A vehicle intelligent dispatching and sharing management system based on big data analysis

Abstract

The present invention relates to the field of intelligent transportation, and discloses a vehicle intelligent dispatching and sharing management system based on big data analysis, comprising: a data acquisition module and a data analysis module, which analyze vehicle use data and user data through a vehicle demand analysis unit to calculate and obtain a vehicle demand index for each area; a traffic condition prediction unit, which predicts future traffic conditions based on historical traffic data and real-time traffic data through a pre-trained traffic condition prediction model, and responds with an output traffic congestion index; an intelligent dispatching module, which is used to judge whether there is a risk of redundancy or shortage of vehicles in each area according to the vehicle demand index and the traffic congestion index, and if there is a risk of redundancy, the redundant vehicles are transferred out, and if there is a risk of shortage, the insufficient vehicles are transferred in; and a sharing management module, which establishes a vehicle sharing database based on the data collected by the data acquisition module, and is used for tracing back the vehicle use records.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation, and in particular to a vehicle intelligent dispatching and sharing management system based on big data analysis. Background Art

[0002] With the continuous development of science and technology, big data technology has increasingly become one of the important tools to promote social progress and improve life. Especially in the field of transportation, with the continuous increase in the number of vehicles, traditional vehicle dispatching and management methods have been unable to meet the needs of modern cities for efficient, intelligent and convenient transportation services. Therefore, the intelligent vehicle dispatching and sharing management system based on big data analysis came into being, providing new ideas and solutions for solving urban traffic problems.

[0003] For example, sharing bicycles has made people's lives more convenient. However, in the actual process of bicycle vehicle dispatching, vehicles are often allocated and dispatched in each area based on historical data or experience, while ignoring the importance of real-time data analysis and dynamic adjustment. Although this method can alleviate the problem of uneven vehicle distribution to a certain extent, it often seems powerless when faced with emergencies or special events. For example, during holidays or special events, the number of tourists in a certain area may surge, resulting in a sharp increase in the demand for shared bicycles in that area. At this time, if the existing vehicle dispatching method is still used, it is obviously unable to meet the surging user demand. Summary of the invention

[0004] The purpose of the present invention is to provide a vehicle intelligent dispatching and sharing management system based on big data analysis, so as to solve the problem of vehicle allocation and dispatching in each area based only on historical data or experience, while ignoring the importance of real-time data analysis and dynamic adjustment. Although it can alleviate the problem of uneven distribution of vehicles to a certain extent, it is often a technical problem that it is powerless when faced with emergencies or special events.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A vehicle intelligent dispatching and sharing management system based on big data analysis, including:

[0007] Data collection module, used to collect vehicle usage data, user data and traffic data;

[0008] Data analysis module, including vehicle demand analysis unit and traffic condition prediction unit;

[0009] The vehicle demand analysis unit analyzes vehicle usage data and user data to identify vehicle changes in different time periods and regions;

[0010] Obtain the travel habits of users in each area, and calculate the user travel index of each area within a working cycle. Based on the vehicle changes and user travel index, calculate the vehicle demand index of each area;

[0011] The traffic condition prediction unit predicts the future traffic conditions based on historical traffic data and real-time traffic data through a pre-trained traffic condition prediction model, and responds with the output of traffic congestion indicators;

[0012] The intelligent scheduling module is used to determine whether there is a risk of redundancy or shortage of vehicles in each area based on vehicle demand indicators and traffic congestion indicators. If there is a risk of redundancy, the redundant vehicles will be transferred out; if there is a risk of shortage, the insufficient vehicles will be transferred in.

