A method and system for calculating the quantity of shared electric motorcycles based on supply and demand balance

Abstract

The present invention discloses a method and system for calculating the amount of shared electric motorcycles based on supply and demand balance. The present invention uses shared electric motorcycle order data, shared electric motorcycle electronic fence data, urban point of interest POI data and traffic community data to build a regression model to analyze the usage characteristics and influencing factors of shared electric motorcycles. According to the model residual, the traffic community is divided into a reasonable deployment area, an oversaturated deployment area and an undersaturated deployment area. By analyzing the borrowing and returning demand of shared electric motorcycles in different partitions, the deployment gap of shared electric motorcycles is predicted, and on the basis of measuring the parking carrying capacity of the city, the deployment amount of shared electric motorcycles is determined from the perspective of supply and demand balance. Based on the analysis of the rationality of the current deployment of shared electric motorcycles, the present invention provides a method for calculating the deployment amount of shared electric motorcycles in the city. The method is simple and easy to understand, highly operational, and can provide guiding suggestions for the deployment of shared electric motorcycles in the city.

Description

Technical Field

[0001] The present invention relates to the technical field of shared electric motorcycle deployment, and in particular to a method and system for calculating the deployment amount of shared electric motorcycles based on supply and demand balance. Background Art

[0002] As a green means of transportation that effectively alleviates traffic congestion and saves energy and reduces emissions, shared electric motorcycles not only alleviate the daily travel difficulties of urban residents such as "crowded buses, expensive taxis, and difficult driving", but also improve the city's green travel efficiency and reduce traffic carbon emissions. It is a transportation method that fits the current low-carbon development of society. However, while shared electric motorcycles are booming, the drawbacks of the system in the actual operation process are becoming increasingly prominent, and problems such as chaotic parking order and traffic safety have emerged. The fundamental reason is the imbalance between supply and demand of vehicles. In order to improve the parking order of shared electric motorcycles and better meet the travel needs of urban residents for shared electric motorcycles, it is necessary to determine a reasonable amount of shared electric motorcycles within the parking capacity of urban shared electric motorcycles. Summary of the invention

[0003] In order to solve the problems existing in the prior art, the present invention provides a method and system for calculating the deployment of shared electric motorcycles based on supply and demand balance. The method uses shared electric motorcycle order data, shared electric motorcycle electronic fence data, city point of interest POI data and traffic community data to mine the usage characteristics of shared electric motorcycles, predict the demand for borrowing and returning shared electric motorcycles, and determine the parking capacity of shared electric motorcycles. On the basis of meeting the maximum parking capacity of the city, the method determines the deployment measurement of shared electric motorcycles from the perspective of supply and demand balance, provides guiding suggestions for the deployment of shared electric motorcycles in the city, and solves the problems mentioned in the above background technology.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solution: a method for calculating the quantity of shared electric motorcycles based on supply and demand balance, comprising the following steps:

[0005] S1. Build a database for calculating the number of shared electric motorcycles put into use. The database information includes shared electric motorcycle order data, shared electric motorcycle electronic fence data, urban point of interest POI data and traffic community data;

[0006] S2. Data preprocessing: Clean the shared electric motorcycle order data in the database of step S1 and remove invalid and abnormal data;

[0007] S3. Statistics on the usage characteristics of shared electric motorcycles: Taking the traffic area as the statistical unit, the average daily borrowing volume, return volume and average daily turnover rate of shared electric motorcycles are counted respectively.

[0008] S4. Analysis of factors affecting the borrowing and returning of shared electric motorcycles: Taking the traffic community as the analysis unit, regression models for the borrowing and returning of shared electric motorcycles were established respectively, and the residuals of the borrowing and returning numbers of shared electric motorcycles in each traffic community were obtained by solving the model. δ1 ;

[0009] S5. Classification of shared electric motorcycle deployment areas: Based on the residual obtained in step S4 δ1 , the traffic communities are divided into three types according to the supply and demand status of shared electric motorcycles: reasonable deployment area, undersaturated deployment area and oversaturated deployment area; according to the division of the three types, the influencing variables retained after solving the regression model in step S4 are used as subsequent explanatory variables and divided into reasonable deployment area explanatory variables, undersaturated deployment area explanatory variables and oversaturated deployment area explanatory variables;

[0010] S6. Determine the reasonable deployment coefficient of shared electric motorcycles: Establish a regression model for the borrowing and returning of shared electric motorcycles in reasonable deployment areas, and solve the model to obtain the reasonable deployment coefficient α of shared electric motorcycles;

[0011] S7. Forecasting the total shortage of shared electric motorcycles: Taking the undersaturated traffic area as the prediction unit, multiply the reasonable deployment coefficient α in step S6 by the explanatory variable of the undersaturated area to predict the demand for shared electric motorcycles in the undersaturated area;

[0012] Then, the difference between the predicted demand and the actual demand is divided by the average daily turnover rate of each traffic zone. Obtain the gap in the deployment of shared electric motorcycles in each traffic community in the under-saturated area;

[0013] Sum up the gaps in all traffic areas to get the total gap in shared electric motorcycles.

[0014] S8. Correction of the gap in the supply of shared electric motorcycles: Based on the total supply gap predicted in step S7, the supply gap is corrected in combination with the potential transfer volume of other modes of transportation and the actual demand in the oversaturated area;

[0015] S9. Shared motorcycle parking capacity measurement: Taking the traffic area as the measurement unit, count the number of shared motorcycle electronic fences in step S1, and use the number of electronic fences to calculate the static parking capacity P of shared motorcycles. i ; Using the static parking capacity P i Combined with the average daily turnover rate of shared electric motorcycles Calculating the parking capacity of shared electric motorcycles

[0016] S10, determine the number of shared electric motorcycles to be deployed: the shared electric motorcycle parking capacity obtained in step S9 is Compare it with the predicted demand M for shared electric motorcycles to obtain the final supply of shared electric motorcycles.

