CN114841610A

CN114841610A - Shared bicycle scheduling method, device, equipment and readable storage medium

Info

Publication number: CN114841610A
Application number: CN202210590148.XA
Authority: CN
Inventors: 姚丽亚; 孙立山; 陈刚; 徐其泰
Original assignee: Beijing Institute of Technology BIT
Current assignee: Beijing Institute of Technology BIT
Priority date: 2022-05-26
Filing date: 2022-05-26
Publication date: 2022-08-02

Abstract

The invention provides a shared bicycle scheduling method, a device, equipment and a readable storage medium, and relates to the technical field of shared bicycles. The method comprises the following steps: determining a dispatching center in a target area; constructing a shared single-vehicle demand prediction model of each placement point in the target area; and determining the number of the shared bicycles dispatched to each placing point by the dispatching center according to the shared bicycle demand prediction model. The scheme of the invention can predict the required quantity of the shared single vehicles at each placing point, and the dispatching is carried out by the dispatching center, so that unbalanced throwing is avoided.

Description

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

Technical Field

The invention belongs to the technical field of shared bicycles, and particularly relates to a shared bicycle scheduling method and device and electronic equipment.

Background

At present, a pile-free shared bicycle is used as a green travel mode, the problem of 'the last kilometer' is solved, convenience is brought to travel of users, urban congestion is relieved to a certain extent, however, the well-spraying development of the shared bicycle also brings certain negative effects, and the following problems need to be solved urgently in the operation and management processes:

(1) unordered delivery of large sharing bicycle manufacturers

In the process of rapid popularization of the shared bicycle, various manufacturers pay attention to the huge economic benefits brought by the industry, so that the unmanaged shared bicycle enterprises increase the investment amount to occupy a larger share of the market. In this large scale, after the unordered throwing, the shared bicycle does not play a role in relieving traffic pressure, but rather aggravates the congestion of the urban road. At present, shared bicycles are not normally parked, so that the bicycles are parked anywhere on motor lanes, sidewalks, green belts and the like on the street, and the public order is seriously influenced. Particularly, the shared bicycles are parked around the subway station in a disorderly manner, so that the periphery of the subway station with large traffic is more congested, and much inconvenience is brought to normal traveling.

(2) Imbalance of shared bicycle borrowing and returning demands

The shared bicycle has different time characteristics and space characteristics from other transportation modes, and due to the unique space-time characteristics, the shared bicycle has the problem of unbalanced borrowing and returning requirements. From the time perspective, the demand of borrowing and returning the bicycles at each station of the shared bicycle is constantly changed in one day, the use frequency of the bicycles is high during the peak period, the amount of the borrowed and returned bicycles is large, and other periods are relatively low; from the perspective of space, the traveling intensity of different areas in a city is different, so that the demand of borrowing and returning the shared bicycle at stations with different land properties in the same time period is different. The imbalance of the demand of the shared bicycle causes the hot spot areas in the peak period to have no vehicles available, which is one of the most main problems faced by the shared bicycle industry.

(3) Shared bicycle damage and maintenance problems

The gradual loss of the shared bicycle after the shared bicycle is put on the market is inevitable, but the phenomena of malicious damage such as handlebar, saddle, tire breakage, two-dimensional code correction and the like are frequently lost, so that the fault rate of the shared bicycle is very high, huge economic loss is brought to a shared bicycle enterprise, and the satisfaction degree of users is reduced.

Disclosure of Invention

The embodiment of the invention aims to provide a shared bicycle scheduling method, a shared bicycle scheduling device, shared bicycle scheduling equipment and a readable storage medium, so that the problem of unbalanced delivery caused by improper shared bicycle scheduling in the prior art is solved.

In a first aspect, an embodiment of the present invention provides a shared bicycle scheduling method, including:

determining a dispatching center in a target area;

constructing a shared bicycle demand prediction model of each placing point in the target area;

and determining the number of the shared bicycles dispatched to each placing point by the dispatching center according to the shared bicycle demand prediction model.

Optionally, the determining a scheduling center in the target area includes:

acquiring coordinate information of each placing point in the target area;

constructing a scheduling center operation cost function according to the coordinate information of each placing point;

obtaining target coordinate information corresponding to the minimum value of the operation cost of the dispatching center based on the operation cost function of the dispatching center;

and determining the dispatching center according to the target coordinate information.

Optionally, the constructing a scheduling center operation cost function according to the coordinate information of each placement point includes:

constructing an operation cost function of the dispatching center according to the coordinate information of each placing point, the coordinate information of the candidate dispatching center, the unit distance cost and the personnel cost;

wherein the candidate dispatching center is a position point in the target area except the placing point.

Optionally, the obtaining target coordinate information corresponding to the minimum value of the scheduling center operation cost based on the scheduling center operation cost function includes:

acquiring a plurality of first candidate dispatching centers, and respectively coding coordinate information of the plurality of first candidate dispatching centers to acquire coded coordinates of the plurality of first candidate dispatching 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;

performing genetic iteration based on the coding coordinates and fitness of the plurality of first candidate scheduling centers to obtain a target scheduling center corresponding to the minimum value of the operation cost of the scheduling center;

acquiring target coordinate information of a target scheduling center;

wherein the first candidate dispatching center is a position point in the target area except the placing point.