[0013] As a technical solution, the process of judging whether there is a risk of redundancy or shortage of vehicles in each area based on vehicle demand indicators and traffic congestion indicators is as follows:

[0014] Get vehicle demand indicators for each area at different times and traffic congestion indicators ;

[0015] Obtain the flow of people in each area based on surveillance cameras ;

[0016] Will , Substituting into the formula: Calculate the scheduling demand index ;

[0017] The scheduling demand index obtained by calculation Vehicle Demand Index Make comparisons;

[0018] like , it is determined that there is a risk of insufficient vehicles in the current area, and vehicles are transferred to the current area;

[0019] like , it is determined that there is a redundancy risk for vehicles in the current area, and vehicles are dispatched from the current area;

[0020] In the formula For the Regions, As a function, the current period is divided into multiple sub-periods at equal intervals. If the traffic congestion index in three consecutive sub-periods increases or the traffic congestion index exceeds the preset traffic congestion warning value, then ,otherwise, ;

[0021] , The preset reference difference, is the curve of the flow of people in the current area changing over time, The reference curve of the flow of people in the current area over time, ~ For the detection period, is the exponential coefficient, , , is the conversion factor.

[0022] As a technical solution, the exponential coefficient The expression is: ;

[0023] In the formula, , is the weight factor, is the maximum value of traffic congestion index in each sub-period, is the minimum value of traffic congestion index in each sub-period, is the curve of traffic congestion index changing over time in each sub-period. For any sub-period, is the total number of sub-periods, , The start and end time of any sub-period.

[0024] As a technical solution, the vehicle demand index The calculation formula is:

[0025] ; In the formula For the The influence coefficient of the region, For the The vehicle variation coefficient of the area, No. User travel indicators for each region, , is the preset proportional coefficient;

[0026] The said Vehicle variation coefficient in each area The expression is: ; In the formula, For the The value of the vehicle usage data item, is the total number of vehicle usage data items, , are the starting point and end point of each time period, For the The curve of vehicle usage data changing over time, For the Reference curve of vehicle usage data changing over time, is the impact factor for each period, is the total number of time periods, For the time period.

[0027] As a technical solution, obtaining the User travel indicators for each region The process is:

[0028] Get the number of users starting from the current area in multiple consecutive time periods , and obtain the travel frequency of each user and average travel distance ;

[0029] Will , and Substitution Calculate the user travel index of the current area ;

[0030] In the formula is the weight coefficient, For the Users, is the preset scale factor, is a function of the fluctuation of the number of users.

[0031] As a technical solution, through the formula:

[0032] Calculate the fluctuation coefficient of the number of users in the current area ; , They are the maximum and minimum number of users in the current area respectively;

[0033] The calculated volatility coefficient With the preset fluctuation range For comparison, if ,but ;like ,but ;like ,but .

[0034] As a technical solution, obtaining The influence coefficient of the region The process is:

[0035] The vehicle change coefficient in each period , User travel indicators Respectively with the preset warning value , Make comparisons;

[0036] If exists ≥ and ≥ The number of time periods is obtained as p1, then ; is a preset fixed coefficient;

[0037] If exists ≥ or ≥ The number of time periods is obtained as p2, then ;

[0038] otherwise, .

[0039] As a technical solution, the system further includes:

[0040] The sharing management module establishes a vehicle sharing database based on the data collected by the data collection module for tracing vehicle usage records;

[0041] The traffic status prediction model includes a fully connected neural network and a recurrent neural network structure.

[0042] Beneficial effects of the present invention:

[0043] The present invention takes into account the actual vehicle usage, user travel information and traffic congestion conditions through vehicle demand indicators and traffic congestion indicators, so as to comprehensively establish a prediction model for vehicle scheduling. When a surge in demand occurs in the current area, the intelligent scheduling system based on big data analysis can quickly perceive this change and immediately allocate vehicles from other areas to meet the demand in this area. This dynamic and real-time scheduling method not only improves the utilization rate of vehicles, but also greatly improves user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The present invention will be further described below in conjunction with the accompanying drawings.

[0045] Figure 1 is a system structure block diagram of the present invention;

[0046] Figure 2 It is a workflow diagram of the intelligent scheduling module in the present invention. DETAILED DESCRIPTION

[0047] 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.