[0017] Preferably, the method for calculating the usage characteristics of shared electric motorcycles in step S3 includes:

[0018] S31. Average daily borrowing volume of shared electric motorcycles: determine the traffic zone to which the shared electric motorcycles belong based on the borrowing latitude and longitude in the order data information, and count the average daily borrowing volume in each traffic zone;

[0019] S32, average daily return volume of shared electric motorcycles: determine the traffic zone to which the shared electric motorcycles belong based on the longitude and latitude of the return in the order data information, and count the average daily return volume in each traffic zone;

[0020] S33. Average daily turnover rate of shared electric motorcycles: The average daily turnover rate of shared electric motorcycles is calculated based on the order data and the total number of vehicles in the traffic area. The calculation formula is as follows:

[0021]

[0022] in, is the average daily turnover rate of shared electric bikes in the ith traffic zone, Q mi is the total number of valid shared motorcycle orders in the i-th traffic zone on the m-th day, S mi is the total number of shared electric motorcycles in the i-th traffic zone on the m-th day, n is the number of traffic zones, and T is the total number of days.

[0023] Preferably, in step S4, the traffic community is used as the analysis unit, the borrowing and returning volumes of shared electric motorcycles are used as the dependent variables of the model, and the POI data of the city's points of interest are used as the explanatory variables of the model, and regression prediction models for the borrowing and returning volumes of shared electric motorcycles are established respectively; the model is solved to obtain the significance level of the explanatory variables and the residual δ1 of the number of shared electric motorcycles borrowed and returned in each traffic community, and the explanatory variables with a significance level ≤0.1 are retained as influencing variables, and the residual δ1 is equal to the actual borrowing and returning volume of shared electric motorcycles minus the value obtained by the regression model.

[0024] Preferably, the regression prediction model includes a linear regression model, a Poisson regression model or a geographically weighted regression model.

[0025] Preferably, in step S5, the basis for dividing the reasonable delivery area, the undersaturated delivery area and the oversaturated delivery area is:

[0026] When the residual δ1 <D min It is the undersaturated zone;

[0027] When D min ≤δ1 <D max It is a reasonable area for placement;

[0028] When the residual δ1≥D max It is the oversaturation zone;

[0029] Among them, D min , D max They represent the minimum and maximum reasonable zone thresholds respectively.

[0030] Preferably, the calculation formula for the total shortage of shared electric motorcycles in step S7 is as follows:

[0031]

[0032] Among them, y is the amount of additional deployment that should be made in under-saturated traffic areas, and y i is the amount of traffic that should be put into the ith traffic zone, y m is the actual delivery volume of the mth traffic zone, is the average daily turnover rate of shared electric motorcycles, and n is the number of traffic communities.

[0033] Preferably, in step S9, the static parking carrying capacity of the shared electric motorcycle is calculated using the following formula:

[0034]

[0035] Among them, i is the traffic area number, P i is the static bearing capacity of traffic zone i, N i is the number of shared bicycle electronic fences in traffic community i, The average number of shared motorcycles parked per electronic fence;

[0036] The formula for calculating the parking capacity of shared electric motorcycles is as follows:

[0037]

[0038] in, To provide the parking capacity for shared electric motorcycles, is the average daily turnover rate of shared electric motorcycles.

[0039] Preferably, determining the quantity of shared electric motorcycles deployed in step S10 includes the following steps:

[0040] S101. Add the current supply of shared electric motorcycles to the supply gap to obtain the predicted demand M for shared electric motorcycles;

[0041] S102. Compare the demand M with the parking capacity of shared electric motorcycles For comparison, if Then take M as the final delivery amount; if Then take As the final delivery amount.

[0042] In addition, to achieve the above-mentioned purpose, the present invention also provides the following technical solutions: a system for calculating the quantity of shared electric motorcycles based on supply and demand balance, the calculation system comprising a database construction module, a data set preprocessing module, a usage feature statistics module, a borrowing and returning influencing factor analysis module, a deployment area type classification module, a reasonable deployment coefficient determination module, a total deployment gap prediction module, a deployment gap correction module, a parking carrying capacity measurement module and a shared electric motorcycle deployment quantity determination module.

[0043] Preferably, the database construction module is used to construct a database for calculating the quantity of shared electric motorcycles deployed;

[0044] The data set preprocessing module is used to clean the shared motorcycle order data in the database and remove invalid data and abnormal data;

[0045] The usage feature statistics module is used to count the average daily borrowing and returning volume of shared electric motorcycles and the average daily turnover rate.

[0046] The borrowing and returning influencing factor analysis module is used to solve the regression model of the borrowing and returning of shared electric motorcycles, and obtain the residual of the borrowing and returning number of shared electric motorcycles in each traffic community. δ1 ;

[0047] The deployment area type classification module is used to divide the traffic area into three types according to the supply and demand status of shared electric motorcycles: a reasonable deployment area, an undersaturated deployment area, and an oversaturated deployment area;

[0048] The reasonable deployment coefficient determination module is used to solve the regression model of the borrowing and returning quantity of shared electric motorcycles in the reasonable deployment area to obtain the reasonable deployment coefficient α of shared electric motorcycles;

[0049] The total gap quantity prediction module is used to predict the total gap quantity of shared electric motorcycles.

[0050] The delivery gap correction module is used to correct the delivery gap;

[0051] The parking capacity measurement module is used to measure the static parking capacity P of the shared electric motorcycle. i And use the static parking capacity P i Calculating the parking capacity of shared electric motorcycles

[0052] The shared electric motorcycle deployment determination module is used to determine the shared electric motorcycle parking capacity. Compare with the predicted demand M for shared electric motorcycles and determine the final release volume of shared electric motorcycles.