Optionally, the performing genetic iteration based on the coding coordinates and the fitness of the plurality of first candidate scheduling centers to obtain a target scheduling center corresponding to the minimum value of the operation cost of the scheduling center includes:

selecting a plurality of second candidate dispatching centers from the plurality of first candidate dispatching centers according to the corresponding fitness of each first candidate dispatching center;

processing the coding coordinates of the plurality of second candidate dispatching centers by adopting a gene recombination and/or gene mutation mode, selecting a plurality of third candidate dispatching centers from the plurality of second candidate dispatching centers, and acquiring the coding coordinates and the fitness of each third candidate dispatching center;

repeating the steps of obtaining the plurality of second candidate dispatching centers and the plurality of third candidate dispatching centers until the preset iteration times is reached, obtaining the coding coordinates of the plurality of fourth candidate dispatching centers, and obtaining the fitness of each fourth candidate dispatching center;

and determining the fourth candidate dispatching center with the maximum fitness as the target dispatching center in the plurality of fourth candidate dispatching centers.

Optionally, the constructing a shared bicycle demand prediction model for each place in the target area includes:

acquiring a plurality of potential factors influencing the traveling of the shared bicycle;

acquiring the travel data of the shared bicycle at each placing point;

determining a plurality of target factors in the plurality of potential factors according to the travel data, and determining an influence factor function of each target factor;

fitting the influence factor function of each target factor to construct the shared bicycle demand prediction model;

wherein the objective factors include at least one of:

age factors;

a sex factor;

a weather factor;

travel distance factors.

Optionally, the obtaining travel data of the shared bicycle at each place includes:

and obtaining the travel data of the shared bicycle of each placing point with the reliability coefficient within a preset range.

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

acquiring a correlation between each potential factor and the travel data;

and determining the plurality of target factors according to the correlation.

Optionally, the scheduling, according to the shared-bicycle demand prediction model, a shared bicycle to a plurality of placement points in the target area by the scheduling center includes:

obtaining the demand of the shared bicycle of each placing point based on the shared bicycle demand prediction model;

determining, based on the demand, a number of shared vehicles to be dispatched by the dispatch center to the each of the placement points.

A second invention, an embodiment of the present invention further provides a shared bicycle scheduling apparatus, including:

the first determining module is used for determining a dispatching center in a target area;

the first construction module is used for constructing a shared bicycle demand prediction model of each placing point in the target area;

and the second determining module is used for determining the number of the shared bicycles dispatched to each placing point by the dispatching center according to the shared bicycle demand forecasting model.

In a third aspect, an embodiment of the present invention further provides a shared bicycle scheduling apparatus, including: a transceiver, a memory, a processor, and a computer program stored on the memory and executable on the processor; the processor is configured to read a program in the memory to implement the steps in the shared bicycle scheduling method according to the first aspect.

In a fourth aspect, an embodiment of the present invention further provides a readable storage medium, including: a processor, a memory and a program stored on and executable on the memory, which when executed by the processor, performs the steps in the shared bicycle scheduling method as described above in relation to the first aspect.

The technical scheme of the invention at least has the following beneficial effects:

in the scheme, the number of the shared bicycles dispatched to each placing point by the dispatching center can be predicted by determining the dispatching center in the target area, constructing the shared bicycle demand prediction model of each placing point in the target area and determining the number of the shared bicycles dispatched to each placing point by the dispatching center according to the shared bicycle demand prediction model, and dispatching the shared bicycles by the dispatching center, so that the on-demand throwing is realized, the throwing imbalance is avoided, and the operation cost is reduced.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive exercise.

Fig. 1 is a flowchart of a shared bicycle scheduling method according to an embodiment of the present invention;

FIG. 2 is a schematic illustration of a target area provided by an embodiment of the present invention;

fig. 3 is a block diagram of a shared bicycle dispatching device according to an embodiment of the present invention;

fig. 4 is a schematic diagram of a hardware structure of a shared bicycle dispatching device according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In order to enable those skilled in the art to better understand the embodiments of the present invention, the following description of the genetic algorithm is given.

Scientific definition of genetic algorithm:

the concept of genetic algorithms was originally derived from a computer science simulation-based study of the genetics of the system for studying the human micro-animal life cycle. The method is a scientific method for carrying out global biological search and gene optimization on randomness established by developing the human biological system by simulating the evolution and evolution mechanism of the human biological system in the modern human research nature, and the theory of the evolution and evolution mechanism of the human biological system in Darwen and the biological gene evolution theory of Mendelian are used for reference. The method is an efficient, parallel and global relevant search information technology and management method, can help a search user to automatically acquire and automatically analyze the accumulated relevant basic knowledge of the relevant network search information space in the whole relevant search process, and can automatically and adaptively adjust and automatically control the whole relevant search process so as to continuously obtain the best user understanding.

Each chromosome in the genetic algorithm corresponds to a solution of the genetic algorithm, and generally, a formula of an adaptive function (fitness function) is adopted to judge and measure the efficiency and the superiority of the solution. Thus, a map is constructed from a gene cluster to its degree of incompatibility. We can consider the process of genetic algorithm as a process of finding the best solution inside a multivariate function. There are numerous "peaks" in the multi-dimensional surface, and the peaks correspond to each other, which is the local optimal solution. And the area where there may be only one "mountain peak" is the highest altitude, it is itself a global optimum understanding. One of the main tasks of the genetic algorithm is to try to climb to the highest peak of the genetic algorithm, rather than falling down or falling down to some smaller mountain peak. In addition, it is worth noting that the genetic algorithm does not have to search for the "highest peak," and if the best fitness for the problem is evaluated to be smaller and best, then the lowest value of the function is the global optimal solution, and accordingly, the "deepest valley" is sought by the genetic algorithm.

The related basic concept of genetic algorithm:

genotype: the genotype is a dominant chromosome internal structural characteristic with strong natural characters;

phenotype is as follows: an external expression of a chromosome-determined trait, or, alternatively, an individual formed according to genotype;

and (3) population evolution, namely, along with the continuous evolution of modern human beings, the population gradually adapts to the natural living environment where the population is expected to be, and the quality is continuously developed and greatly improved. The evolution of the main biological population of human and the evolution of human mainly take a certain biological population relationship as a theoretical basis.