[0048] See also Figure 1-Figure 2 As shown, the present invention is a vehicle intelligent dispatching and sharing management system based on big data analysis, comprising:

[0049] Data collection module, used to collect vehicle usage data, user data and traffic data;

[0050] Data analysis module, including vehicle demand analysis unit and traffic condition prediction unit;

[0051] The vehicle demand analysis unit analyzes vehicle usage data and user data to identify vehicle changes in different time periods and regions;

[0052] Obtain the travel habits of users in each area, and calculate the user travel index of each area within a working cycle. Based on the vehicle changes and user travel index, calculate the vehicle demand index of each area;

[0053] The traffic condition prediction unit predicts the future traffic conditions based on historical traffic data and real-time traffic data through a pre-trained traffic condition prediction model, and responds with the output of traffic congestion indicators;

[0054] The intelligent scheduling module is used to determine whether there is a risk of redundancy or shortage of vehicles in each area based on vehicle demand indicators and traffic congestion indicators. If there is a risk of redundancy, the redundant vehicles will be transferred out; if there is a risk of shortage, the insufficient vehicles will be transferred in.

[0055] The process of judging whether there is a risk of redundancy or shortage of vehicles in each area based on vehicle demand indicators and traffic congestion indicators is as follows:

[0056] Get vehicle demand indicators for each area at different times and traffic congestion indicators ;

[0057] Obtain the flow of people in each area based on surveillance cameras ;

[0058] Will , Substituting into the formula: Calculate the scheduling demand index ;

[0059] The scheduling demand index obtained by calculation Vehicle Demand Index Make comparisons;

[0060] like , it is determined that there is a risk of insufficient vehicles in the current area, and vehicles are transferred to the current area;

[0061] like , then it is judged that there is a redundancy risk for vehicles in the current area, and vehicles are dispatched from the current area; if neither of the above two risks exists, the vehicle dispatch plan for the current area remains unchanged;

[0062] In the formula For the Regions, As a function, the current period is divided into multiple sub-periods at equal intervals. If the traffic congestion index in three consecutive sub-periods increases or the traffic congestion index exceeds the preset traffic congestion warning value, then ,otherwise, ;

[0063] , The preset reference difference is determined based on historical data and experience data. is the curve of the flow of people in the current area changing over time, The reference curve of the flow of people in the current area over time, ~ For the detection period, is the exponential coefficient, , , is the conversion factor, which is selected and determined based on historical data.

[0064] The exponential coefficient The expression is: ;

[0065] In the formula, , is a weighting factor, formulated based on historical data analysis. is the maximum value of traffic congestion index in each sub-period, is the minimum value of traffic congestion index in each sub-period, is the curve of traffic congestion index changing over time in each sub-period. For any sub-period, is the total number of sub-periods, , The start and end time of any sub-period.

[0066] In this embodiment, in order to solve the problem that the existing vehicle dispatching system mentioned in the background technology cannot cope with the surge in vehicle demand, the vehicle usage data, user data and traffic data are first collected through the data collection module, and then analyzed respectively through the vehicle demand analysis unit and the traffic condition prediction unit in the data analysis module, so as to finally obtain the vehicle demand index and traffic congestion index of each area. The vehicle demand index and traffic congestion index are used to take into account the actual vehicle usage, user travel information and traffic congestion, so as to comprehensively establish a prediction model for vehicle dispatching, so that when the demand surges in the current area, the intelligent dispatching system based on big data analysis can quickly perceive this change and immediately allocate vehicles from other areas to meet the demand in the area; this dynamic and real-time dispatching method not only improves the utilization rate of vehicles, but also greatly improves user satisfaction;

[0067] In addition, the system can further optimize vehicle dispatch through predictive analysis. By deeply mining and analyzing historical data, the system can predict the vehicle demand in various regions in a certain period of time in the future. Based on these prediction results, the system can dispatch and allocate vehicles in advance to avoid vehicle shortages or surpluses.

