wherein the processor is used to read the program in the memory to implement the steps in the shared bicycle dispatching method as described in any one of claims 1 to 7.

10. A readable storage medium, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps in the shared bicycle dispatching method as described in any one of claims 1 to 7.

Images