Fitness is as follows: measure the adaptation of a species to the living environment.

Selecting: several individuals are selected from the population with a certain probability. Generally, the selection process is a process based on the eligibility of fitness.

DNA replication: that is, when a senescent cell is unable to undergo benign division, DNA replication in the genetic material will transfer its gene out to a new senescent cell by re-replicating it, from which it re-inherits the legacy senescent granulocyte genetic gene.

And (3) crossing: the DNA of the same position of the two chromosomes is cut off, and the two front and rear strings are respectively combined in a crossed manner to form two new chromosomes. Also known as gene recombination or hybridization;

mutation: the whole process of genetic replication has certain mutation probability, and the mutation can directly generate a new chromosome, which shows that the new chromosome creates a new genetic characteristic, namely a new trait for us.

And (3) encoding: genetic information in DNA is encoded by the sequential arrangement of longer spots in a pattern on a strand. Genetic coding can therefore also be broadly considered as a mapping from phenotype to genotype.

And (3) decoding: one genotype is mapped to a phenotype decode.

Individual: individuals are broadly defined as entities with distinct characteristics of the chromosome.

Population: a collection of individuals, the number of individuals in the collection being referred to as the size of the population.

Referring to fig. 1, fig. 1 is a flowchart of a shared bicycle scheduling method according to an embodiment of the present invention, and as shown in fig. 1, the method includes the following steps:

step 101, determining a dispatching center in a target area;

in the step, taking a campus as an example, the campus is regionalized, a plurality of areas are divided, and a scheduling center is established for each area, wherein at least one area is used as a target area, and address selection is performed in the target area to determine the scheduling center.

102, constructing a shared bicycle demand prediction model of each placing point in the target area;

in this step, since each place in the target area faces different user demands, a shared bicycle demand prediction model needs to be constructed for each place.

It is noted that the shared-vehicle demand prediction model may be determined using a time series prediction method or an influence factor analysis method.

And 103, determining the number of the shared bicycles dispatched to each placing point by the dispatching center according to the shared bicycle demand prediction model.

That is, based on the shared-bicycle demand prediction model, the number of shared bicycles dispatched to each placing point by the dispatching center can be determined, and the shared bicycles can be dispatched to each placing point uniformly by the dispatching center in the target area, so that the on-demand delivery is realized, the management level of the shared bicycles is improved, and the appearance of the city are improved.

In the embodiment, a scheduling center in a target area is determined, a shared bicycle demand prediction model of each placing point in the target area is built, the number of shared bicycles scheduled to each placing point by the scheduling center is determined according to the shared bicycle demand prediction model, the number of the shared bicycles of each placing point can be predicted, the shared bicycles are scheduled by the scheduling center, on-demand delivery is achieved, disorder delivery and imbalance of delivery are avoided, operation cost is reduced, meanwhile, the shared bicycles are centrally managed by the scheduling center, and the damaged shared bicycles are delivered to each placing point in the target area after being maintained.

In one embodiment, step 101 includes:

acquiring coordinate information of each placing point in the target area;

In order to select a dispatching center in a target area, namely solve the problem of site selection of the dispatching center, firstly, the placing points of all shared bicycles in the target area need to be represented on a map by using coordinates.

The position of the shared bicycle placing point arranged in the target area can be determined according to the field investigation result, and the coordinate information of the placing point is obtained, wherein the coordinate information is the center coordinate.

For example, according to the results of the on-site investigation, as shown in fig. 2, it can be found that 1 placement point is respectively provided at east side of No. 11 student dormitory building, No. 5 and No. 6 student dormitory building, No. 13 and No. 14 student dormitory building, No. 7 and No. 8 student dormitory building, No. 12 student dormitory building and No. 10 office building, and the sports culture hall, and 6 placement points are provided in total.

Taking the corner of the No. 16 student dormitory building and the No. 5 student dormitory building as the origin of coordinates, constructing a rectangular coordinate system of the target area, and obtaining the coordinates of the following 6 placing points:

(1.7，1.1)；(1.7，3.4)；(1.7，4.8)；(5.1，5.2)；(7.2，3.1)；(6.0，1.2)。

the position of the scheduling center is selected in the target area, namely, the cost optimal problem is selected, namely, the address is selected when the cost optimal value is solved, so that the cost optimal value is converted into the problem of the cost minimum value, the operation planning linear programming is used for carrying out model construction, the following scheduling center operation cost function is constructed, and the target coordinate information with the minimum distance corresponding to the cost minimum value is obtained based on the scheduling center operation cost function.

Further, according to the fitness function, target coordinate information corresponding to the minimum value is obtained, and therefore the address of the dispatching center is determined.

In an embodiment, the constructing a scheduling center operation cost function according to the coordinate information of each placement point includes:

It should be noted that the following scheduling center operation cost function is constructed:

wherein, C represents the total operation cost of the dispatching center;

representing an objective function, representing the cost of the scheduling center in the actual use process due to the scheduling distance;

a represents the hiring cost of hiring the relevant scheduler, which can be considered a constant.

The fitness function to be solved is the lowest value of the distance-related cost, and belongs to the problem of minimum value taking, wherein the objective function and the fitness function need to be converted, and the fitness function is the opposite number of the objective function, which is specifically as follows:

in an embodiment, the obtaining, based on the scheduling center operation cost function, target coordinate information corresponding to a minimum value of the scheduling center operation cost includes:

acquiring target coordinate information of a target scheduling center;

It should be noted that, a target scheduling center corresponding to the minimum value of the operation cost of the scheduling center is obtained by using a genetic algorithm. The genetic algorithm is realized by the following steps:

the implementation process of genetic algorithm is better than the biological evolution process in the natural process. First, a potential text solution to this problem, namely, digitized text encoding, needs to be found and obtained. Then, a random population is initialized by random numbers, and the individual data in the random population is the digitalized data codes of the populations. Next, after decoding through the appropriate genes, the same gene fitness function evaluation is performed once for each gene using one gene fitness evaluation function. The selected optimal value of the specified parameter is selected by a function of the selectable value according to a certain specific condition. The variation of the progeny genome is initiated and then allowed to produce a progeny. Wherein, the operation steps of the genetic algorithm are defined as follows:

step 1: and evaluating the fitness of the individual corresponding to each chromosome.

