Method and system for analyzing operation capability of energy increasing station

Abstract

The invention relates to an energy increasing station operation capability analysis method and system. The method comprises the following steps: acquiring operation data in a preset area; the operation data at least comprises traffic flow data, people flow data and client interaction data; calculating a local operation index based on the operation data; the local operation indexes at least comprise a traffic flow index, a service efficiency index and a business capability index; based on a preset weight combination and the local operation index, calculating a comprehensive operation index of the energy increasing station; and generating an operation report based on the operation data, the local operation index and the comprehensive operation index, and displaying the operation report. By adopting the method, the comprehensiveness and the accuracy of the operation capability analysis of the energy increasing station can be improved.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and in particular to a method and system for analyzing the operating capacity of a fueling station. Background Art

[0002] With the vigorous development of the automobile industry, the demand for operational analysis of gas stations is also increasing. However, current gas stations still rely on data sources such as gas pumps and cash register systems. Correspondingly, the analytical indicators used are also focused on sales, fuel volume, etc., which cannot objectively and comprehensively reflect the overall operational efficiency of gas stations.

[0003] It can be seen that the current analysis method of the operating capacity of refueling stations still has problems of low comprehensiveness and accuracy. Summary of the invention

[0004] Based on this, it is necessary to provide a method and system for analyzing the operating capacity of a fueling station that can improve comprehensiveness and accuracy in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for analyzing the operating capacity of an energy station, the method comprising: Acquiring operation data within a preset area; the operation data at least includes vehicle flow data, passenger flow data, and customer interaction data; Based on the operation data, a local operation index is calculated; the local operation index at least includes a traffic flow index, a service efficiency index and an operating capacity index; Calculating a comprehensive operation index of the energy station based on a preset weight combination and the local operation index; An operation report is generated based on the operation data, local operation index and comprehensive operation index, and the operation report is displayed.

[0006] In one embodiment, obtaining the operation data in the preset area includes: Acquire video data collected by a plurality of video acquisition devices in a preset area, and customer interaction data collected by at least one customer interaction system; Based on the video data, vehicle information recognition and customer information recognition are performed to obtain vehicle flow data and passenger flow data; Based on the vehicle flow data, the pedestrian flow data and the customer interaction data, operation data within a preset area is determined.

[0007] In one embodiment, the determining the operation data in a preset area based on the vehicle flow data, the pedestrian flow data and the customer interaction data includes: Performing data source alignment and standardization on the vehicle flow data, the pedestrian flow data, and the customer interaction data to obtain standardized operation data; Extracting features from the standardized operation data to obtain a plurality of operation features corresponding to each collection device; Based on the weighted voting method, the same operation characteristics of multiple collection devices are fused at the decision level to obtain fused operation data; Based on the isolation forest algorithm, abnormal data is detected and filtered on the fused operation data to obtain the operation data in the preset area.

[0008] In one embodiment, the calculating the local operation index based on the operation data includes: Based on the traffic flow data, a plurality of traffic flow sub-indexes are calculated; the traffic flow sub-indexes include one or more of the following: the conversion rate of vehicles outside the station, the fluctuation coefficient of vehicle flow entering the station, and the trend index of vehicle flow entering the station; A traffic flow index is determined based on the plurality of traffic flow sub-indices and preset traffic flow index weights.

[0009] In one embodiment, the calculating the local operation index based on the operation data includes: Based on the passenger flow data and the customer interaction data, a plurality of service efficiency sub-indices are calculated; the service efficiency sub-indices include one or more of an intelligent queuing time evaluation index, a service process efficiency index, and a customer satisfaction feedback index; A service efficiency index is determined based on the plurality of service efficiency sub-indices and preset service efficiency index weights.

[0010] In one embodiment, the calculating the local operation index based on the operation data includes: Based on the customer interaction data, a plurality of business capability sub-indexes are calculated; the business capability sub-indexes include one or more of an energy sales efficiency index, a non-oil product business performance index, and a customer loyalty and expansion index; Based on the plurality of business capability sub-indices and preset business capability index weights, a business capability index is determined.

