Integrated smart management platform for urban public services

Through the integrated smart management platform of urban public services, cloud database and feature matrix analysis is used to solve the problems of information asymmetry and uneven resource allocation in traditional management, the accurate acquisition of user information and efficient allocation of resources are achieved, and the user has been improved Experience and service efficiency.

CN118798580BActive Publication Date: 2024-12-06XIAMEN C&D CITY SERVICE DEV CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411265058.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-10
Publication Date
2024-12-06
Estimated Expiration
2044-09-10

AI Technical Summary

Technical Problem

There are problems of information asymmetry and uneven resource allocation in traditional urban public service management, which makes it difficult for users to obtain accurate service point operation status and waiting time information, and the resource allocation efficiency is low, resulting in waste.

Method used

An integrated intelligent management platform is designed, including basic data acquisition module, service point path map module, traffic flow prediction module, data acquisition module, data preprocessing module, service point carrying capacity prediction module, priority queue prediction module and weather prediction module. Through cloud database and feature matrix analysis, an optimized service point recommendation queue is generated to improve user experience and resource utilization efficiency.

Benefits of technology

It realizes users to obtain accurate service point information and optimize resource allocation, improves user satisfaction and service efficiency, avoids resource waste, and meets users' personalized needs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118798580B_ABST
    Figure CN118798580B_ABST
Patent Text Reader

Abstract

The present invention is applicable to the technical field of management platform, and provides an integrated intelligent management platform for urban public services, including: a basic data acquisition module; a service point route map module; a traffic flow prediction module; a data collection module; a data preprocessing module; a service point carrying capacity prediction module; a priority queue prediction module, which is used to generate a priority queue on the day of the appointment time based on the preprocessing data and the carrying capacity prediction result; a weather prediction module; a screening module, which is used to optimize the priority queue according to the weather conditions on the day of the appointment time; and a recommendation module, which is used to recommend a corresponding service point to a user. The present invention has a good appointment service function and can recommend a priority queue to a user for the user to select a service point.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of management platforms, and in particular to an integrated smart management platform for urban public services. Background Art

[0002] In traditional urban public service management, information asymmetry is a significant problem. Users often cannot obtain key information about the operation status, waiting time, and service quality of service points in a timely manner. This information asymmetry not only leads to blindness in the selection of service points, but also may cause users to find that the waiting time is too long or the service quality does not meet expectations after arriving at the service point, thus wasting a lot of time and energy.

[0003] In addition, uneven resource allocation is also a prominent problem in traditional urban public service management. Due to the lack of accurate data analysis and prediction capabilities, it is difficult for service providers to dynamically adjust resources according to actual conditions. This often leads to some popular service points being overcrowded, while other service points are deserted. This uneven resource allocation not only reduces the overall service efficiency, but may also lead to a waste of resources. Summary of the invention

[0004] The purpose of the present invention is to provide an integrated intelligent management platform for urban public services, aiming to solve the problems mentioned in the background technology.

[0005] The present invention is implemented as follows: an integrated urban public service smart management platform comprises:

[0006] The basic data acquisition module is used to obtain the user's basic information, including the user's reservation time, departure location and selected travel mode;

[0007] The service point path map module is used to obtain the service point locations in the target area according to the user's starting location and generate a route map from the starting location to each service point;

[0008] The traffic flow prediction module is used to predict the traffic flow data of each route within the reservation time period using the generated route map;

[0009] The data collection module is used to continuously collect raw data in the target area within the reservation time T and at a preset time interval h, including the real-time carrying capacity and traffic flow of each service point;

[0010] A data preprocessing module is used to preprocess the collected raw data, generate preprocessed data, and construct a feature matrix for analysis based on the preprocessed data;

[0011] The service point carrying capacity prediction module is used to predict the carrying capacity of each service point within the reservation time period by using the pre-processed data and the feature matrix to obtain the carrying capacity prediction result;

[0012] The priority queue prediction module is used to generate the priority queue on the day of the appointment time based on the preprocessed data and the carrying capacity prediction results;

[0013] A weather forecast module is used to use the generated route map to predict the weather forecast data of the target area within the reservation time period;

[0014] A screening module is used to optimize the priority queue according to weather forecast data to obtain an optimized priority queue;

[0015] The recommendation module is used to recommend a corresponding service point to the user based on the optimized priority queue.

[0016] Preferably, the basic data acquisition module acquires the user's starting position through the user's input device or positioning device.

[0017] Preferably, the service point path map module forms a service point route map of the target area based on the pre-stored service point location information and the user's starting location. The calculation formula of the service point location information and the user's starting location path is:

[0018] ,in, Indicates at a specific time From point The user's starting location to the point The cost of the optimal path to the service point location; Indicates that at all points To point The set of possible paths Find the path that minimizes the total cost ; For a given path , for all the road segments The weighted lengths of are summed up; Indicates at a specific time Slave Node To Node Length of road section; Indicates at a specific time Road Section Traffic congestion coefficient; Indicates at a specific time Road Section The road condition coefficient; For a given path , for all its nodes The additional costs are summed up; Indicates at a specific time Passing Node The additional costs include waiting time at traffic lights, tolls and parking fees.