[0053] The beneficial effects of the present invention are:

[0054] (1) Based on the supply and demand status of the number of shared electric motorcycles, the present invention divides the traffic area into a reasonable deployment area, an oversaturated deployment area, and an undersaturated deployment area, and calculates the deployment amount for different deployment areas, so that the deployment results are more reasonable;

[0055] (2) The present invention uses the electronic fence data of shared electric motorcycles to calculate their parking capacity based on the daily average turnover rate of shared electric motorcycles, making the calculation result more accurate;

[0056] (3) The calculation method of the present invention is simple and easy to understand, highly operational, and suitable for decision-making and deployment by the government and related departments. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a schematic diagram of the process of the present invention;

[0058] Figure 2 This is the average daily turnover rate of shared electric motorcycles. Figure 2 (a) is a working day, Figure 2 (b) is a weekend;

[0059] Figure 3 This is the residual graph of shared electric motorcycle borrowing and returning. Figure 3 (a) To borrow a vehicle, Figure 3 (b) to return the vehicle;

[0060] Figure 4 This is a map of the types of areas where shared electric motorcycles are deployed. Figure 4 (a) is the number of vehicles loaned out, Figure 4 (b) The number of vehicles returned;

[0061] Figure 5 This is the static load-bearing capacity distribution diagram of shared electric motorcycles;

[0062] Figure 6 This is the distribution map of the parking capacity of shared electric motorcycles. Figure 6 (a) is a working day, Figure 6 (b) is the weekend. DETAILED DESCRIPTION

[0063] 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 only 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.

[0064] Example 1

[0065] A method for calculating the quantity of shared electric motorcycles based on supply and demand balance, such as Figure 1 As shown, the following steps are included:

[0066] S1. Build a database for calculating the number of shared electric motorcycles deployed. The database information includes shared electric motorcycle order data, shared electric motorcycle electronic fence data, urban point of interest (POI) data, and traffic community data.

[0067] Urban POI data includes three types of information: socio-economic attributes, transportation facilities, and land use.

[0068] Socioeconomic attribute information includes: population size and per capita GDP.

[0069] Transportation facility information includes: the number of bus stops, the number of subway stations, and the number of transportation facility services.

[0070] Land use information includes: the number of healthcare services, the number of restaurant services, and the number of shopping services.

[0071] S2. Data preprocessing: Clean the shared electric motorcycle order data in the database of step S1 and remove invalid and abnormal data.

[0072] Furthermore, the removal of invalid data and abnormal data in step S2 includes:

[0073] S21. Incomplete records of shared motorcycle order data information. The shared motorcycle order data information includes: vehicle ID, user ID, borrowing date, borrowing time, borrowing longitude, borrowing latitude, return date, return time, return longitude, return latitude;

[0074] S22, for a record where the travel distance is less than the minimum travel distance threshold Y, the travel distance is obtained by calculating the Euclidean distance between the longitude and latitude of the returned vehicle and the longitude and latitude of the borrowed vehicle;

[0075] S23, the duration of a trip is less than the minimum travel time threshold U min and greater than the maximum travel time threshold U of the shared electric bike max The trip duration can be obtained by subtracting the borrowing time from the return time on the same date.

[0076] S24, the average speed of a trip is less than the minimum travel speed threshold V min and greater than the maximum travel speed threshold V of the shared electric motorcycle max Based on the records of travel, the average travel speed can be obtained by the ratio of travel distance to travel time.

[0077] S3. Statistics on the usage characteristics of shared electric motorcycles: Taking the traffic area as the statistical unit, the average daily borrowing volume, return volume and average daily turnover rate of shared electric motorcycles are counted respectively.

[0078] Furthermore, the method for calculating the usage characteristics of shared electric motorcycles in step S3 includes:

[0079] S31. Average daily borrowing volume of shared electric motorcycles: determine the traffic zone to which the shared electric motorcycles belong based on the borrowing latitude and longitude in the order data information, and count the average daily borrowing volume in each traffic zone;

[0080] S32, average daily return volume of shared electric motorcycles: determine the traffic zone to which the shared electric motorcycles belong based on the longitude and latitude of the return in the order data information, and count the average daily return volume in each traffic zone;

[0081] S33. Average daily turnover rate of shared electric motorcycles: The average daily turnover rate of shared electric motorcycles is calculated based on the order data and the total number of vehicles in the traffic area. The calculation formula is as follows:

[0082]

[0083] in, is the average daily turnover rate of shared electric bikes in the ith traffic zone, Q mi is the total number of valid shared motorcycle orders in the i-th traffic zone on the m-th day, S mi is the total number of shared electric motorcycles in the i-th traffic zone on the m-th day, n is the number of traffic zones, and T is the total number of days.

[0084] S4. Analysis of factors affecting the borrowing and returning of shared electric motorcycles: Taking the traffic community as the analysis unit, regression models for the borrowing and returning of shared electric motorcycles are established respectively. The residual δ1 of the number of borrowed and returned shared electric motorcycles in each traffic community is obtained by solving the model.

[0085] Furthermore, in step S4, the traffic community is used as the analysis unit, the number of shared electric motorcycles borrowed and returned is used as the dependent variable of the model, and the city's point of interest POI data is used as the explanatory variable of the model, and regression prediction models for the number of shared electric motorcycles borrowed and returned are established respectively; the model is solved to obtain the significance level of the explanatory variables and the residual δ1 of the number of shared electric motorcycles borrowed and returned in each traffic community, and the explanatory variables with a significance level ≤0.1 are retained as influencing variables, and the residual δ1 is equal to the actual number of shared electric motorcycles borrowed and returned minus the value obtained by the regression model.

[0086] Furthermore, the established regression prediction model includes a linear regression model, a Poisson regression model or a geographically weighted regression model.

[0087] Furthermore, before establishing a regression model, the explanatory variables need to be tested for multicollinearity. When the variance inflation factor (VIF) value of the explanatory variables is less than 10, it indicates that there is no multicollinearity between the explanatory variables and they can be used for modeling. Otherwise, multicollinearity exists and the explanatory variables with multicollinearity need to be eliminated before modeling can be carried out.