Step 2: and respectively selecting two individuals from one population as a father party and a mother party according to the principle that the higher the adaptability degree is, the higher the selection probability is.

And step 3: two male chromosomes with different aspects between different ancestors and parents are extracted by the method, and filial generation exchange is carried out to generate a new filial generation.

And 4, step 4: and (5) carrying out mutation on the chromosomes of the offspring.

And 5: repeating steps 2, 3 and 4 until a new population is generated.

Step 6: the loop is ended.

Several important concepts in the application of genetic algorithms:

(1) binary coding

Inspired by the profound influence and research of biologists on the basic structure of human chromosomes, it can be assumed that currently there are only two bases of '0' and '1', and they are directly connected together in series in order by a simple staining chain, because each basic unit must have basic information quantity capable of accurately showing 1bit for its representation, so that only a chain chromosome long enough is enough to accurately outline all the characteristics of an individual. This is a binary-based staining encoding method, and the chromosome has the following approximate structure:

010010011011011110111110

(3) roulette as a selection method in genetic algorithms

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

The cumulative total fitness is therefore:

so the probability of each individual being selected is:

when a new wheel is rotated in the air and has stopped, the pointer may randomly point to a location or area represented by an individual, meaning that the individual is directly selected.

(2) Genetic recombination in genetic algorithms

Genetic recombination and mutation are fundamental to the differentiation between offspring and parents. For the genetic operation of these two genes, there is a great difference in the way signals are processed between binary coding and floating-point coding. The process of binary-coded gene exchange is very similar to the process of homologous chromosome association taught in the organisms in the high-school period, namely, several codes at the same time or position are randomly exchanged to generate a new individual.

(3) Genetic algorithm gene mutation

The process of gene mutation formation: the term "gene mutation" refers to a change in the structure of a gene in a chromosome at a specific gene locus. Allelic mutations of a gene can usually directly transform one allele into another allele, and often mutation directly results in a certain genetic appearance and phenotype variation of the gene. That is, the operation process of binary gene-encoded genetic gene operation sequence is very similar to that of various genetic gene operation processes in genetic biology, and the "0" or "1" sequence on the gene string is converted into the completely opposite "1" or "0" sequence with a certain probability.

Wherein a first candidate dispatch center is obtained in the target area by roulette. Here, the process of acquiring the first candidate dispatch center may be regarded as a process of selecting a population in a genetic algorithm, and the specific steps are as follows:

and (4) performing optimal reservation operation, sequencing NP individuals in the contemporary population from large to small according to fitness, marking the individual with the maximum fitness as an optimal individual elite, ranking the optimal individual elite at the first position, and performing mountain climbing operation without participating in crossing and variation processes.

And performing wheel selection operation and selecting NP-1 filial generations.

Firstly, calculating the probability of each chromosome in the current population being selected, and calculating by using a probability calculation formula:

then, the cumulative probability for each individual is calculated:

wherein, i is 1, 2.

And then carrying out individual selection on the current population, and if the selected individual is not the optimal individual, carrying out repeated selection until the selected individual is NP-1 in scale.

In an embodiment, the performing genetic iteration based on the coding coordinates and the fitness of the multiple first candidate scheduling centers to obtain a target scheduling center corresponding to the minimum value of the operation cost of the scheduling center includes:

It should be noted that, a second candidate dispatch center with the highest fitness is selected from the first candidate dispatch centers, and then a plurality of second candidate dispatch centers are subjected to gene recombination and/or gene mutation, wherein the gene recombination is similar to a biological association process and is an array exchange between two binary codes, and the gene mutation is that one or several '0's, which are independent of a group of two-dimensional codes, are changed into '1' or changed from '1's into '0's. Each gene recombination or mutation will result in a change in the binary code and thus in the value of its corresponding coordinate. That is, each gene recombination or gene mutation will generate a new individual, i.e., a second candidate dispatch center.

In the actual use and development process of the genetic algorithm, when a new individual, namely a new population is iterated, the parent data of the new individual needs to be calculated with high adaptability, so that one parent data in the most adaptable population is selected to be iterated next time. The specific operation method is as follows:

finding out individual elite with the optimal fitness function in the male parent, carrying out gene recombination and/or gene mutation by using the individual elite to obtain new individual elite-1, calculating the fitness of the elite-1, and if the fitness function of the elite-1 is greater than that of the elite, taking the elite-1 as the optimal individual in the new generation of population. The above operations are repeated until no better elite-1 individuals can be found.

In the genetic iteration process, the iteration times are preset as termination conditions of iteration, and the iteration is continued until the iteration times are reached to finish the genetic iteration.

An exemplary procedure for the genetic algorithm is as follows.

In one embodiment, step 102 comprises:

acquiring the travel data of the shared bicycle at each placing point;

determining a plurality of target factors in the plurality of potential factors and determining an influence factor function of each target factor according to the travel data;

wherein the objective factors include at least one of:

age factors;

a sex factor;

a weather factor;

travel distance factors.

It should be noted that, in the embodiment of the present invention, since the area in the campus is taken as an example of the target area, an influence factor analysis method may be used herein to determine the shared bicycle demand prediction model.

Wherein for the target area, a latent factor is determined, which may include at least one of a gender factor, an age factor, an occupation factor, an income factor, a travel habit factor, a road factor, and a weather factor.