[0019] Preferably, the traffic flow prediction module includes a data acquisition submodule and a feature extraction module. The data acquisition submodule is used to clean the collected raw traffic data; the feature extraction module is used to calculate the average vehicle speed based on the timestamp and distance information of the vehicle passing through a specific road section in the GPS data or traffic sensor data, and count the total number of vehicles passing through a certain road section within a specific time period; calculate the average road travel time by analyzing the timestamp data of the vehicle passing through the road section; calculate the traffic density based on the average vehicle speed, the total number of vehicles and the average road travel time, combined with the road length.

[0020] Preferably, the method is used to predict the carrying capacity of each service point within the reservation time period by using the preprocessed data and the feature matrix to obtain a carrying capacity prediction result, including:

[0021] Receive a feature matrix X, which contains preprocessed data of each sub-region in the target region at different time points;

[0022] Input the feature matrix X into the carrying capacity prediction model to output the predicted number of people arriving without reservations ; The carrying capacity prediction model is constructed based on the ARIMA model and trained based on the historical data of the platform;

[0023] The bearing capacity coefficient is calculated by the calculation formula. The calculation formula of the bearing capacity coefficient is:

[0024] ,in, represents the carrying capacity coefficient, which indicates the carrying capacity of the service point during the reservation period; Represents the number of people who have made reservations; Represents the predicted number of people arriving without a reservation; The maximum number of people that can be carried at the service point; Represents the actual usage area of ​​the service point; Represents the maximum available area of ​​the service point; represents the number of facilities currently available at the service point; The maximum number of facilities that can be accommodated at the representative service point; Represents the number of service personnel currently available at the service point; The maximum number of service personnel that a representative service point can accommodate.

[0025] Preferably, the collected data is preprocessed to obtain preprocessed data, and a feature matrix is ​​constructed according to the preprocessed data. The specific steps include:

[0026] Cleaning the collected raw data to obtain cleaned data;

[0027] Normalize the cleaned data to convert data of different dimensions to the same dimension to obtain normalized data;

[0028] Smoothing the normalized data to obtain smoothed data;

[0029] According to the smoothed data, the feature matrix X is constructed according to the number of sub-regions divided by the target area and the acquisition time points. The size of the feature matrix X is M×N, where M represents the number of sub-regions divided by the target area, N represents the number of time intervals collected at the preset time interval h within the time range T, and each element Representative in the sub-regions, The feature vector at each time point.

[0030] Preferably, the method is used to generate a priority queue for the appointment time based on the preprocessed data and the carrying capacity prediction result, including:

[0031] The particle swarm is randomly initialized, and each particle represents a service point priority queue;

[0032] For each particle, calculate the value of the evaluation function corresponding to each particle, where the value of the evaluation function is the comprehensive score of the priority queue;

[0033] According to the value of the evaluation function, the individual final solution and the global final solution of each particle are updated, the speed and position of each particle are updated, and the steps are repeated until the preset number of iterations is reached to obtain the final priority queue. The final priority queue is the final service point recommendation order on the day of the appointment time.

[0034] Preferably, the calculation formula of the evaluation function is:

[0035] ;

[0036] in, , and is the weight coefficient; and Respectively represent the maximum number of people and the current number of people at the service point; represents the total number of facilities at the service point; Indicates the total number of service personnel at the service point; and They represent the actual travel time under current traffic conditions and the travel time under free flow conditions, respectively; and Represents the user's latitude and longitude coordinates respectively; and Respectively represent the latitude and longitude coordinates of the service point.

[0037] Preferably, in updating the speed and position of each particle, the speed update formula is:

[0038] ;

[0039] in, Indicates Particles in In dimensional space, The speed of the moment; is the inertia weight; Indicates Particles in In dimensional space, The speed of the moment; , and is the learning factor; , and is a random number in the range [0, 1]; Indicates Particles in The best position of an individual in dimensional space; Indicates that all particles are The global optimal position in dimensional space; Indicates The best position in the neighborhood of a particle; Indicates Particles in In dimensional space, The location at the moment;

[0040] The update formula of position is:

[0041] ;

[0042] in, is a parameter; and are the upper and lower bounds of the d-th dimension search space respectively; Generate a random number between [0, 1].