[0011] In one embodiment, the operation data includes historical operation data and real-time operation data; the preset weight combination is a dynamic weight combination; and the calculation of the comprehensive operation index of the energy station based on the preset weight combination and the local operation index includes: Generate current forecasted operating data based on historical operating data; Based on the predicted operation data and the real-time operation data, a dynamic weight combination is determined; the dynamic weight combination includes a dynamic weight corresponding to each type of the real-time operation data.

[0012] In one embodiment, determining the dynamic weight combination based on the predicted operation data and the real-time operation data includes: Acquire an initial weight combination; the initial weight combination includes an initial weight corresponding to each of the local operation indexes; Calculating a mean square error between the predicted operation data and the real-time operation data; Based on the mean square error and the genetic algorithm, a tournament selection method is used to select at least two initial weight combinations as parent weight combinations; Performing a single-point crossover operation on the at least two initial weight combinations and applying Gaussian noise to obtain an updated weight combination; Based on multiple iterative optimizations, the weight combination with the smallest mean square error is selected as the dynamic weight combination.

[0013] In one embodiment, generating an operation report based on the operation data, the local operation index and the comprehensive operation index, and displaying the operation report includes: Based on the reinforcement learning algorithm, the operation data, the local operation index, the comprehensive operation index and the dynamic weight combination, training is performed to obtain a strategy mapping relationship; the strategy mapping relationship includes a mapping relationship between the operation status corresponding to the real-time operation data and the strategy recommendation; Based on the real-time operation data and the strategy mapping relationship, obtaining a strategy suggestion corresponding to the real-time operation data; An operation report is generated based on the operation data, local operation index, comprehensive operation index and the strategy recommendation, and the operation report is displayed.

[0014] In a second aspect, the present application provides a system for analyzing the operating capacity of a fueling station, the system comprising: A data collection module, used to obtain operation data in a preset area; the operation data at least includes vehicle flow data, passenger flow data and customer interaction data; A data analysis module, for calculating a local operation index based on the operation data; the local operation index at least includes a traffic flow index, a service efficiency index and an operating capacity index; and calculating a comprehensive operation index of the energy station based on a preset weight combination and the local operation index; The data display module is used to generate an operation report based on the operation data, local operation index and comprehensive operation index, and to display the operation report.

[0015] The above-mentioned method and system for analyzing the operational capacity of the energy station obtains operational data within a preset area; the operational data at least includes vehicle flow data, passenger flow data and customer interaction data; based on the operational data, a local operational index is calculated; the local operational index at least includes a vehicle flow index, a service efficiency index and a business capacity index; based on a preset weight combination and the local operational index, a comprehensive operational index of the energy station is calculated; an operational report is generated based on the operational data, the local operational index and the comprehensive operational index, and the operational report is displayed. Starting from obtaining multi-dimensional operational data, multiple local operational indexes are constructed, and a comprehensive operational index is calculated through a preset weight combination. Finally, the operational report is displayed in a variety of forms, which can achieve the effect of improving the comprehensiveness and accuracy of the operational capacity analysis of the energy station. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 An application environment diagram of a method for analyzing the operating capacity of a fueling station in an embodiment; Figure 2 A schematic diagram of a flow chart of a method for analyzing the operating capacity of a fueling station in one embodiment; Figure 3 It is a structural block diagram of a system for analyzing the operating capacity of a fueling station in one embodiment; Figure 4 It is a structural block diagram of a system for analyzing the operating capacity of a fueling station in another embodiment. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application 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 application and are not used to limit the present application.

[0018] The method for analyzing the operating capacity of a fueling station provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 obtains the operation data in the preset area; the operation data at least includes traffic data, passenger flow data and customer interaction data; based on the operation data, the local operation index is calculated; the local operation index at least includes the traffic flow index, the service efficiency index and the business capacity index; based on the preset weight combination and the local operation index, the comprehensive operation index of the energy station is calculated; based on the operation data, the local operation index and the comprehensive operation index, an operation report is generated, and the operation report is displayed. Among them, the terminal 102 can be but is not limited to various personal computers, laptops, smart phones, tablet computers and Internet of Things devices. The server 104 can be implemented with an independent server or a server cluster consisting of multiple servers.

[0019] In one embodiment, Figure 2 As shown in the figure, a method for analyzing the operating capacity of a fueling station is provided. Figure 1 The terminal 102 in the example is used as an example to illustrate, and the following steps are included: Step S100, obtaining operation data within a preset area.