In order to obtain travel data, a field survey method and a questionnaire survey method are adopted to collect the shared bicycle travel data in a target area. First, the field survey results were as follows:

(1) the main distribution points of the shared single vehicles in the target area are each dormitory building area and each individual teaching building area, and the number of the shared single vehicles at each placement point is basically between 30 and 50.

(2) The shared bicycle in the dormitory building area is basically lent by students before nine morning hours and then is distributed in each area in the campus.

(3) Through the squat survey of the canteens, the flow peak of the canteens sharing bicycle is found to be 11 noon: 30 to 12: 00, and during dinner time, the flow of the shared bicycle is reduced.

Then, in order to understand the degree of dependence of students and teachers on the shared bicycle in the target area and the degree of satisfaction of management of the shared bicycle in the current target area, subjective survey needs to be conducted on the students and teachers, and a questionnaire survey method is adopted as the most direct and simple method. Because one of the questionnaires is biased to the subjective survey of the subject, the anti-choice questions need to be set reasonably in the questionnaire design stage, and the effective degree of the questionnaires and the concentration degree of the questionnaires are improved.

Further, according to the travel data, analyzing the influence factors, and determining the target factors comprises: gender, age, weather, and distance factors, and determining an influence factor function for each target factor.

Specifically, the impact factor function for each target factor is as follows:

(1) sex-affecting factor function:

since the usage of the sharing bicycle has a linear correlation with the target factor of gender, it is assumed that 100a sharing bicycles are supplied for every hundred males, where a is the influence factor function corresponding to the gender of the male, and k × a is the influence factor function corresponding to the gender of the female, where k is the linear coefficient between the male and female users.

According to the questionnaire survey result, among the male users, the high-frequency user proportion of the shared bicycle is 22%, the medium-frequency user proportion is 47%, and the low-frequency user proportion is 31%, and then the three frequencies are weighted to obtain the average value, so that the influence factor function a can be obtained.

Here, in combination with the actual situation and the comprehensive consideration of the function fitting, the weight coefficient ratio corresponding to the frequency may be set to 6: 3: 1, thereby obtaining:

male gender corresponding impact factor function:

a＝0.22×0.6+0.47×0.3+0.31×0.1＝0.304

from the questionnaire and the shared single-vehicle industry usage report, the linear coefficient k is 0.716. Thus, the impact factor function for female gender:

k×a＝0.716×0.304＝0.218

(2) age influence factor function:

in the research environment in the target area, the age difference is basically represented as the difference between the teacher and the student, so that the questionnaire is processed according to the age group, and the student is far higher in dependence on the sharing bicycle than the teacher and accords with the linear correlation relationship. It is assumed that for each hundred students, a shared bicycle of 100b is supplied, b is the student influence factor function, and n × b is the teacher influence factor function, where n is the linear coefficient between students and teacher users in both age groups.

According to the questionnaire survey result, in students, the high-frequency user proportion of the shared bicycle is 29%, the medium-frequency user proportion is 52%, the low-frequency user proportion is 19%, and the three frequencies are weighted to obtain the average value, so that the value of the influence factor b can be obtained.

In the present study, the weighting factor ratio was set to 6, in combination with the actual situation and the comprehensive consideration of the function fitting: 3: 1, so that finally:

student impact factor function:

b＝0.29×0.6+0.52×0.3+0.19×0.1＝0.349

from the questionnaire, the linear coefficient n was 0.193.

Thus, the teacher influences the factor function:

n×b＝0.193×0.349＝0.061

(3) weather effect factor function: the use of a shared bicycle inside a target area is bound to be affected by the weather. Suppose that no one in the target area goes out by using the sharing bicycle in severe weather conditions such as rainy days. Therefore, the weather conditions are classified into high-temperature days, sunny days, cloudy days, low-temperature days, and windy days. Meanwhile, the judgment can be carried out according to the conventional principle. The number of people using the shared bicycle is the most in sunny days, and the number of people in low-temperature days and windy days is the least in high-temperature days and cloudy days, so that the influence of weather on the use condition of the shared bicycle in a target area is assumed to be in accordance with the following quadratic function:

y＝px ² +qx+c

and then establishing the corresponding relation between the weather and the function independent variable x as follows:

when the weather is sunny, making x equal to 1; when the weather is high-temperature day, making x be 2, and when the weather is cloudy day, making x be-1; when the weather is low-temperature days, making x be-2; and when the weather is strong wind, making x equal to-2.5.

When x is 1, let y be 1, y be a weather influence factor, and according to the question "in the questionnaire, it is assumed that in a certain case on a sunny day, you will select the shared bicycle for going out, and now replace the weather with the following conditions, and what the weather that you will still select the shared bicycle for going out" the following weather influence factor function can be obtained:

y＝-0.238x ² +0.114x+1.124

(4) distance impact factor function:

in the target area, whether students use the sharing bicycle or not when going out, the distance between students and the sharing bicycle has a certain causal relationship, if the distance is long, the students are difficult to go out by riding the bicycle, and if the distance is short, most of the students still can choose to walk to reach. And fitting the data collected by the questionnaire, so that the influence factors of the distance on the use condition of the shared bicycle in the target area can be found to basically accord with the function characteristics of the exponential function.

By collecting and analyzing data of a question "how far you can choose to ride a sharing bicycle when traveling inside a campus" in a questionnaire, four options of "arrive within five minutes of walking", "arrive within five to ten minutes of walking", "arrive within ten to twenty minutes of walking", and "arrive more than twenty minutes of walking" in the questionnaire are associated with a variable t:

when the travel distance is within five minutes of walking, making t equal to 1; when the travel distance is 'five to ten minutes of walking' to arrive, let t be 2; when the travel distance is 'ten to twenty minutes to arrive on foot', t is made to be 3; when the travel distance is "arrive twenty minutes or more by walking", let t be 5.