[0043] Preferably, the priority queue is optimized according to the weather forecast data to obtain an optimized priority queue, including:

[0044] The severity of the weather is assessed based on the wind volume and rainfall, and the weather impact coefficient is calculated. The weather impact coefficient includes the wind volume coefficient and the rainfall coefficient. The calculation formula of the wind volume coefficient is:

[0045] ;

[0046] The calculation formula of rainfall coefficient is:

[0047] ;

[0048] The calculation formula of weather influence coefficient is:

[0049] ;

[0050] in, is the current wind volume, is the current rainfall amount, is the maximum value in the air volume data, is the minimum value in the air volume data. is the maximum value in the rainfall data. is the minimum value in the rainfall data. and To adjust the parameters;

[0051] According to the weather influence coefficient, the time to reach the service point is calculated. The time formula to reach the service point is: ,in is the original path time, The additional time due to weather conditions. ≥0, is the impact coefficient determined based on the severity of the weather;

[0052] According to the time of arrival at the service point, the priority queue is re-sorted to obtain the optimized result.

[0053] Compared with the prior art, the beneficial effects of the present invention are: by setting up a cloud database, a basic data acquisition module, a service point route map module, a service point carrying capacity prediction module, a service point carrying capacity prediction module, a traffic flow prediction module, a weather prediction module, a data collection module, a priority queue prediction module, a screening module and a recommendation module, the management platform incorporates various public services into one system, so that the required basic data can be retrieved and loaded from the cloud database, so as to directly obtain the past historical data of each public service, thereby facilitating integrated intelligent processing, and there is no need to establish an additional data collection terminal for collection, wherein the cloud database serves as background support.

[0054] When the user uses it, the user's appointment time, departure location and travel method can be obtained through a terminal, such as an APP on a mobile phone. The traffic flow prediction module uses the route map to predict the traffic flow data of each route within the appointment time period. Then, the data acquisition module continuously collects the original data in the target area within the appointment time T according to the preset time interval t. The data preprocessing module processes, preprocesses and characterizes it, thereby constructing a feature matrix with area and time as rows and columns. Each element of the feature matrix is ​​a spliced ​​vector composed of carrying capacity and traffic flow. Then, the service point carrying capacity prediction module obtains the carrying capacity prediction result by preprocessing data and the feature matrix. The priority queue prediction module generates the priority queue on the day of the appointment time based on the preprocessed data and the carrying capacity prediction result. The priority queue is a combination of service points, and then it is optimized in combination with weather conditions, so as to recommend to the user a service point priority queue that can complete the appointment service the fastest, thereby improving user satisfaction.

[0055] The service point carrying capacity prediction module and the traffic flow prediction module are both updated and optimized through past historical data and real-time data. Since the management platform is based on urban areas, the service point carrying capacity and traffic flow data have obvious periodicity. For example, from Monday to Friday, the service point has less traffic, and on Saturdays and Sundays, there is more traffic. The traffic flow is based on the daily working hours, resulting in periodic situations such as rush hour. Through these data, better appointment service effects can be provided for local public services. On this basis, the weather forecast service system can predict the weather conditions on the day of the appointment time, and produce different sizes of impacts according to different weather conditions, and then further update the priority queue to meet user needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0057] The drawings of the present invention include but are not limited to the following:

[0058] Figure 1 A schematic diagram of the architecture of an integrated smart management platform for urban public services provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0061] like Figure 1 FIG. 1 is a schematic diagram of the architecture of an integrated smart management platform for urban public services provided by an embodiment of the present invention, including:

[0062] Cloud database, used to store data of various public services in the target area;

[0063] The basic data acquisition module is used to obtain the user's basic information, including the user's reservation time, departure location and selected travel mode;

[0064] The service point path map module is used to obtain the service point locations in the target area according to the user's starting location and generate a route map from the starting location to each service point;

[0065] The traffic flow prediction module is used to predict the traffic flow data of each route within the reservation time period using the generated route map;

[0066] The data collection module is used to continuously collect raw data in the target area within the reservation time T and at a preset time interval h, including the real-time carrying capacity and traffic flow of each service point;

[0067] A data preprocessing module is used to preprocess the collected raw data, generate preprocessed data, and construct a feature matrix for analysis based on the preprocessed data;

[0068] The service point carrying capacity prediction module is used to predict the carrying capacity of each service point within the reservation time period by using the pre-processed data and the feature matrix to obtain the carrying capacity prediction result;

[0069] The priority queue prediction module is used to generate the priority queue on the day of the appointment time based on the preprocessed data and the carrying capacity prediction results;

[0070] A weather forecast module is used to use the generated route map to predict the weather forecast data of the target area within the reservation time period;

[0071] A screening module is used to optimize the priority queue according to weather forecast data to obtain an optimized priority queue;

[0072] The recommendation module is used to recommend a corresponding service point to the user based on the optimized priority queue.