[0020] The preset area can refer to a specific spatial range defined when analyzing the operational capacity of a fueling station. The preset area can be a circular area extending outwards a certain distance from a single fueling station as the center; or it can be a polygonal area containing multiple adjacent fueling stations and their surrounding related traffic roads and other elements.

[0021] Operational data may be a collection of information reflecting various activities of the charging station during operation, wherein the operational data at least includes vehicle flow data, passenger flow data and customer interaction data.

[0022] The traffic data may be data on vehicles entering, exiting or passing through the energy station, and may further include data on the number, type, speed, and dwell time of vehicles entering, exiting or passing through the energy station.

[0023] The crowd flow data may include data on people entering, exiting or passing through the energy station. Furthermore, the crowd flow data may include data on the number, distribution, and behavior trajectory of people entering or passing through the energy station. For example, the crowd flow data may be obtained by using cameras installed at crowded areas such as the entrances and exits of the energy station and convenience stores, and by using computer vision technology to count the number of people. Bluetooth beacons may also be used to sense mobile phone Bluetooth signals to estimate the flow of people and determine the movement trajectory of people.

[0024] Customer interaction data may include interaction data between customers and various settings in the gas station, including but not limited to operation data and transaction data, etc. Further, customer interaction data may include interaction information between customers and various facilities (such as gas pumps, cash registers, self-service terminals, etc.) in the gas station, as well as service personnel.

[0025] Step S200, calculating a local operation index based on the operation data.

[0026] Among them, the local operation index can be a quantitative indicator constructed for specific aspects of the operation process of the refueling station, which is used to measure the performance of a certain local operation link.

[0027] The local operation index includes at least a traffic flow index, a service efficiency index and an operating capacity index.

[0028] The vehicle flow index may be an evaluation index reflecting the number of vehicles entering a preset area per unit time. The service efficiency index may be a comprehensive index measuring the speed and quality of services provided by a refueling station to customers. The business capability index may be an index used to evaluate the sales capability of a refueling station.

[0029] Step S300, calculating the comprehensive operation index of the energy station based on the preset weight combination and the local operation index.

[0030] The preset weight combination may be a plurality of pre-set weight values, which are used to indicate the importance ratio of each local operation index in the calculation of the comprehensive operation index. The weight setting may be determined according to the operation management focus and strategic objectives of the energy station.

[0031] The comprehensive operation index may be a comprehensive quantitative evaluation index of the overall operation capability of the energy station, and may be calculated by weighted summation, etc. Exemplarily, the calculation of the comprehensive operation index may be achieved by determining a preset weight combination and calculating the comprehensive operation index.

[0032] Step S400, generating an operation report based on the operation data, the local operation index and the comprehensive operation index, and displaying the operation report.

[0033] The operation report can be a text or graphic document summarizing and analyzing the operation of the energy station, and can include operation data, local operation index and comprehensive operation index, etc., to provide detailed information on the operation of the energy station. The operation report can be a regular report compiled regularly or a special report prepared temporarily to deal with special circumstances.

[0034] Display can be the process of presenting the contents of the operation report to users in an intuitive and easy-to-understand way. The display can be realized through visual charts, multimedia and other display forms.

[0035] Operation report generation and presentation can be achieved, for example, by designing operation report templates, writing operation report content, developing presentation tools, and publishing operation reports.

[0036] The present embodiment provides a method for analyzing the operational capacity of a refueling station. It starts with acquiring multi-dimensional operational data, constructs multiple local operational indexes, calculates a comprehensive operational index through a combination of preset weights, and finally displays the operational report in a variety of forms, thereby achieving the effect of improving the comprehensiveness and accuracy of the operational capacity analysis of the refueling station.

[0037] In one embodiment, obtaining the operation data in the preset area includes: Acquire video data collected by a plurality of video acquisition devices in a preset area, and customer interaction data collected by at least one customer interaction system; Based on the video data, vehicle information recognition and customer information recognition are performed to obtain vehicle flow data and passenger flow data; Based on the vehicle flow data, the pedestrian flow data and the customer interaction data, operation data within a preset area is determined.