Let T be the distance influence factor, which is obtained by questionnaire data analysis, and when T is 5, the distance influence factor function T is 0.961, when T is 1, the distance influence factor function T is 0.107, and when T is 2, the distance influence factor function T is 0.403.

Thus, the distance impact factor function can be obtained as an exponential function:

T＝0.0705e ^0.539t

that is to say, the influence factors are respectively solved according to the four influence factors, so that four accumulation functions, namely influence factor functions, can be obtained, and when the demand prediction is performed on the station, the four influence factor functions need to be summarized and fitted, so that the shared bicycle demand prediction model is obtained.

It can be easily inferred that the shared bicycle demand prediction model is a composite function formed by combining one or more of the four influence factor functions, so that the four influence factor functions are required to be considered as a basis function and placed on the first layer of the composite function, and the shared bicycle demand prediction model is finally obtained by analogy.

Since the influence of gender and age on the traveling of the shared bicycle in the target area has linear characteristics, and the influence factor functions of the gender and the age are the most basic linear functions, the two influence factor functions can be combined as the functions of the bottommost layer of the composite function, and besides, the two influence factors are the factors owned by the shared bicycle user and are quite reasonable as the most basic influence factors.

The two influence factors are independent and do not influence each other as the distance and weather of the external influence condition, so that the two influence factors can be called as environmental influence factor factors.

As for the sex and age of the influence factors of the user, the sex and age influence the use of the shared bicycle at the same time, so that the sex and age cannot be directly multiplied, and the requirement prediction needs to be carried out by considering the construction of new influence factors of the user.

Wherein, let a be a sex influence factor function, stipulate as follows:

male sex influence factor function a ₁ ＝0.304；

Female sex influence factor function a ₂ ＝0.218；

Let b be an age-influencing factor function, specified as follows:

student influence factor function b ₁ ＝0.349；

Teacher influence factor function b ₂ ＝0.061。

To construct a new user's own influence factor function for demand prediction, the gender influence factor function and the age influence factor function are combined, and here, the new influence factor function is constructed by using geometric mean, and finally, the following results are obtained:

male student influence factor function f ₁ ＝0.326；

Male teacher influence factor function f ₂ ＝0.136；

Female student influence factor function f ₃ ＝0.275；

Female teacher influence factor function f ₄ ＝0.115。

Because the influence factor function of the testee is fixed and constant, the travel distance is almost constant in a short period, and the travel distance of the teacher in the same school is only changed in a fixed interval in a school period, the distance influence factor function and the geometric mean of the influence factor function of the user can be used as an inherent influence factor function, and the weather is multiplied by the constant influence factor function:

finally, the shared bicycle demand prediction model is as follows:

in an embodiment, the obtaining the travel data of the shared bicycle at each placing point includes:

and acquiring the travel data of the shared bicycle of each placing point with the reliability coefficient in a preset range.

It should be noted that the reliability analysis is a method for measuring samples, that is, measuring whether the subject really answers during answering. The clonal Bach confidence coefficient was used as a criterion for confidence. If the Bach confidence coefficient of the clone is more than 0.8, the confidence scale can be directly verified or the test confidence is good; if the Bach confidence coefficient of the clone is above 0.7, the confidence is generally considered to be widely accepted; if the clone Bach confidence coefficient is in the range of 0.6 to 0.7, the confidence scale or test confidence is generally considered to have some international academic research value, but may require more revisions, and if the clone Bach confidence coefficient is below 0.6, the test requires re-topical to ensure the scientificity of the test.

The SPSSAU (a data science analysis platform system) is used for analysis, and the clone Bach reliability coefficient of the questionnaire is 0.941, so that the reliability of the questionnaire is very good, and the collected result can completely support subsequent analysis.

In addition, validity analysis is needed to be performed on the questionnaire, validity is the validity of the questionnaire, validity can be used to measure whether the design correctness and reliability of the question items (including quantitative data) of the questionnaire are correct or not, verification is performed by methods such as factor analysis (exploratory factor analysis), and due to the fact that the corresponding relation between the preset variables and the question items is performed, after the factor analysis is performed, the corresponding relation between the factor (namely, one variable, which is called as a factor when the factor analysis is used) and other question items is performed, and if the two are basically the same, it can be said that the question items have good working validity.

The validity analysis KMO value of the questionnaire is 0.778 by analyzing the result by using SPSSAU. When the KMO value is used as the most direct validity judgment standard, the KMO value reaches 0.6, namely the validity criterion is met, and 0.7 is more suitable, so that the validity of the questionnaire completely meets the criterion and can be used for subsequent analysis.

In an embodiment, the determining, according to the travel data, a plurality of target factors of the plurality of potential factors includes:

acquiring a correlation between each potential factor and the travel data;

and determining the plurality of target factors according to the correlation.

It should be noted that the questionnaire analysis of this time is an internal rule of each potential factor and the shared bicycle usage in the target area, but before the research result is determined, it cannot be said that there is a strong correlation between the two, so after sufficient answers are collected, a simple correlation analysis needs to be performed first to determine a correlation between each potential factor and the shared bicycle usage, that is, the travel data, and here, a cross analysis method and a correlation analysis method may be used to obtain the correlation between each potential factor and the travel data.

The cross analysis method is that the existing collected questionnaire survey answer is used, several questions are selected as dependent variables in personality test questions, one question is selected as independent variable according to the use frequency, the two questions are analyzed, a relevant chart is drawn, and whether objective relevant relationship exists between the two questions is visually observed. In order to facilitate the cross analysis, the counter topic is not considered as a dependent variable, so that the selection topic "your sex is" as a dependent variable, the selection topic "the number of times you share a single vehicle per month" is used as a dependent variable, and the correlation between the two is analyzed.