[0073] In the embodiment of the present invention, the management platform integrates various public services into one system, so it can retrieve and load the required basic data from the cloud database, so as to directly obtain the past historical data of various public services, so as to facilitate integrated intelligent processing, and no additional data collection terminal needs to be established for collection, wherein the cloud database serves as background support; when the user uses it, the user's appointment time, departure location and travel method can be obtained through the terminal, such as the APP on the mobile phone, and the traffic flow prediction module uses the route map to predict the traffic flow data of each route within the appointment time period, and then the data collection module continuously collects the original data in the target area within the appointment time T according to the preset time interval t, and the data preprocessing module processes it, preprocesses it and characterizes it, so as to construct a feature matrix with area and time as rows and columns, and each element of the feature matrix is ​​a spliced ​​vector composed of carrying capacity and traffic flow, and then the service point carrying capacity prediction module obtains the carrying capacity prediction result by preprocessing data and feature matrix, and the priority queue prediction module Based on the preprocessed data and the carrying capacity prediction results, the block generates a priority queue on the day of the appointment time. The priority queue is a combination of service points, and then it is optimized in combination with the weather conditions, so as to recommend to the user a service point priority queue that can complete the appointment service as quickly as possible, thereby improving user satisfaction; note that the service point carrying capacity prediction module and the traffic flow prediction module are updated and optimized through past historical data and real-time data, and because the management basis of the management platform is the urban area, the service point carrying capacity and traffic flow data have obvious periodicity. For example, from Monday to Friday, the service point has less traffic, and on Saturdays and Sundays, there are more traffic. The traffic flow is based on the daily working hours, resulting in cyclical situations such as rush hour. Through these data, better appointment service effects can be provided for local public services. On this basis, the weather forecast service system can predict the weather conditions on the day of the appointment time, and different weather conditions will produce different impacts, and then further update the priority queue to meet user needs.

[0074] As a preferred embodiment of the present invention, the basic data acquisition module acquires the user's departure location through the user's input device or positioning device, and stores it in the cloud database.

[0075] As a preferred embodiment of the present invention, the service point path map module forms a service point route map of the target area based on the pre-stored service point location information and the user's starting location. The calculation formula of the service point location information and the user's starting location path is:

[0076] ,in, Indicates at a specific time From point The user's starting location to the point The cost of the optimal path to the service point location; Indicates that at all points To point The set of possible paths Find the path that minimizes the total cost ; For a given path , for all the road segments The weighted lengths of are summed up; Indicates at a specific time Slave Node To Node Length of road section; Indicates at a specific time Road Section Traffic congestion coefficient; Indicates at a specific time Road Section The road condition coefficient; For a given path , for all its nodes The additional costs are summed up; Indicates at a specific time Passing Node The additional costs include waiting time at traffic lights, tolls and parking fees.

[0077] In the embodiment of the present invention, the service point path map module plays a vital role in the integrated smart management platform of urban public services. By forming a service point route map in the target area through the pre-stored service point location information and the user's departure location, the efficiency of path optimization and traffic management can be greatly improved. Determine the optimal path cost, taking into account the length of the road segments , Traffic congestion coefficient , Road condition coefficient And additional costs , thereby ensuring the accuracy and rationality of path selection.

[0078] Specifically, the service point path map module can be used to In the process, based on the above comprehensive factors, the path with the lowest total cost is selected, which not only effectively reduces the user's travel time and improves travel efficiency, but also can flexibly respond to traffic changes in different time periods and provide real-time optimal path recommendations; by dynamically adjusting path planning, this module also plays an important role in reducing traffic congestion and optimizing road use. Taking into account additional costs such as waiting time at traffic lights, tolls and parking fees, this module can provide users with more optimized travel plans in terms of economic cost, thereby improving user satisfaction and travel experience.

[0079] In addition, the service point route map module alleviates traffic pressure on certain roads in the city by optimizing route selection and diverting traffic, improves the operating efficiency of the overall transportation system, and provides technical support and guarantee for the development of smart cities. This route optimization method that takes multiple factors into consideration is not only suitable for different times, weather and traffic conditions, but also ensures that the platform can provide efficient and reliable solutions when dealing with complex and changeable urban traffic environments by improving the intelligence level of the system.

[0080] Therefore, the core role of the service point route map module in the integrated urban public service smart management platform is not only reflected in route optimization and the improvement of user travel experience, but also in its improvement of the overall efficiency of the urban transportation system and the promotion of smart city construction. Through this module, the platform can achieve multiple goals in providing personalized services, improving the efficiency of public resource utilization, and optimizing urban management, thereby bringing significant value to users and city managers.