[0038] Among them, the video acquisition device can be a hardware device used to capture images or videos of scenes inside and outside the gas station, and can be deployed in crowded areas such as gas station entrances and exits, refueling areas, and parking lots.

[0039] The customer interaction system may refer to the facilities and service platform used for interacting with customers in a gas station, and may include, for example, a gas station terminal, a cash register system, a member management system, and a customer service center.

[0040] Vehicle information and customer information are identified based on video data to obtain vehicle flow data and pedestrian flow data. For example, this can be achieved through license plate recognition algorithms and face recognition algorithms. In addition, the attributes of people in the video can be classified, the proportion of customers of different age groups and genders can be counted, and crowd attribute analysis can be achieved.

[0041] Based on the vehicle flow data, the pedestrian flow data and the customer interaction data, the operation data in the preset area is determined, and the vehicle flow data, the pedestrian flow data and the customer interaction data can be integrated to obtain comprehensive charging station operation data.

[0042] The present embodiment provides a method for analyzing the operational capacity of a refueling station. By acquiring video data and customer interaction data in a preset area and using computer vision and deep learning technology to identify vehicle information and customer information, detailed vehicle flow data and passenger flow data are obtained, thereby improving the quality and richness of the data and achieving the effect of improving the comprehensiveness and accuracy of the analysis of the operational capacity of the refueling station.

[0043] In one embodiment, the determining the operation data in a preset area based on the vehicle flow data, the pedestrian flow data and the customer interaction data includes: Performing data source alignment and standardization on the vehicle flow data, the pedestrian flow data, and the customer interaction data to obtain standardized operation data; Extracting features from the standardized operation data to obtain a plurality of operation features corresponding to each collection device; Based on the weighted voting method, the same operation characteristics of multiple collection devices are fused at the decision level to obtain fused operation data; Based on the isolation forest algorithm, abnormal data is detected and filtered on the fused operation data to obtain the operation data in the preset area.

[0044] Among them, data source alignment can be to synchronize data from different acquisition devices according to key information such as timestamps and spatial locations to ensure that data between different data sources are consistent in time and space. Standardization can be to convert different types of data into a unified format and dimension so that they can be compared and processed on the same scale.

[0045] Feature extraction can be to mine key attributes or indicators that can reflect the operation status of the energy station from standardized operation data. Common feature extraction methods include mean value, variance, maximum value, minimum value, etc. Exemplarily, the extracted operation features may include time series features, spatial features, behavioral features, etc.

[0046] Weighted voting is an integrated learning method that assigns different weights to the prediction results of multiple models or data sources to obtain the final result. The weights are determined by determining the corresponding weights according to the performance and importance of each acquisition device. Decision-level fusion is a method of fusing the results of multiple data sources at the decision-making level. It directly performs a weighted combination of the final output results. It is suitable for scenarios where different data sources are quite different but each has its own unique advantages.

[0047] Through the technical operation of detecting and filtering abnormal data based on the isolation forest algorithm, it is possible to remove erroneous data caused by sensor failure, external interference and other factors, and retain data that truly represents the normal operating conditions in the preset area, thereby ensuring the accuracy and credibility of the final analysis results.

[0048] The present embodiment provides a method for analyzing the operational capacity of a refueling station, which ensures the consistency and comparability of data by aligning and standardizing the data sources of vehicle flow data, passenger flow data, and customer interaction data; extracts rich operational features through feature extraction, covering multiple dimensions such as statistics, time, space, and behavior; finally, performs decision-level fusion through a weighted voting method, fully utilizes the advantages of multi-source data, and improves the accuracy and reliability of operational data; detects and filters abnormal data through an isolation forest algorithm, further ensuring the quality of operational data, thereby achieving the technical effect of ensuring high-quality operational data, accurately reflecting the operational status of the refueling station, and providing strong support for optimizing management and decision-making.

[0049] In one embodiment, the calculating the local operation index based on the operation data includes: Based on the traffic flow data, calculating a plurality of traffic flow sub-indices; A traffic flow index is determined based on the plurality of traffic flow sub-indices and preset traffic flow index weights.

[0050] Among them, the traffic flow sub-index includes one or more of the off-station vehicle conversion rate, the in-station vehicle flow fluctuation coefficient, and the in-station rate change trend index.