[0088] S5. Classification of shared electric motorcycle deployment area types: Based on the residual δ1 obtained in step S4, the traffic community is divided into three types according to the supply and demand status of shared electric motorcycle deployment: reasonable deployment area, under-saturated deployment area and over-saturated deployment area; based on the three-type division, the influencing variables retained after solving the regression model in step S4 are used as subsequent explanatory variables and divided into explanatory variables for reasonable deployment area, explanatory variables for under-saturated deployment area and explanatory variables for over-saturated deployment area.

[0089] The basis for dividing the reasonable supply area, undersaturated supply area and oversaturated supply area is:

[0090] When the residual δ1 <D min It is the undersaturated zone;

[0091] When D min ≤δ1 <D max It is a reasonable area for placement;

[0092] When the residual δ1≥D max It is the oversaturation zone;

[0093] Among them, D min , D max They represent the minimum and maximum reasonable zone thresholds respectively.

[0094] S6. Determine the reasonable deployment coefficient of shared electric motorcycles: Establish a regression model for the borrowing and returning volume of shared electric motorcycles in reasonable deployment areas, and solve the model to obtain the reasonable deployment coefficient α of shared electric motorcycles.

[0095] Taking the reasonable deployment area as the analysis unit, a regression model of the shared electric motorcycle borrowing and returning volume in the reasonable deployment area is established. The borrowing and returning volume of shared electric motorcycles are taken as the dependent variables of the model, and the influencing variables in step S4 are taken as the explanatory variables of the model. The model is solved to obtain the reasonable deployment coefficient α of shared electric motorcycles.

[0096] S7. Prediction of the total gap in the deployment of shared electric motorcycles: Taking the undersaturated traffic area as the prediction unit, multiply the reasonable deployment coefficient α in step S6 by the explanatory variable of the undersaturated area to predict the demand for shared electric motorcycles in the undersaturated area.

[0097] Then, the difference between the predicted demand and the actual demand is divided by the average daily turnover rate of each traffic zone. Obtain the gap in the deployment of shared electric motorcycles in each traffic community in the under-saturated area;

[0098] Sum up the gaps in all traffic communities to get the total gap in shared electric motorcycle deployment.

[0099] The calculation formula for the total shortage of shared electric motorcycles in step S7 is as follows:

[0100]

[0101] Among them, y is the amount of additional deployment that should be made in under-saturated traffic areas, and y i is the amount of traffic that should be put into the ith traffic zone, y m is the actual delivery volume of the mth traffic zone, is the average daily turnover rate of shared electric motorcycles, and n is the number of traffic communities.

[0102] S8. Correction of the gap in the supply of shared electric motorcycles: Based on the total supply gap predicted in step S7, the supply gap is corrected in combination with the potential transfer volume of other modes of transportation and the actual demand in the oversaturated areas.

[0103] The transfer volume of other modes of transportation mainly includes: 1) With the elimination of non-new national standard electric bicycles, a certain proportion of non-new national standard electric vehicle users will transfer to shared electric motorcycles, and the transfer ratio needs to be determined according to the specific situation of the city; 2) There is a certain replacement ratio of fixed-pole public bicycles being replaced with shared electric motorcycles, but the replacement ratio needs to be determined according to the specific situation of the city.

[0104] Actual demand in oversaturated areas: Taking oversaturated traffic areas as prediction units, multiply the reasonable deployment coefficient α in step S6 by the explanatory variable information of the oversaturated area to obtain the actual demand for shared electric motorcycles in the oversaturated area. According to the actual demand for shared bicycles in the oversaturated area, appropriately reduce or stop increasing the deployment in the area.

[0105] S9. Shared motorcycle parking capacity measurement: Taking the traffic area as the measurement unit, count the number of shared motorcycle electronic fences in step S1, and use the number of electronic fences to calculate the static parking capacity P of shared motorcycles. i ; Using the static parking capacity P i Combined with the average daily turnover rate of shared electric motorcycles Calculating the parking capacity of shared electric motorcycles

[0106] The calculation formula for the static parking capacity of shared electric motorcycles is as follows:

[0107]

[0108] Among them, i is the traffic area number, P i is the static bearing capacity of traffic zone i, N i is the number of shared bicycle electronic fences in traffic community i, The average number of shared motorcycles parked per electronic fence;

[0109] The calculation formula for the parking capacity of shared electric motorcycles is as follows:

[0110]

[0111] in, To provide the parking capacity for shared electric motorcycles, is the average daily turnover rate of shared electric motorcycles.

[0112] S10, determine the number of shared electric motorcycles to be deployed: the shared electric motorcycle parking capacity obtained in step S9 is Compare it with the predicted demand M for shared electric motorcycles to obtain the final supply of shared electric motorcycles.

[0113] Determining the number of shared electric motorcycles to be put into use includes the following steps:

[0114] S101. Add the current supply of shared electric motorcycles to the supply gap to obtain the predicted demand M for shared electric motorcycles;

[0115] S102. Compare the demand M with the parking capacity of shared electric motorcycles For comparison, if Then take M as the final delivery amount; if Then take As the final delivery amount.

[0116] Based on the analysis of the rationality of the current deployment of shared electric motorcycles, the present invention provides a method for calculating the deployment amount of shared electric motorcycles in cities. The method is simple and easy to understand, and has strong operability, and can provide guidance for the deployment of shared electric motorcycles in cities.