[0068] At the same time, it also provides a method to accurately determine whether there is a risk of redundancy or shortage of vehicles in each area. Specifically, the traffic congestion index and passenger flow of each area are first substituted into Calculate the scheduling demand index Obviously, if the flow of people in an area is larger, the probability of using shared bicycles is higher. If the current time period is divided into multiple sub-periods at equal intervals, if the traffic congestion index in three consecutive sub-periods increases or the traffic congestion index exceeds the preset traffic congestion warning value, it means that the traffic congestion in the area is more serious. Due to the convenience of shared bicycles, the probability of choosing shared bicycles is obviously higher; therefore ,otherwise, In order to accurately express the impact of traffic congestion on the overall dispatch demand index, an exponential coefficient is set , through the dynamically updated exponential coefficient adjustment of the impact on the entire traffic congestion index, and ultimately reflects the impact on the dispatch demand index of each area, where the exponential coefficient The expression is: ;like The greater the change in the traffic congestion index in each sub-period, the higher the probability that there may be a surge in demand in the current area. Combined with the extreme difference of the traffic congestion index, the purpose of accurately judging the change in traffic congestion situation can be achieved, and the accuracy of the dispatch demand index can be improved; It reflects the change in passenger flow during the previous detection period. Obviously, the greater the change, the higher the probability of a surge in vehicle demand in the area. Therefore, the larger the scheduling demand index, the more urgent it is to transfer vehicles to the area.

[0069] The vehicle demand index The calculation formula is:

[0070] ; In the formula For the The influence coefficient of the region, For the The vehicle variation coefficient of the area, No. User travel indicators for each region, , is the preset proportional coefficient;

[0071] The said Vehicle variation coefficient in each area The expression is: ; In the formula, For the The value of the vehicle usage data item, is the total number of vehicle usage data items, , are the starting point and end point of each time period, For the The curve of vehicle usage data changing over time, For the Reference curve of vehicle usage data changing over time, The impact factor for each period is determined based on historical data and empirical data. is the total number of time periods, For the The vehicle usage data includes but is not limited to the number of shared bicycles used per unit time in each area, the riding time of shared bicycles, the starting and ending positions of shared bicycles, and the riding distance and riding speed can be calculated based on the starting and ending positions;

[0072] The present invention provides a method for obtaining vehicle demand indicators in each area. By combining user data and vehicle usage data, the vehicle demand in each area is comprehensively analyzed, thereby obtaining accurate vehicle demand in each area, providing data support for subsequent vehicle scheduling in the area; specifically, by The vehicle demand index of the current area is calculated, where ; Through the above formula, it can be obtained that if the vehicle change coefficient in the current area is larger, the vehicle demand in the current area is larger, and vice versa, if the vehicle change coefficient in the current area is smaller, the vehicle demand in the current area is smaller; and through the cumulative change of vehicles in each time period, it is obvious that if the cumulative change of vehicles is larger, the vehicle demand in the current area is larger, and if the user travel index in the current area is larger, it means that the vehicle demand in the current area is larger, and vice versa.

[0073] Get the User travel indicators in each region The process is:

[0074] Get the number of users starting from the current area in multiple consecutive time periods , and obtain the travel frequency of each user and average travel distance ;

[0075] Will , and Substitution Calculate the user travel index of the current area ;

[0076] In the formula is the weight coefficient, obtained based on historical data analysis, For the Users, is a preset proportionality factor, determined based on historical data analysis. is a function of the fluctuation of the number of users.

[0077] By formula:

[0078] Calculate the fluctuation coefficient of the number of users in the current area ; , They are the maximum and minimum number of users in the current area respectively;

[0079] The calculated volatility coefficient With the preset fluctuation range For comparison, if ,but ;like ,but ;like ,but .