[0051] The off-station vehicle conversion rate can be an indicator that measures the proportion of all passing vehicles in the preset area that actually enter the refueling station for refueling or other services, reflecting the attractiveness and conversion capacity of the refueling station to the surrounding traffic flow. The inbound vehicle flow fluctuation coefficient can be an indicator that describes the stability of vehicle flow in the refueling station. Exemplarily, it can be obtained by dividing the ratio of the number of vehicles entering the station per unit time to the average number of vehicles entering the station by the time fluctuation weight to reflect the degree of traffic fluctuation. The entry rate change trend index can be an indicator of the changing trend of the vehicle entry rate in the refueling station over time. It reflects the changing trend of the entry rate by calculating the year-on-year growth rate or month-on-month growth rate of the number of vehicles entering the station in different time periods.

[0052] Based on the multiple traffic flow sub-indices and the preset traffic flow index weights, the traffic flow index can be determined by assigning preset traffic flow index weights to the above three traffic flow sub-indices (external vehicle conversion rate, inbound vehicle flow fluctuation coefficient, inbound rate change trend index), and then multiplying each sub-indices with their corresponding weights and summing them. The preset traffic flow index weights can be dynamically adjusted according to different application scenarios, the characteristics of the energy station, and the focus of the operator.

[0053] The present embodiment provides a method for analyzing the operating capacity of a refueling station. By calculating a plurality of traffic flow sub-indices based on traffic flow data, and determining a traffic flow index based on the plurality of traffic flow sub-indices and preset traffic flow index weights, the method can achieve the technical effect of improving the comprehensiveness and accuracy of the analysis of the operating capacity of a refueling station.

[0054] In one embodiment, the calculating the local operation index based on the operation data includes: Calculating a plurality of service efficiency sub-indices based on the pedestrian flow data and the customer interaction data; A service efficiency index is determined based on the plurality of service efficiency sub-indices and preset service efficiency index weights.

[0055] Among them, the service efficiency sub-index includes one or more of an intelligent queuing time evaluation index, a service process efficiency index, and a customer satisfaction feedback index.

[0056] The intelligent queuing time evaluation index may be an indicator used to measure the length of time that customers wait for services in a refueling station and its rationality. The service process efficiency index may be an indicator used to evaluate the execution speed and quality of various service processes in a refueling station, such as refueling, settlement, and consultation. The customer satisfaction feedback index may be an indicator used to measure customers' subjective feelings and evaluations of the service quality of a refueling station.

[0057] Based on the plurality of service efficiency sub-indices and the preset service efficiency index weights, the service efficiency index can be determined by adding the sub-indices according to the preset service efficiency index weights. The preset service efficiency index weights can be dynamically adjusted according to different application scenarios, the characteristics of the energy station, and the focus of the operator, which will not be described in detail in this embodiment.

[0058] The present embodiment provides a method for analyzing the operational capacity of a refueling station. By calculating a plurality of service efficiency sub-indices based on passenger flow data and customer interaction data, and determining a service efficiency index based on a plurality of service efficiency sub-indices and preset service efficiency index weights, it is possible to optimize queue management, resource allocation, and service processes, improve customer satisfaction, and achieve the technical effect of improving the comprehensiveness and accuracy of the analysis of the operational capacity of a refueling station.

[0059] In one embodiment, the calculating the local operation index based on the operation data includes: Calculating a plurality of business capability sub-indices based on the customer interaction data; Based on the plurality of business capability sub-indices and preset business capability index weights, a business capability index is determined.

[0060] The business capability sub-index may be a numerical indicator for measuring the operational capabilities of different aspects of the fueling station. The business capability sub-index may include one or more of the energy sales efficiency index, the non-oil product business performance index, and the customer loyalty and expansion index.

[0061] The energy sales efficiency index can be an indicator used to measure the profitability and efficiency of a gas station in energy sales. The energy sales efficiency index can be obtained by obtaining data on energy sales, energy sales costs, and the number of refueling vehicles per unit time from the cash register system and the gas station terminal, and then calculating them comprehensively. The non-oil business performance index can be an indicator for evaluating the operating performance of a gas station in non-oil businesses such as convenience stores, catering, and car services. The customer loyalty and expansion index can be an indicator for measuring the loyalty of gas station customers and the development of new customers. It can be calculated by comprehensively considering indicators such as repurchase rate, membership growth rate, and customer churn rate, thereby reflecting the effectiveness of customer relationship management.