[0117] Example 2

[0118] Combined with the current operating status of a city's shared electric motorcycle system and actual shared electric motorcycle order data in a city, a method for calculating the amount of shared electric motorcycles based on supply and demand balance of the present invention is adopted, which specifically includes the following steps:

[0119] (1) Constructing a database for calculating the number of shared electric motorcycles

[0120] As of May 2022, the total number of shared bicycles (including shared electric motorcycles) in a certain city is 125,000, and there are more than 200 parking spots. This case uses the shared electric motorcycle order data of a certain city from April 20 to May 20, 2022 (a total of 3,490,699 orders), and selects the data for one week without special circumstances such as holidays and heavy rain for operation. The data covers the five administrative districts of the city, which are clustered with various land use functions such as residential, work areas, schools, large commercial areas, and transportation hubs. The demand for shared electric motorcycle riding is large, indicating that the data is representative.

[0121] The constructed shared electric motorcycle deployment measurement database includes shared electric motorcycle order data, shared electric motorcycle electronic fence data, point of interest POI data and traffic community data in a certain city. Among them, the specific contents of shared electric motorcycle electronic fence data, urban point of interest POI data and traffic community data are shown in Table 1. Specifically, the shared electric motorcycle electronic fence data includes: the number and area of ​​electronic fences; the urban point of interest POI data includes three types of information: socio-economic attributes, transportation facilities and land use; socio-economic attribute information includes population density and GDP per capita; transportation facility information includes bus station density, number of subway stations and transportation facility service density; land use information includes medical care service density, catering service density and shopping service quantity. In addition, this case divides the main urban area of ​​a certain city into 717 traffic communities, each of which contains: traffic community ID, traffic community area, and traffic community centroid latitude and longitude coordinates.

[0122] Table 1 Data information example (partial)

[0123]

[0124] Table 1 continued

[0125]

[0126]

[0127] (2) Data preprocessing

[0128] After the shared electric motorcycle order data is cleaned and processed, the following travel record table is formed, as shown in Table 2. A complete travel record consists of 10 parts: order ID, user ID, borrowing date, borrowing time, borrowing longitude, borrowing latitude, return date, return time, return longitude, and return latitude.

[0129] Table 2 Shared electric motorcycle travel data record table (part)

[0130]

[0131] Table 2 continued

[0132]

[0133] (3) Statistics on the usage characteristics of shared electric motorcycles

[0134] Taking the traffic community as the statistical unit, the average daily number of shared electric motorcycles borrowed and returned, as well as the average daily turnover rate in a certain city were counted, and ArcGIS software was used to visualize the statistical data of the average daily turnover rate on weekdays and weekends, as shown in the figure below. Figure 2 shown.

[0135] (4) Analysis of factors affecting the borrowing and returning of shared electric motorcycles

[0136] The regression modeling used in the specific implementation example of this method is a linear regression model. Taking the traffic community as the analysis unit, a linear regression model of the borrowing and returning volume of shared electric motorcycles is established respectively: the borrowing and returning volume of shared electric motorcycles is used as the dependent variable of the model, and the urban point of interest POI data in the database of step (1) is used as the explanatory variable of the model.

[0137] (4.1) Before establishing a linear regression model, in order to avoid high correlation between explanatory variables in the model, which may lead to distortion or inaccuracy of model estimation results, it is necessary to first perform a multicollinearity test on the explanatory variables. SPSS software was used to test the multicollinearity between variables. The results showed that the VIF values ​​of the explanatory variables were all less than 10, and there was no multicollinearity between the variables, so regression analysis could be performed.

[0138] (4.2) Linear regression models for the borrowing and returning of shared electric bikes were established respectively. The model results are shown in Table 3, and only explanatory variables with a significance level of ≤ 0.1 were retained as influencing variables. δ1 It is equal to the actual borrowing (returning) volume of shared electric bikes minus the value obtained by the regression model. The residual of each traffic zone is obtained based on the established regression model. δ1 , using ArcGIS software to visualize the data, such as Figure 3 shown.

[0139] Table 3. Regression model results

[0140]

[0141]

[0142] (5) Classification of delivery areas

[0143] According to the residual δ1 obtained in step (4), the traffic areas are divided into three types according to the supply and demand status of shared electric motorcycles: reasonable deployment area, undersaturated deployment area and oversaturated deployment area. Specifically, through analysis, the traffic areas with δ1<-0.9 are divided into undersaturated deployment areas, the traffic areas with -0.9≤δ1<-1.22 are divided into reasonable deployment areas, and the traffic areas with δ1≥1.22 are divided into oversaturated deployment areas. The results of the deployment area type classification are visualized using ArcGIS software, as shown in Figure 4 shown.

[0144] (6) Determine the reasonable deployment coefficient of shared electric motorcycles

[0145] Taking the traffic community as the analysis unit, a regression model for the borrowing and returning volume of shared electric motorcycles in the reasonable deployment area is established. The borrowing and returning volume of shared electric motorcycles are taken as the dependent variables of the model, and the influencing variables in step (4) are taken as the explanatory variables of the model. The model is solved to obtain the reasonable deployment coefficient of shared electric motorcycles. The results are shown in Table 4.

[0146] Table 4 Results of regression model for reasonable area placement

[0147] variable B t Significance VIF constant 755.9 1.320 0.187 <![CDATA[Population density (persons / km 2 )]]> 0.298 4.129 0.000 1.254 GDP per capita (yuan / person) -0.006 -1.288 0.198 1.334 <![CDATA[Bus stop density (stops / km 2 )]]> 9.065 0.644 0.520 1.514 Number of subway stations 578.693 3.044 0.002 1.031 <![CDATA[Medical and healthcare service density (number / km 2 )]]> 14.604 5.321 0.000 1.991 <![CDATA[Traffic facility service density (number / km 2 )]]> 13.192 4.210 0.000 2.974 <![CDATA[Food service density (number / km 2 )]]> -0.159 -0.297 0.766 2.417 Number of shopping services 7.428 10.120 0.000 1.413

[0148] (7) Forecasting the gap in the number of shared electric motorcycles and correcting the gap

[0149] (7.1) Sharing motorcycle deployment gap: Multiply the POI information of the city's under-saturated traffic areas by the reasonable deployment coefficient to obtain the demand for sharing motorcycles in the under-saturated areas. Then, divide the difference between the demand and the actual deployment by the average daily turnover rate of shared motorcycles. The gap in the number of shared electric motorcycles in each traffic area in the undersaturated area is obtained, and the sum of the gaps in all traffic areas is the total gap in the number of shared electric motorcycles. The formula is as follows:

[0150]

[0151] Among them, y is the amount of additional deployment that should be made in under-saturated traffic areas, and y i is the amount of traffic that should be put into the ith traffic zone, y m is the actual delivery volume of the mth traffic zone, is the average daily turnover rate of shared electric motorcycles, and n is the number of traffic communities.