[0080] The present invention provides a method for obtaining user behavior indicators in each area. Specifically, the number of users who actually use shared bicycles in each area is first obtained, that is, the starting point of using shared bicycles is the current area. Obviously, if the number of users is larger, it means that the user behavior in the current area is more inclined to travel by shared bicycles, so the demand for vehicles is greater. The greater the travel frequency and average travel distance of each user in a time period, the greater the probability that the current user tends to travel by shared bicycles, so the demand for vehicles is greater. Then, through the formula Calculate and obtain the user behavior indicators in each area. By accumulating, summing and averaging, more accurate user behavior indicators can be obtained, providing accurate data support for subsequent vehicle dispatch judgments. At the same time, in order to improve the accuracy of user behavior indicators, The method of taking values ​​is dynamically adjusted. Specifically, the fluctuation coefficient of the number of users in the current area is obtained by means of variance. Obviously, the larger the fluctuation coefficient, the greater the fluctuation of the current number of users in the previous cycle, and the more likely it is that there will be a surge in users, so the urgency of vehicle scheduling is greater.

[0081] Get the The influence coefficient of the region The process is:

[0082] The vehicle change coefficient in each period , User travel indicators Respectively with the preset warning value , Make comparisons;

[0083] If exists ≥ and ≥ The number of time periods is obtained as p1, then ; It is a preset fixed coefficient, determined based on the analysis of historical data and experimental data;

[0084] If exists ≥ or ≥ The number of time periods is obtained as p2, then ;

[0085] otherwise, .

[0086] The present invention provides a method for obtaining an influence coefficient. Specifically, by comparing the vehicle change coefficient, the user travel index and the respective warning values ​​in each time period, if the vehicle change coefficient and the user travel index both exceed the warning values, it means that the probability of a surge in vehicle demand in the area is greater, and therefore the impact on the vehicle demand index is greater. If at least one of them exceeds the warning value, it means that there is also a probability of a surge in vehicle demand in the area, and therefore the impact on the vehicle demand index also needs to be increased.

[0087] The system further comprises:

[0088] The sharing management module establishes a vehicle sharing database based on the data collected by the data collection module for tracing vehicle usage records;

[0089] The traffic status prediction model includes a fully connected neural network and a recurrent neural network structure.

[0090] The process of building a traffic state prediction model composed of a fully connected neural network FCN and a recurrent neural network RNN ​​is as follows:

[0091] Data preprocessing: preprocessing historical traffic data, including normalization, standardization, padding or truncating time series data, and converting non-sequence data into a format suitable for FCN;

[0092] Construct the RNN layer, select the RNN variant LSTM, determine the number of RNN layers and the number of units in each layer to process the time series data in the historical traffic data;

[0093] Connect the FCN layer, use the output of the RNN layer as the input of the FCN, and build a fully connected layer for feature extraction and classification;

[0094] Model compilation, select loss function, optimizer and evaluation index to compile traffic state prediction model;

[0095] Model training: Use the training data to train the model and validate it on the validation set to adjust the hyperparameters;

[0096] Model evaluation and tuning: Evaluate the performance of the model on the test set and tune it as needed.

[0097] In the field of deep learning, in order to process complex data and tasks, it is often necessary to combine different types of neural networks to make full use of their respective advantages. The composite model of RNN and FCN is one of the mature technologies. Therefore, the construction process of the above model is an existing technology and will not be elaborated here.

[0098] It should be noted that the calculation formulas and various parameters involved in the calculation in the present invention have been dimensionally processed in advance, and the process of dimensionless processing is well known in the industry and will not be described here.