[0062] The business capability index is determined based on the multiple business capability sub-indices and preset business capability index weights. For example, this can be achieved by setting preset business capability index weights according to the characteristics of each business capability sub-indices and performing weighted summation of the multiple business capability sub-indices.

[0063] The present embodiment provides a method for analyzing the operational capability of a charging station. By calculating a plurality of operational capability sub-indexes based on customer interaction data and determining the operational capability index based on preset weights, the method can provide the charging station with a more scientific and detailed basis for operational decision-making, thereby achieving the technical effect of further improving the comprehensiveness and accuracy of operational capability analysis.

[0064] In one embodiment, the operation data includes historical operation data and real-time operation data; the preset weight combination is a dynamic weight combination; and the calculation of the comprehensive operation index of the energy station based on the preset weight combination and the local operation index includes: Generate current forecasted operating data based on historical operating data; A dynamic weight combination is determined based on the predicted operation data and the real-time operation data.

[0065] The historical operation data may be operation data collected at historical moments, and the historical moments may be multiple historical moments within a preset time range. Generating current forecast operation data based on historical operation data may be to analyze trends and patterns in the historical operation data, and to estimate the operation of the energy station in the future in combination with the current market environment and other relevant fluctuation factors.

[0066] Real-time operation data may be data that reflects the current instant operation status of the energy station. The dynamic weight combination may be a set of weight values ​​set for the difference in importance of the overall operation, which may be determined based on the comparative relationship between the predicted operation data and the real-time operation data. Exemplarily, if the difference in traffic volume between the predicted operation data and the real-time operation data is large, it indicates that external factors may have occurred, and the weights may be adjusted accordingly so that the comprehensive operation index more accurately reflects the actual operation status of the energy station, thereby better guiding operation management decisions, ensuring the efficiency and adaptability of operations, and further improving the operation management level of the energy station. In a specific embodiment, when the difference in traffic volume between the predicted operation data and the real-time operation data is large, the dynamic weight corresponding to the traffic volume may be reduced accordingly, thereby reducing the impact of external factors on the evaluation of the operation capacity of the energy station itself.

[0067] The present embodiment provides a method for analyzing the operational capacity of a refueling station. By deeply mining historical operational data to generate predicted operational data, and combining it with real-time operational data to determine a dynamic weight combination, a comprehensive operational index of the refueling station is calculated based on the dynamic weight combination. This can achieve a scientific assessment of the operational level of the refueling station and further improve the comprehensiveness and accuracy of the operational capacity analysis.

[0068] In one embodiment, determining the dynamic weight combination based on the predicted operation data and the real-time operation data includes: Acquire an initial weight combination; the initial weight combination includes an initial weight corresponding to each of the local operation indexes; Calculating a mean square error between the predicted operation data and the real-time operation data; Based on the mean square error and the genetic algorithm, a tournament selection method is used to select at least two initial weight combinations as parent weight combinations; Performing a single-point crossover operation on the at least two initial weight combinations and applying Gaussian noise to obtain an updated weight combination; Based on multiple iterative optimizations, the weight combination with the smallest mean square error is selected as the dynamic weight combination.

[0069] The initial weight combination may be the initial importance ratio assigned to each local operation index, and these weights determine the contribution of each local operation index in the calculation of the comprehensive operation index. For example, the initial weight combination may include a traffic flow index weight, a service efficiency index weight, an operating capacity index weight, etc.

[0070] The mean square error (MSE) is a statistical indicator used to measure the difference between the predicted value and the actual value. The average of the squares of the difference between the predicted value and the actual value can reflect the prediction accuracy of the model. Genetic algorithm is an optimization algorithm that simulates the biological evolution process in nature. It gradually optimizes the quality of the solution through operations such as selection, crossover, and mutation. The tournament selection method is a selection strategy of the genetic algorithm. By randomly selecting a number of individuals to compete, the best performing individual is selected as the parent individual. Single-point crossover is a crossover operation in the genetic algorithm. By randomly selecting a crossover point, some gene fragments of two parent individuals are exchanged to generate new offspring individuals. Gaussian noise is a random perturbation that conforms to the normal distribution. It is used to introduce mutations to prevent the algorithm from falling into the local optimum.