The results of the cross-analyses are shown in table 1 below.

Sharing the number of uses of a single vehicle per month	Sex of male	Female with a view to preventing the formation of wrinkles
			5 times or less	2(4.17％)	24(50％)
5 times to 10 times	20(18.18％)	54(49.09％)
			10 times to 20 times	19(27.94％)	30(44.12％)
More than 20 times	1(11.11％)	4(44.44％)

TABLE 1

In addition, with the above table 1, it can be known that the usage of the shared bicycle by the male subject and the female subject in the target area are obviously different, and the male subject and the female subject may have a proportional relationship in number, so that after the number of the female subjects is doubled, the score of the male subject on the monthly usage number of the shared bicycle is 1.4 times that of the female subject, and therefore, it can be determined that the sex factor has a certain correlation with the trip data of the shared bicycle.

The correlation analysis method includes the steps of respectively selecting a question from a test subject influence factor condition investigation part of a questionnaire and a test subject shared single-vehicle condition investigation part, and calculating correlation parameters of the question and the test subject by using a mathematical statistics principle, so that the evidence that potential factors and travel data of a shared single vehicle in a campus have a determined correlation relationship. The following two different correlation coefficients were used to determine the correlation between the two questionnaires:

the items entitled "distance for each trip (class, meal, express, etc.) and" the number of times you use the shared bicycle per month "are selected for analysis here.

The following table 2 shows the correlation analysis using the Kendalls correlation coefficient.

Question topic in questionnaire	Distance of going out (class, meal or express delivery, etc.) each time
		Number of times you use shared bicycle per month	0.487

TABLE 2

As can be seen from table 2 above, the correlation between "distance of you going out (class, eat or take express delivery, etc.) each time" and "the number of times you use the shared bicycle per month" is analyzed by using the correlation analysis method, and the Kendall correlation coefficient is used to represent the strength of the correlation. A concrete analysis revealed that the correlation between "distance of you going out (class, eat or take express delivery, etc.) each time" and "the number of times you used the shared bicycle each month" had a numerical value of 0.487, and exhibited significance at a level of 0.01.

Therefore, it can be judged that there is a significant positive correlation between "the distance you travel (class, eat or take express, etc.) each time" and "the number of times you use the shared bicycle per month".

The correlation analysis was performed using Spearman correlation coefficients as in table 3 below.

Question topic in questionnaire	Distance of going out (class, meal or express delivery, etc.) each time
		Number of times you use shared bicycle per month	0.574

TABLE 3

As can be seen from table 3 above, the correlation between "distance of each trip (class, meal or express delivery, etc.) and" number of times you use shared bicycle per month "is analyzed by using the correlation analysis method, and the Spearman correlation coefficient is used to represent the strength of the correlation. Specific analysis revealed that the correlation value between "the distance you travel (class, eat or take courier, etc.) each time" and "the number of times you use the shared bicycle each month" was 0.574, and a significance of the 0.01 level was exhibited.

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

In one embodiment, step 103 comprises:

It should be noted that after the shared bicycle demand forecasting model is constructed, the shared bicycle demand forecasting model needs to be applied to an example to verify the validity of the shared bicycle demand forecasting model, 400 student apartments in a target area are taken as an example, and in order to ensure that the demand is met, clear weather is selected as a research object, so that the weather influence factor function y is 1. And the walking distances of the student apartment from the frequently-visited library, the information teaching building and the central teaching building are all about 10 minutes to 15 minutes, so that the distance influence factor is calculated by substituting the distance influence factor function with t being 3 into the distance influence factor function formula:

T＝0.0705e ^0.539t ＝0.355

and obtaining an influence factor F of 0.340 according to the shared bicycle demand prediction model.

Thus, the demand of the shared bicycle of the student apartment is obtained as 400 × 0.340-136.

Further, after knowing that the daily demand of the shared bicycles in the student apartment is 136 vehicles on a sunny day, it does not mean that the shared bicycles in the demand need to be dispatched to the placement point of the student apartment by the dispatching center, and the dispatching center needs to dispatch the number of the shared bicycles to the placement point based on the demand and in combination with the reuse rate and the operation cost of the shared bicycles. That is, according to the shared bicycle demand prediction model, demand data is provided for the dispatching center, so that the dispatching center can consider how to dispatch and how to operate.

In summary, the embodiment of the invention determines the dispatching center in the target area through genetic iteration, constructs the shared bicycle demand prediction model of the placing points, and determines the number of shared bicycles dispatched to each placing point by the dispatching center, so that the problem of unbalanced release caused by disordered release is solved, the operation cost is reduced, the shared bicycles are uniformly managed by the dispatching center, the damaged vehicles can be dispatched after being maintained, and the influence on the use of the damaged vehicles is avoided.

As shown in fig. 3, an embodiment of the present invention further provides a shared bicycle scheduling apparatus, including:

a first determining module 301, configured to determine a scheduling center in a target area;

a first construction module 302, configured to construct a shared bicycle demand prediction model for each placement point in the target area;

a second determining module 303, configured to determine, according to the shared-bicycle demand prediction model, the number of shared bicycles dispatched to each placement point by the dispatch center.

Optionally, the first determining module 301 includes:

the first acquisition unit is used for acquiring coordinate information of each placing point in the target area;

the first construction unit is used for constructing a scheduling center operation cost function according to the coordinate information of each placing point;

the first obtaining unit is used for obtaining target coordinate information corresponding to the minimum value of the operation cost of the dispatching center based on the operation cost function of the dispatching center;

and the first determining unit is used for determining the dispatching center according to the target coordinate information.