[0081] As a preferred embodiment of the present invention, the method for predicting the carrying capacity of each service point within the reservation time period by using the preprocessed data and the feature matrix to obtain the carrying capacity prediction result includes:

[0082] Receive a feature matrix X, which contains preprocessed data of each sub-region in the target region at different time points;

[0083] Input the feature matrix X into the carrying capacity prediction model to output the predicted number of people arriving without reservations ;

[0084] The bearing capacity coefficient is calculated by the calculation formula. The calculation formula of the bearing capacity coefficient is:

[0085] ,in, represents the carrying capacity coefficient, which indicates the carrying capacity of the service point during the reservation period; Represents the number of people who have made reservations; Represents the predicted number of people arriving without a reservation; The maximum number of people that can be carried at the service point; Represents the actual usage area of ​​the service point; Represents the maximum available area of ​​the service point; represents the number of facilities currently available at the service point; The maximum number of facilities that can be accommodated at the representative service point; Represents the number of service personnel currently available at the service point; The maximum number of service personnel that a representative service point can accommodate.

[0086] In the embodiment of the present invention, by accurately predicting the carrying capacity of each service point in different time periods, the platform can perform resource scheduling and optimal configuration in advance to ensure efficient operation of the service point and timely service for users.

[0087] First, the module forms a feature matrix by comprehensively collecting and processing multi-dimensional data, including the number of reservations, area used, number of facilities, and number of service personnel, laying the foundation for accurate prediction of carrying capacity.

[0088] Secondly, by inputting the feature matrix into the carrying capacity prediction model, the platform can output the number of people who may arrive directly in the future in real time, thereby realizing dynamic monitoring and prediction of the service point load. Furthermore, the calculation formula of the carrying capacity coefficient comprehensively considers multiple key factors, including the number of people who have made reservations, the actual area used, the number of available facilities, and the number of service personnel, ensuring the comprehensiveness and reliability of the prediction results; through this calculation formula, the platform can quantify the actual carrying capacity of the service point, providing a scientific basis for subsequent resource scheduling and optimal configuration.

[0089] Overall, this module not only improves the management efficiency and service quality of service points, but also effectively avoids service overload or resource waste, and optimizes the overall operating efficiency of urban public services. In addition, the carrying capacity prediction module is highly adaptable and flexible, and can provide personalized and accurate carrying capacity predictions based on the specific conditions of different time periods and different service points. The realization of this function has greatly improved the intelligence level of the platform, enabling it to provide efficient and reliable solutions when responding to complex and changeable urban public service needs.

[0090] In summary, the carrying capacity prediction module plays a vital role in the integrated intelligent management platform of urban public services. Through precise calculation and comprehensive analysis, it provides strong support for optimizing the management and operation of urban public services.

[0091] In an embodiment of the present invention, the carrying capacity prediction model may specifically adopt an ARIMA model, which is trained based on the platform's past historical data to obtain a model, thereby predicting the number of people who arrive directly without an appointment, and then obtaining the carrying capacity through a calculation formula.

[0092] Specific steps to predict the number of people who arrive without an appointment:

[0093] Collect historical data of service points from the cloud database. These data should include the number of reservations and actual arrivals at the service point in each time period. Data types include timestamp, number of reservations, actual arrivals, weather conditions, special events, and other factors that may affect the number of arrivals.

[0094] Remove missing values, outliers and noise data to ensure the accuracy and consistency of the data; normalize the data and convert data of different dimensions to the same dimension.

[0095] Use the ARIMA model for time series forecasting, divide the historical data into training set and test set, use 80% of the data for training and 20% for testing;

[0096] Use the training set data to fit the ARIMA model; the ARIMA model consists of three parameters:

[0097] p is the number of autoregressive terms, d is the number of differences, and q is the number of moving average terms; the calculation formula of the ARIMA model is:

[0098] ;

[0099] in, is the observed value at time t; c is a constant; is the autoregressive coefficient, which represents the weighted value of the first p time points; is the moving average coefficient, which represents the weighted value of the first q error terms; is the white noise or error term at time point t;

[0100] Use the test set data to evaluate the model performance; commonly used evaluation indicators include mean square error (MSE), mean absolute error (MAE), etc. The MSE formula is:

[0101] , where n is the number of samples, is the actual value, is the predicted value;

[0102] Based on the evaluation results, the model parameters are adjusted to improve the prediction accuracy.

[0103] Integrate the prediction module into the integrated smart management platform for urban public services to achieve real-time data updates and predictions; establish a feedback mechanism to adjust the model based on the actual number of arrivals to improve prediction accuracy; through the above steps, effective predictions can be made on the number of people who arrive at the service point without an appointment, thereby better planning and managing the resources and carrying capacity of the service point.

[0104] As a preferred embodiment of the present invention, the traffic flow prediction module includes a data acquisition submodule and a feature extraction module. The data acquisition submodule is used to clean the collected original traffic data; the feature extraction module is used to calculate the average vehicle speed based on the timestamp and distance information of the vehicle passing through a specific road section in the GPS data or traffic sensor data, and count the total number of vehicles passing through a certain road section within a specific time period; by analyzing the timestamp data of the vehicle passing through the road section, the average road travel time is calculated; based on the average vehicle speed, the total number of vehicles and the average road travel time, combined with the road length, the traffic density, that is, the number of vehicles per unit road length, is calculated; the extracted features are standardized to eliminate the dimensional differences between different features.