[0152] Specifically, the average daily turnover rate of shared electric motorcycles in a city is calculated according to the formula is 5.15. It can be seen from Table 4 that the regression model constant of the reasonable area is 755.9 and the population density coefficient is 0.298. Taking the traffic community No. 172 in the under-deployed area as a prediction example, the population density of the traffic community is 1141.4, the per capita GDP is 113352, the bus station density is 3.0, the number of subway stations is 0, the medical care service density is 0, the transportation facility service density is 0, the catering service density is 3.0, and the number of shopping services is 0. It can be calculated that the amount of investment that should be made in this under-deployed area at this stage is:

[0153] y=(1141.4×0.298+113352×(-0.006)+3×9.065+3×(-0.159)+755.9-32) / 5.15=79.7

[0154] Therefore, 797,000 shared electric motorcycles should be put into use in Traffic Zone No. 172. The sum of the number of shared electric motorcycles that should be put into use in all traffic zones is the current gap in the number of shared electric motorcycles, and finally the gap in the number of shared electric motorcycles in a certain city is 28,200.

[0155] (7.2) Shared electric motorcycles correction gap: On the one hand, the current utilization rate of public bicycles is extremely low and cannot meet market demand. It is recommended that all these public bicycles be replaced with shared electric motorcycles; on the other hand, with the elimination of non-new national standard electric vehicles, some citizens who use non-new national standard electric vehicles may switch to shared electric motorcycles. In addition, the supply in oversaturated areas should be appropriately reduced or not increased.

[0156] Specifically, as of May 2022, the total number of public bicycles in a certain city is 15,000, but the annual turnover rate of public bicycles is only 0.83 times. It is recommended to replace all these public bicycles with shared electric motorcycles. On the other hand, according to the safety technical specifications for electric bicycles, there are currently 1.58 million electric vehicles in a certain city that do not meet the new national standard for electric vehicles. With the elimination of non-new national standard electric vehicles, it is conservatively estimated that about 1% of citizens who use non-new national standard electric vehicles may transfer to shared electric motorcycles, so it is necessary to increase the number of shared bicycles by about 16,000 to meet the transfer needs of these citizens. In summary, the total correction gap of shared electric motorcycles in a certain city at this stage is 59,200.

[0157] (8) Measurement of shared electric motorcycle parking capacity

[0158] Taking the traffic zone as the measurement unit, count the number of electronic fences in step (1), multiply the number of electronic fences by the average number of parked vehicles in each electronic fence to obtain the static parking carrying capacity of shared electric motorcycles, multiply the static parking carrying capacity by the average parking turnover rate of each traffic zone and then sum them to obtain the parking carrying capacity of shared electric motorcycles.

[0159] (8.1) Static parking capacity

[0160] The survey shows that the average number of shared bicycles that can be parked in each electronic fence is 15. Therefore, without considering the turnover rate, the calculation formula for the static bearing capacity of a city's electronic fence is:

[0161]

[0162] Among them, i is the traffic area number, P i is the static bearing capacity of traffic zone i, N i is the number of shared bicycle electronic fences in traffic community i, is the average number of shared bicycles parked in each electronic fence. Based on the calculation results, ArcGIS software can be used to obtain the static carrying capacity distribution map of shared electric bicycles based on traffic communities in a certain city, such as Figure 5 shown.

[0163] (8.2) Calculation of parking capacity

[0164] The calculation formula for the parking capacity of shared electric motorcycles is:

[0165]

[0166] in, To provide the parking capacity for shared electric motorcycles, For the i The average daily turnover rate of shared electric motorcycles in the traffic community.

[0167] The parking capacity of shared electric motorcycles in each traffic zone on weekdays and weekends is calculated, and the calculation results are displayed using ArcGIS software. Figure 6 The parking capacity of shared electric motorcycles can be obtained by summing up the parking capacity of each traffic zone. According to calculations, the parking capacity of shared electric motorcycles in a certain city is 211,000, which means that the maximum number of shared electric motorcycles that can be accommodated is 211,000.

[0168] (9) Determination of the estimated number of shared electric motorcycles

[0169] The shared motorcycle parking capacity obtained in step (8) Compare with the predicted number of shared electric motorcycles M to determine the final number of shared electric motorcycles.

[0170] (9.1) Demand for delivery

[0171] The total demand for shared electric bikes is obtained by adding the current supply of shared electric bikes to the supply gap. Specifically, at present, the total number of shared bikes (including shared electric bikes) in a certain city is 125,000, but there is a supply gap of 59,200 shared electric bikes (taking into account the predicted demand for shared bikes, the demand for replacement of public bikes, and the demand generated by some citizens turning to shared bikes after the gradual elimination of non-new national standard electric bikes). The total demand for shared bikes M is between 184,200 and 184,200.

[0172] (9.2) Final delivery volume determination

[0173] The demand M and the parking capacity of shared electric motorcycles For comparison, if Then take M as the final delivery amount; if Then take As the final deployment quantity. Specifically, according to step (9.1), the total deployment demand M of shared electric motorcycles in a certain city is between 184,200 vehicles; according to step (8.2), the parking capacity of shared electric motorcycles in a certain city is Therefore, the final number of shared electric motorcycles in a certain city is determined to be 184,200.