Optionally, the first building unit is specifically configured to:

Optionally, the first obtaining unit includes:

the first obtaining subunit is configured to obtain a plurality of first candidate scheduling centers, and encode the coordinate information of the plurality of first candidate scheduling centers respectively to obtain encoded coordinates of the plurality of first candidate scheduling centers;

the calculating subunit is configured to calculate, according to the scheduling center operation cost function and the coordinate information of the plurality of first candidate scheduling centers, the fitness of the plurality of first candidate scheduling centers respectively;

the obtaining subunit is configured to perform genetic iteration based on the coding coordinates and the fitness of the multiple first candidate scheduling centers, and obtain a target scheduling center corresponding to the minimum value of the operation cost of the scheduling center;

the second acquisition subunit is used for acquiring target coordinate information of the target scheduling center;

Optionally, the obtaining subunit is specifically configured to:

Optionally, the first building module 302 includes:

the second acquisition unit is used for acquiring a plurality of potential factors influencing the traveling of the shared bicycle;

a third obtaining unit, configured to obtain travel data of the shared bicycle at each placement point;

a second determining unit, configured to determine, according to the travel data, a plurality of target factors in the plurality of potential factors, and determine an influence factor function of each target factor;

the second construction unit is used for fitting the influence factor function of each target factor to construct the shared bicycle demand prediction model;

wherein the objective factors include at least one of:

age factors;

a sex factor;

a weather factor;

travel distance factors.

Optionally, the third obtaining unit is specifically configured to:

Optionally, the second determining unit is specifically configured to:

acquiring a correlation between each potential factor and the travel data;

and determining the plurality of target factors according to the correlation.

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

The apparatus provided in the embodiment of the present invention may implement the method embodiments, and the implementation principle and the technical effect are similar, which are not described herein again.

As shown in fig. 4, the shared bicycle scheduling apparatus according to the 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 for storing programs and data used by the processor 400 in executing operations, and the processor 400 calls and executes the programs and data stored in the memory 420.

The processor 400 is used for reading the program in the memory 420 and executing the following processes:

determining a dispatching center in a target area;

constructing a shared single-vehicle demand prediction model of each placement point in the target area;

A transceiver 410 for receiving and transmitting data under the control of the processor 400.

Where in fig. 3, the bus architecture may include any number of interconnected buses and bridges, with various circuits of one or more processors, represented by processor 400, and memory, represented by memory 420, being linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The transceiver 410 may be a number of elements, including a transmitter and a transceiver, providing a means for communicating with various other apparatus over a transmission medium. For different user devices, the user interface 430 may also be an interface capable of interfacing with a desired device externally, including but not limited to a keypad, display, speaker, microphone, joystick, etc.

The processor 400 is responsible for managing the bus architecture and general processing, and the memory 420 may store data used by the processor 400 in performing operations.

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

acquiring coordinate information of each placing point in the target area;

acquiring target coordinate information of a target scheduling center;

acquiring the travel data of the shared bicycle at each placing point;

wherein the objective factors include at least one of:

age factors;

a sex factor;

a weather factor;

travel distance factors.

acquiring a correlation between each potential factor and the travel data;

and determining the plurality of target factors according to the correlation.

The device provided by the embodiment of the present invention may implement the above method embodiment, and the implementation principle and technical effect are similar, which are not described herein again.

Those skilled in the art will appreciate that all or part of the steps for implementing the above embodiments may be performed by hardware, or may be instructed to be performed by associated hardware by a computer program that includes instructions for performing some or all of the steps of the above methods; and the computer program may be stored in a readable storage medium, which may be any form of storage medium.

In addition, the specific embodiment of the present invention further provides a readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps in the shared bicycle scheduling method, and can achieve the same technical effects, and in order to avoid repetition, the detailed description is omitted here.

In the several embodiments provided in the present application, it should be understood that the disclosed method and apparatus may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be physically included alone, or two or more units may be integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.

The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) to execute some steps of the transceiving method according to various embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A shared bicycle scheduling method, comprising:

determining a dispatching center in a target area;

2. The method of claim 1, wherein determining the dispatch center in the target area comprises:

acquiring coordinate information of each placing point in the target area;

3. The method of claim 2, wherein constructing a scheduling center operation cost function according to the coordinate information of each placing point comprises:

4. The method according to claim 2, wherein the obtaining target coordinate information corresponding to a minimum value of the scheduling center operation cost based on the scheduling center operation cost function includes:

acquiring target coordinate information of a target scheduling center;

5. The method of claim 4, wherein performing genetic iteration based on the coded coordinates and fitness of the first candidate scheduling centers to obtain a target scheduling center corresponding to a minimum value of an operating cost of the scheduling center comprises:

6. The method of claim 1, wherein the constructing a shared bicycle demand prediction model for each pose point in the target area comprises:

acquiring the travel data of the shared bicycle at each placing point;

wherein the objective factors include at least one of:

age factors;

a sex factor;

a weather factor;

travel distance factors.

7. The method of claim 6, wherein the obtaining of travel data of the shared bicycle for each place comprises:

8. The method of claim 6, wherein said determining a plurality of target factors from said plurality of potential factors from said travel data comprises:

acquiring a correlation between each potential factor and the travel data;

and determining the plurality of target factors according to the correlation.

9. The method of claim 1, wherein determining the number of shared vehicles to dispatch to each of the placement points by the dispatch center according to the shared vehicle demand prediction model comprises:

10. A shared bicycle dispatching device, comprising:

11. A shared bicycle scheduling apparatus comprising: a transceiver, a memory, a processor, and a computer program stored on the memory and executable on the processor; characterized in that the processor, reading the program in the memory, implements the steps in the shared bicycle scheduling method according to any one of claims 1 to 9.

12. A readable storage medium, comprising: a processor, a memory, and a program stored on and executable on the memory, the program, when executed by the processor, implementing the steps in the method of shared bicycle scheduling of any of claims 1 to 9.