[0105] As a preferred embodiment of the present invention, the steps of preprocessing the collected data and constructing a feature matrix include:

[0106] Data cleaning to remove noise and outliers;

[0107] Data normalization is used to convert data of different dimensions to the same dimension using the formula , where x is the original data, is the normalized data, is the minimum value of the data set, is the maximum value of the data set;

[0108] Data smoothing, using the moving average method, the formula is ,in is the smoothed value at time t, is the original data at time ti;

[0109] Define a feature matrix X, whose size is M×N, where M is the number of sub-regions divided into the target area, and N is the number of time points collected within the time range T at the preset time interval h; the elements of each feature matrix Representative in the The sub-region is The preprocessed data collected at each time point.

[0110] The construction process of the feature matrix X is as follows:

[0111] Create a zero matrix of size M×N; fill the preprocessed data into the corresponding positions of the matrix. The specific steps are:

[0112] Get the first The sub-region is The service point carrying capacity and traffic flow data at each time point; fill the acquired data into the first Row, No. The position of the column.

[0113] In the embodiment of the present invention, through the combined effect of these steps, an efficient and accurate feature matrix is ​​formed, which provides a solid data foundation for subsequent carrying capacity prediction, priority queue generation and path optimization. The feature matrix construction process not only ensures the scientificity and systematicity of data processing, but also plays a key role in other modules.

[0114] First, in the carrying capacity prediction module, the feature matrix is ​​used as input data. Through the carrying capacity prediction model, the carrying capacity of each service point in different time periods can be accurately predicted, providing a scientific basis for optimizing resource allocation; secondly, in the priority queue generation module, the feature matrix provides detailed data in time and space dimensions, supporting the system to generate efficient priority queues to ensure that users receive services in the shortest time; in addition, in the path optimization module, the data of the feature matrix supports the system to calculate the optimal path in real time, taking into account the real-time traffic conditions and the load conditions of the service points, and providing users with the best travel plan; through the synergy of these modules, the feature matrix not only improves the system's data processing capabilities and prediction accuracy, but also significantly improves the overall operation efficiency and user satisfaction of the platform.

[0115] In the embodiment of the present invention, specifically, the target area is divided by spreading with a fixed radius distance with the user's starting position as the center;

[0116] Divide the target area into M sub-areas, each sub-area represents a specific geographical range, and the relationship between adjacent sub-areas is the union;

[0117] Assuming there are 3 sub-regions (M=3) and 4 time points (N=4), the collected data after preprocessing is as follows:

[0118] Sub-area 1: The data at time points 1, 2, 3, and 4 are 0.2, 0.4, 0.3, and 0.5, respectively;

[0119] Sub-area 2: The data at time points 1, 2, 3, and 4 are 0.1, 0.3, 0.5, and 0.2, respectively;

[0120] Sub-area 3: The data at time points 1, 2, 3, and 4 are 0.4, 0.2, 0.6, and 0.3, respectively;

[0121] The constructed feature matrix is:

[0122]

[0123] Therefore, through the above steps, the data acquisition module constructs a feature matrix for subsequent processing and analysis, providing basic data for the priority queue judgment model. Each element in the matrix is ​​a combination vector, which is converted from the service point capacity and communication volume data.

[0124] As a preferred embodiment of the present invention, receiving the pre-processed data and processing it to obtain the priority queue on the day of the appointment time includes:

[0125] Receive a feature matrix X, which contains preprocessed data of each sub-region in the target region at different time points;

[0126] The particle swarm is randomly initialized, and each particle represents a service point priority queue;

[0127] For each particle, calculate the value of the evaluation function corresponding to each particle, where the value of the evaluation function is the comprehensive score of the priority queue;

[0128] According to the value of the evaluation function, the individual final solution and the global final solution of each particle are updated, the speed and position of each particle are updated, and the steps are repeated until the preset number of iterations is reached to obtain the final priority queue. The final priority queue is the final service point recommendation order on the day of the appointment time.

[0129] The calculation formula of the evaluation function is:

[0130] ;

[0131] in, , and is the weight coefficient; and Respectively represent the maximum number of people and the current number of people at the service point; represents the total number of facilities at the service point; Indicates the total number of service personnel at the service point; and They represent the actual travel time under current traffic conditions and the travel time under free flow conditions, respectively; and Represents the user's latitude and longitude coordinates respectively; and Respectively represent the latitude and longitude coordinates of the service point.