[0174] Through the above steps, this method predicts the demand for borrowing and returning shared electric motorcycles on the basis of determining the parking capacity of shared electric motorcycles in the city, and gives the appropriate amount of shared electric motorcycles. It has good practicality, accuracy and scalability.

[0175] Example 3

[0176] A system for calculating the quantity of shared electric motorcycles to be put into use based on supply-demand balance comprises a database construction module, a data set preprocessing module, a usage feature statistics module, a borrowing and returning influencing factor analysis module, a put area type classification module, a reasonable put coefficient determination module, a total put gap prediction module, a put gap correction module, a parking carrying capacity measurement module and a shared electric motorcycle put amount determination module.

[0177] Database construction module: used to build a database for calculating the number of shared electric motorcycles deployed.

[0178] Dataset preprocessing module: used to clean the shared electric motorcycle order data in the database and remove invalid and abnormal data.

[0179] Use feature statistics module: used to count the average daily borrowing and returning volume of shared electric motorcycles and the average daily turnover rate

[0180] Borrowing and Returning Factor Analysis Module: used to solve the regression model of the borrowing and returning of shared electric motorcycles, and obtain the residual of the borrowing and returning number of shared electric motorcycles in each traffic community δ1.

[0181] The module for classifying deployment areas is used to classify traffic communities into three types according to the supply and demand status of shared electric motorcycles: reasonable deployment areas, under-saturated deployment areas, and over-saturated deployment areas.

[0182] Reasonable deployment coefficient determination module: used to solve the regression model of the shared electric motorcycle borrowing and returning volume in the reasonable deployment area, and obtain the reasonable deployment coefficient α of shared electric motorcycles.

[0183] Total gap quantity prediction module: used to predict the total gap quantity of shared electric motorcycles.

[0184] Delivery gap correction module: used to correct delivery gap.

[0185] Parking capacity measurement module: used to measure the static parking capacity P of shared electric motorcycles i And use the static parking capacity P i Calculating the parking capacity of shared electric motorcycles

[0186] Shared electric motorcycle deployment determination module: used to determine the parking capacity of shared electric motorcycles Compare with the predicted demand M for shared electric motorcycles and determine the final release volume of shared electric motorcycles.

[0187] The system of the present invention divides the traffic area into a reasonable deployment area, an oversaturated deployment area and an undersaturated deployment area based on the supply and demand status of the deployment of shared electric motorcycles, and calculates the deployment amount for different deployment areas to make the deployment results more reasonable.

[0188] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for calculating the quantity of shared electric motorcycles based on supply and demand balance, characterized in that: The steps include: S1. Build a database for calculating the number of shared electric motorcycles put into use. The database information includes shared electric motorcycle order data, shared electric motorcycle electronic fence data, urban point of interest POI data and traffic community data; The city POI data includes three types of information: socio-economic attributes, transportation facilities and land use; among which socio-economic attribute information includes population and GDP per capita; transportation facility information includes the number of bus stops, the number of subway stations and the number of transportation facility services; land use information includes the number of medical care services, the number of catering services and the number of shopping services; S2. Data preprocessing: Clean the shared electric motorcycle order data in the database of step S1 and remove invalid and abnormal data; S3. Statistics on the usage characteristics of shared electric motorcycles: Taking the traffic area as the statistical unit, the average daily borrowing volume, return volume and average daily turnover rate of shared electric motorcycles are counted respectively. S4. Analysis of influencing factors of shared electric motorcycle borrowing and returning: Taking the traffic community as the analysis unit, respectively establish regression models for the borrowing and returning of shared electric motorcycles, and solve the model to obtain the residual δ1 of the number of shared electric motorcycles borrowed and returned in each traffic community; specifically including: taking the traffic community as the analysis unit, taking the borrowing and returning of shared electric motorcycles as the dependent variables of the model, and taking the urban point of interest POI data as the explanatory variables of the model, respectively establish regression prediction models for the borrowing and returning of shared electric motorcycles; solve the model to obtain the significance level of the explanatory variables and the residual δ1 of the number of shared electric motorcycles borrowed and returned in each traffic community, retain the explanatory variables with a significance level ≤ 0.1 as the influencing variables, and the residual δ1 is equal to the actual borrowing and returning of shared electric motorcycles minus the value obtained by the regression model; S5. Classification of shared electric motorcycle deployment area types: Based on the residual δ1 obtained in step S4, the traffic community is divided into three types according to the supply and demand status of shared electric motorcycle deployment: reasonable deployment area, undersaturated deployment area, and oversaturated deployment area; Based on the three types of division, the influencing variables retained after solving the regression model in step S4 are used as subsequent explanatory variables and divided into reasonable deployment area explanatory variables, undersaturated deployment area explanatory variables, and oversaturated deployment area explanatory variables; S6. Determine the reasonable deployment coefficient of shared electric motorcycles: Establish a regression model for the borrowing and returning of shared electric motorcycles in reasonable deployment areas, and solve the model to obtain the reasonable deployment coefficient α of shared electric motorcycles; S7. Forecasting the total shortage of shared electric motorcycles: Taking the undersaturated traffic area as the prediction unit, multiply the reasonable deployment coefficient α in step S6 by the explanatory variable of the undersaturated area to predict the demand for shared electric motorcycles in the undersaturated area; Then, the difference between the predicted demand and the actual demand is divided by the average daily turnover rate of each traffic zone. Obtain the gap in the deployment of shared electric motorcycles in each traffic community in the under-saturated area; Sum up the gaps in all traffic areas to get the total gap in shared electric motorcycles. S8. Correction of the gap in the supply of shared electric motorcycles: Based on the total supply gap predicted in step S7, the supply gap is corrected in combination with the potential transfer volume of other modes of transportation and the actual demand in the oversaturated area; S9. Shared motorcycle parking capacity measurement: Taking the traffic area as the measurement unit, count the number of shared motorcycle electronic fences in step S1, and use the number of electronic fences to calculate the static parking capacity P of shared motorcycles. i ; Using the static parking capacity P i Combined with the average daily turnover rate of shared electric motorcycles Calculating the parking capacity of shared electric motorcycles S10, determine the number of shared electric motorcycles to be deployed: the shared electric motorcycle parking capacity obtained in step S9 is Compare it with the predicted demand M for shared electric motorcycles to obtain the final supply of shared electric motorcycles.