[0099] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A vehicle intelligent dispatching and sharing management system based on big data analysis, characterized in that: include: Data collection module, used to collect vehicle usage data, user data and traffic data; Data analysis module, including vehicle demand analysis unit and traffic condition prediction unit; The vehicle demand analysis unit analyzes vehicle usage data and user data to identify vehicle changes in different time periods and regions; Obtain the travel habits of users in each area, and calculate the user travel index of each area within a working cycle. Based on the vehicle changes and user travel index, calculate the vehicle demand index of each area; The traffic condition prediction unit predicts the future traffic conditions based on historical traffic data and real-time traffic data through a pre-trained traffic condition prediction model, and responds with the output of traffic congestion indicators; The intelligent dispatching module is used to determine whether there is a risk of redundancy or shortage of vehicles in each area based on vehicle demand indicators and traffic congestion indicators. If there is a risk of redundancy, the redundant vehicles will be dispatched; if there is a risk of shortage, the insufficient vehicles will be dispatched; The process of judging whether there is a risk of redundancy or shortage of vehicles in each area based on vehicle demand indicators and traffic congestion indicators is as follows: Get vehicle demand indicators for each area at different times and traffic congestion indicators ; Obtain the flow of people in each area based on surveillance cameras ; Will , Substituting into the formula: Calculate the scheduling demand index ; The scheduling demand index obtained by calculation Vehicle Demand Index Make comparisons; like , it is determined that there is a risk of insufficient vehicles in the current area, and vehicles are transferred to the current area; like , it is determined that there is a redundancy risk for vehicles in the current area, and vehicles are dispatched from the current area; In the formula For the Regions, As a function, the current period is divided into multiple sub-periods at equal intervals. If the traffic congestion index in three consecutive sub-periods increases or the traffic congestion index exceeds the preset traffic congestion warning value, then ,otherwise, ; , The preset reference difference, is the curve of the flow of people in the current area changing over time, The reference curve of the flow of people in the current area over time, ~ For the detection period, is the exponential coefficient, , , is the conversion factor.

2. The vehicle intelligent dispatching and sharing management system based on big data analysis according to claim 1 is characterized in that: The exponential coefficient The expression is: ; In the formula, , is the weight factor, is the maximum value of traffic congestion index in each sub-period, is the minimum value of traffic congestion index in each sub-period, is the curve of traffic congestion index changing over time in each sub-period. For any sub-period, is the total number of sub-periods, , The start and end time of any sub-period.

3. The vehicle intelligent dispatching and sharing management system based on big data analysis according to claim 1 or 2 is characterized in that: The vehicle demand index The calculation formula is: ; In the formula For the The influence coefficient of the region, For the The vehicle variation coefficient of the area, No. User travel indicators for each region, , is the preset proportional coefficient; The said Vehicle variation coefficient in each area The expression is: ; In the formula, For the The value of the vehicle usage data item, is the total number of vehicle usage data items, , are the starting point and end point of each time period, For the The curve of vehicle usage data changing over time, For the Reference curve of vehicle usage data changing over time, is the impact factor for each period, is the total number of time periods, For the time period.

4. The vehicle intelligent dispatching and sharing management system based on big data analysis according to claim 3 is characterized in that: Get the User travel indicators in each region The process is: Get the number of users starting from the current area in multiple consecutive time periods , and obtain the travel frequency of each user and average travel distance ; Will , and Substitution Calculate the user travel index of the current area ; In the formula is the weight coefficient, For the Users, is the preset scale factor, is a function of the fluctuation of the number of users.

5. The vehicle intelligent dispatching and sharing management system based on big data analysis according to claim 4 is characterized in that: By formula: Calculate the fluctuation coefficient of the number of users in the current area ; , They are the maximum and minimum number of users in the current area respectively; The calculated volatility coefficient With the preset fluctuation range For comparison, if ,but ;like ,but ;like ,but .

6. The vehicle intelligent dispatching and sharing management system based on big data analysis according to claim 3 is characterized in that: Get the The influence coefficient of the region The process is: The vehicle change coefficient in each period , User travel indicators Respectively with the preset warning value , Make comparisons; If exists ≥ and ≥ The number of time periods is obtained as p1, then ; is the preset fixed coefficient, If exists ≥ or ≥ The number of time periods is obtained as p2, then ; otherwise, .

7. The vehicle intelligent dispatching and sharing management system based on big data analysis according to claim 1 is characterized in that: The system further comprises: The sharing management module establishes a vehicle sharing database based on the data collected by the data collection module for tracing vehicle usage records; The traffic status prediction model includes a fully connected neural network and a recurrent neural network structure.

Images