[0132] When updating the speed and position of each particle, the speed update formula is:

[0133] ;

[0134] in, Indicates Particles in In dimensional space, The speed of the moment; is the inertia weight; Indicates Particles in In dimensional space, The speed of the moment; , and is the learning factor; , and is a random number in the range [0, 1]; Indicates Particles in The best position of an individual in dimensional space; Indicates that all particles are The global optimal position in dimensional space; Indicates The best position in the neighborhood of a particle; Indicates Particles in In dimensional space, The location at the moment;

[0135] The update formula of position is:

[0136] ;

[0137] in, is a parameter; and are the upper and lower bounds of the d-th dimension search space respectively; Generate a random number between [0, 1].

[0138] As a preferred embodiment of the present invention, the priority queue is optimized according to weather forecast data to obtain an optimized priority queue, including:

[0139] The severity of the weather is assessed based on the wind volume and rainfall, and the weather impact coefficient is calculated. The weather impact coefficient includes the wind volume coefficient and the rainfall coefficient. The calculation formula of the wind volume coefficient is:

[0140] ;

[0141] The calculation formula of rainfall coefficient is:

[0142] ;

[0143] The calculation formula of weather influence coefficient is:

[0144] ;

[0145] in, is the current wind volume, is the current rainfall amount, is the maximum value in the air volume data, is the minimum value in the air volume data. is the maximum value in the rainfall data. is the minimum value in the rainfall data. and To adjust the parameters;

[0146] According to the weather influence coefficient, the time to reach the service point is calculated. The time formula to reach the service point is: ,in is the original path time, The additional time due to weather conditions. ≥0, is the impact coefficient determined based on the severity of the weather;

[0147] According to the time of arrival at the service point, the priority queue is re-sorted to obtain the optimized result.

[0148] In an embodiment of the present invention, by reordering the priority queue, it is ensured that users can obtain accurate service point recommendations even in bad weather, thereby avoiding service delays or user inconvenience caused by weather changes; the recommendation module not only improves the flexibility and adaptability of the platform, but also improves user satisfaction, making the platform more intelligent and reliable when dealing with complex and changeable actual environments, thereby playing an important role in optimizing urban public service management and improving the overall service level.

[0149] As a preferred embodiment of the present invention, the recommendation module displays the optimal recommended service point information to the user through the user interface, and if the user modifies the recommended service point, the optimal path is recalculated.

[0150] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. The integrated smart management platform for urban public services is characterized by: include: The basic data acquisition module is used to obtain the user's basic information, including the user's reservation time, departure location and selected travel mode; The service point path map module is used to obtain the service point locations in the target area according to the user's starting location and generate a route map from the starting location to each service point; The traffic flow prediction module is used to predict the traffic flow data of each route within the reservation time period using the generated route map; The data collection module is used to continuously collect raw data in the target area within the reservation time T and at a preset time interval h. The raw data includes the real-time carrying capacity and traffic flow of each service point. A data preprocessing module is used to preprocess the collected raw data, generate preprocessed data, and construct a feature matrix for analysis based on the preprocessed data; The service point carrying capacity prediction module is used to predict the carrying capacity of each service point within the reservation time period by using the pre-processed data and the feature matrix to obtain the carrying capacity prediction result; The priority queue prediction module is used to generate the priority queue on the day of the appointment time based on the preprocessed data and the carrying capacity prediction results; A weather forecast module is used to use the generated route map to predict the weather forecast data of the target area within the reservation time period; A screening module is used to optimize the priority queue according to weather forecast data to obtain an optimized priority queue; The recommendation module is used to recommend a corresponding service point to the user based on the optimized priority queue; It is used to predict the carrying capacity of each service point within the reservation time period using preprocessed data and feature matrix to obtain the carrying capacity prediction results, including: Receive a feature matrix X, which contains preprocessed data of each sub-region in the target region at different time points; Input the feature matrix X into the carrying capacity prediction model to output the predicted number of people arriving without reservations ; The carrying capacity prediction model is constructed based on the ARIMA model and trained based on the historical data of the platform; The bearing capacity coefficient is calculated by the calculation formula. The calculation formula of the bearing capacity coefficient is: ,in, represents the carrying capacity coefficient, which indicates the carrying capacity of the service point during the reservation period; Represents the number of people who have made reservations; Represents the predicted number of people arriving without a reservation; The maximum number of people that can be carried at the service point; Represents the actual usage area of ​​the service point; Represents the maximum available area of ​​the service point; represents the number of facilities currently available at the service point; The maximum number of facilities that can be accommodated at the representative service point; Represents the number of service personnel currently available at the service point; The maximum number of service personnel that a representative service point can accommodate; Used to generate the priority queue for the appointment time based on the preprocessed data and the carrying capacity prediction results, including: The particle swarm is randomly initialized, and each particle represents a service point priority queue; For each particle, calculate the value of the evaluation function corresponding to each particle, where the value of the evaluation function is the comprehensive score of the priority queue; The calculation formula of the evaluation function is: ; in, , and is the weight coefficient; and Respectively represent the maximum number of people and the current number of people at the service point; Represents the total number of facilities at the service point; Indicates the total number of service personnel at the service point; and They represent the actual travel time under current traffic conditions and the travel time under free flow conditions, respectively; and Represents the user's latitude and longitude coordinates respectively; and Respectively represent the latitude and longitude coordinates of the service point; According to the value of the evaluation function, the individual final solution and the global final solution of each particle are updated, the speed and position of each particle are updated, and the steps are repeated until the preset number of iterations is reached to obtain the final priority queue. The final priority queue is the final service point recommendation order on the day of the appointment time.