2. The method for calculating the quantity of shared electric motorcycles based on supply and demand balance according to claim 1 is characterized in that: The method for statistically analyzing the usage characteristics of shared electric motorcycles in step S3 includes: S31. Average daily borrowing volume of shared electric motorcycles: determine the traffic zone to which the shared electric motorcycles belong based on the borrowing latitude and longitude in the order data information, and count the average daily borrowing volume in each traffic zone; S32, average daily return volume of shared electric motorcycles: determine the traffic zone to which the shared electric motorcycles belong based on the longitude and latitude of the return in the order data information, and count the average daily return volume in each traffic zone; S33. Average daily turnover rate of shared electric motorcycles: The average daily turnover rate of shared electric motorcycles is calculated based on the order data and the total number of vehicles in the traffic area. The calculation formula is as follows: in, is the average daily turnover rate of shared electric bikes in the ith traffic zone, Q mi is the total number of valid shared motorcycle orders in the i-th traffic zone on the m-th day, S mi is the total number of shared electric motorcycles in the i-th traffic zone on the m-th day, n is the number of traffic zones, and T is the total number of days.

3. The method for calculating the quantity of shared electric motorcycles based on supply and demand balance according to claim 1 is characterized in that: The regression prediction model includes a linear regression model, a Poisson regression model or a geographically weighted regression model.

4. The method for calculating the quantity of shared electric motorcycles based on supply and demand balance according to claim 1 is characterized in that: In step S5, the basis for dividing the reasonable delivery area, the undersaturated delivery area and the oversaturated delivery area is: When the residual δ1 <D min It is the undersaturated zone; When D min ≤δ1 <D max It is a reasonable area for placement; When the residual δ1≥D max It is the oversaturation zone; Among them, D min , D max They represent the minimum and maximum reasonable zone thresholds respectively.

5. The method for calculating the quantity of shared electric motorcycles based on supply and demand balance according to claim 1 is characterized in that: The calculation formula for the total shortage of shared electric motorcycles in step S7 is as follows: Among them, y is the amount of additional deployment that should be made in under-saturated traffic areas, and y i is the amount of traffic that should be put into the ith traffic zone, y m is the actual delivery volume of the mth traffic zone, is the average daily turnover rate of shared electric motorcycles, and n is the number of traffic communities.

6. The method for calculating the quantity of shared electric motorcycles based on supply and demand balance according to claim 1 is characterized in that: In step S9, the static parking carrying capacity of the shared electric motorcycle is calculated using the following formula: Among them, i is the traffic area number, P i is the static bearing capacity of traffic zone i, N i is the number of shared bicycle electronic fences in traffic community i, The average number of shared motorcycles parked per electronic fence; The formula for calculating the parking capacity of shared electric motorcycles is as follows: in, To provide the parking capacity for shared electric motorcycles, is the average daily turnover rate of shared electric motorcycles.

7. The method for calculating the quantity of shared electric motorcycles based on supply and demand balance according to claim 1 is characterized by: Determining the number of shared electric motorcycles in step S10 includes the following steps: S101. Add the current supply of shared electric motorcycles to the supply gap to obtain the predicted demand M for shared electric motorcycles; S102. Compare the demand M with the parking capacity of shared electric motorcycles For comparison, if Then take M as the final delivery amount; if Then take As the final delivery amount.

8. A calculation system for the method for calculating the quantity of shared electric motorcycles based on supply and demand balance according to any one of claims 1 to 7, characterized in that: The calculation system includes a database construction module, a data set preprocessing module, a usage feature statistics module, a borrowing and returning influencing factor analysis module, a deployment area type classification module, a reasonable deployment coefficient determination module, a total deployment gap prediction module, a deployment gap correction module, a parking carrying capacity measurement module and a shared electric motorcycle deployment quantity determination module.

9. The measuring system according to claim 8, characterized in that: The database construction module is used to construct a database for calculating the number of shared electric motorcycles deployed; The data set preprocessing module is used to clean the shared motorcycle order data in the database and remove invalid data and abnormal data; The usage feature statistics module is used to count the average daily borrowing and returning volume of shared electric motorcycles and the average daily turnover rate. The borrowing and returning influencing factor analysis module is used to solve the regression model of the borrowing and returning quantity of shared electric motorcycles, and obtain the residual δ1 of the borrowing and returning quantity of shared electric motorcycles in each traffic community; The deployment area type classification module is used to divide the traffic area into three types according to the supply and demand status of shared electric motorcycles: a reasonable deployment area, an undersaturated deployment area, and an oversaturated deployment area; The reasonable deployment coefficient determination module is used to solve the regression model of the borrowing and returning quantity of shared electric motorcycles in the reasonable deployment area to obtain the reasonable deployment coefficient α of shared electric motorcycles; The total gap quantity prediction module is used to predict the total gap quantity of shared electric motorcycles. The delivery gap correction module is used to correct the delivery gap; The parking capacity measurement module is used to measure the static parking capacity P of the shared electric motorcycle. i And use the static parking capacity P i Calculating the parking capacity of shared electric motorcycles The shared electric motorcycle deployment determination module is used to determine the shared electric motorcycle parking capacity. Compare with the predicted demand M for shared electric motorcycles and determine the final release volume of shared electric motorcycles.

Images