2. The urban public service integrated smart management platform according to claim 1 is characterized in that: The basic data acquisition module acquires the user's starting position through the user's input device or positioning device.

3. The integrated smart management platform for urban public services according to claim 2 is characterized in that: The service point path map module forms a service point route map of the target area based on the pre-stored service point location information and the user's starting location. The calculation formula of the service point location information and the user's starting location path is: ,in, Indicates at a specific time From point The user's starting location to the point The cost of the optimal path to the service point location; Indicates that at all points To point The set of possible paths Find the path that minimizes the total cost ; For a given path , for all the road segments The weighted lengths of are summed up; Indicates at a specific time Slave Node To Node Length of road section; Indicates at a specific time Road Section Traffic congestion coefficient; Indicates at a specific time Road Section The road condition coefficient; For a given path , for all its nodes The additional costs are summed up; Indicates at a specific time Passing Node The additional costs include waiting time at traffic lights, tolls and parking fees.

4. The integrated smart management platform for urban public services according to claim 3 is characterized in that: The traffic flow prediction module includes a data acquisition submodule and a feature extraction module. The data acquisition submodule is used to clean the collected raw traffic data; the feature extraction module is used to calculate the average vehicle speed based on the timestamp and distance information of the vehicle passing through a specific road section in the GPS data or traffic sensor data, and count the total number of vehicles passing through a certain road section within a specific time period; by analyzing the timestamp data of the vehicle passing through the road section, the average road travel time is calculated; based on the average vehicle speed, the total number of vehicles and the average road travel time, combined with the road length, the traffic density is calculated.

5. The integrated smart management platform for urban public services according to claim 4 is characterized in that: The collected data is preprocessed to obtain preprocessed data, and a feature matrix is ​​constructed based on the preprocessed data. The specific steps include: Cleaning the collected raw data to obtain cleaned data; Normalize the cleaned data to convert data of different dimensions to the same dimension to obtain normalized data; Smoothing the normalized data to obtain smoothed data; According to the smoothed data, the feature matrix X is constructed according to the number of sub-regions divided by the target area and the acquisition time points. The size of the feature matrix X is M×N, where M represents the number of sub-regions divided by the target area, N represents the number of time intervals collected at the preset time interval h within the time range T, and each element Representative in the sub-regions, The feature vector at each time point.

6. The integrated smart management platform for urban public services according to claim 5 is characterized in that: When updating the speed and position of each particle, the speed update formula is: ; in, Indicates Particles in In dimensional space, The speed of the moment; is the inertia weight; Indicates Particles in In dimensional space, The speed of the moment; , and is the learning factor; , and is a random number in the range [0, 1]; Indicates Particles in The best position of an individual in dimensional space; Indicates that all particles are The global optimal position in dimensional space; Indicates The best position in the neighborhood of a particle; Indicates Particles in In dimensional space, The location at the moment; The update formula of position is: ; in, is a parameter; and are the upper and lower bounds of the d-th dimension search space respectively; Generate a random number between [0, 1].

7. The integrated smart management platform for urban public services according to claim 6 is characterized in that: According to the weather forecast data, the priority queue is optimized to obtain an optimized priority queue, including: The severity of the weather is assessed based on the wind volume and rainfall, and the weather impact coefficient is calculated. The weather impact coefficient includes the wind volume coefficient and the rainfall coefficient. The calculation formula of the wind volume coefficient is: ; The calculation formula of rainfall coefficient is: ; The calculation formula of weather influence coefficient is: ; in, is the current wind volume, is the current rainfall amount, is the maximum value in the air volume data, is the minimum value in the air volume data. is the maximum value in the rainfall data. is the minimum value in the rainfall data. and To adjust the parameters; According to the weather influence coefficient, the time to reach the service point is calculated. The time formula to reach the service point is: ,in is the original path time, The additional time due to weather conditions. ≥0, is the impact coefficient determined based on the severity of the weather; According to the time of arrival at the service point, the priority queue is re-sorted to obtain the optimized result.

Citation Information

Patent Citations

  • Station recommendation method and apparatus

    CN108917782A

  • Artificial intelligence service system

    CN117273992A