[0071] Multiple iteration optimization can be a process of continuously repeating the above steps and gradually optimizing the weight combination. After multiple rounds of iterations, by selecting the weight combination with the smallest mean square error as the dynamic weight combination, the comprehensive operation index obtained by the real-time operation data can be obtained to characterize the actual operation efficiency of the energy station, thereby achieving the effect of improving prediction accuracy, avoiding overfitting, and enhancing adaptability.

[0072] A method for analyzing the operational capacity of a fueling station provided in this embodiment can significantly improve the accuracy and robustness of the operational capacity analysis of a fueling station by obtaining an initial weight combination, calculating a mean square error, selecting a parent weight combination based on a genetic algorithm, performing a single-point crossover and applying Gaussian noise to generate an updated weight combination, and selecting a dynamic weight combination based on multiple iterative optimizations, thereby further improving the comprehensiveness and accuracy of the operational capacity analysis.

[0073] In one embodiment, generating an operation report based on the operation data, the local operation index and the comprehensive operation index, and displaying the operation report includes: Based on the reinforcement learning algorithm, the operation data, the local operation index, the comprehensive operation index and the dynamic weight combination, training is performed to obtain a strategy mapping relationship; the strategy mapping relationship includes a mapping relationship between the operation status corresponding to the real-time operation data and the strategy recommendation; Based on the real-time operation data and the strategy mapping relationship, obtaining a strategy suggestion corresponding to the real-time operation data; An operation report is generated based on the operation data, local operation index, comprehensive operation index and the strategy recommendation, and the operation report is displayed.

[0074] Among them, reinforcement learning is a machine learning method that learns the optimal behavior strategy through the interaction between the intelligent agent and the environment. Exemplarily, reinforcement learning includes four parts: state, action, reward and strategy. In this embodiment, the state space can be constructed based on the combination of operation data, local operation index, comprehensive operation index and dynamic weight, and multiple actions that can be taken can be defined. Exemplarily, it can include increasing or decreasing the number of gas station attendants, starting or stopping promotional activities, adjusting the inventory of convenience store goods, etc. After multiple iterations of training, the intelligent agent will gradually learn the best actions to be taken under the conditions corresponding to different operation data, forming a strategy mapping relationship.

[0075] Based on the real-time operation data and the strategy mapping relationship, a strategy suggestion corresponding to the real-time operation data is obtained, which can be to convert the real-time operation data into a state vector, input it into a trained agent, identify the current operation state, and output the corresponding strategy suggestion according to the strategy mapping relationship. For example, when it is detected that the traffic volume is too large during peak hours, it is recommended to increase the number of gas station attendants; when it is found that a certain product is not selling well, it is recommended to adjust the promotion strategy.

[0076] The present embodiment provides a method for analyzing the operational capacity of a refueling station. By performing reinforcement learning training on operational data, local operational indexes, comprehensive operational indexes, and dynamic weight combinations, a strategy mapping relationship is obtained, and then strategy recommendations are generated based on real-time operational data. Finally, an operational report is generated for display, thereby achieving the effect of improving the comprehensiveness and accuracy of the operational capacity analysis of the refueling station.

[0077] In order to more clearly illustrate the technical solution of the present application, the present application also provides a detailed embodiment.

[0078] In one embodiment, a method for analyzing the operational capacity of a refueling station is provided, which uses video AI analysis technology to collect initial data to achieve real-time, accurate and objective data; by integrating the traffic flow index, service efficiency index and business capacity index, the operational efficiency of the refueling station is comprehensively and accurately evaluated; the analysis model also has self-learning and dynamic adjustment capabilities, and can evolve over time to maintain its advancement and applicability. The method for analyzing the operational capacity of a refueling station in this embodiment is applied to an operational capacity analysis system for a refueling station.

[0079] like Figure 3 As shown, the energy station operation capacity analysis system includes a data acquisition layer, a data processing layer, a data analysis and modeling layer, and a visualization and decision-making layer.

[0080] Among them, the data acquisition layer includes video acquisition equipment and other data acquisition equipment interfaces.

[0081] Video acquisition equipment includes: entrance and exit and surrounding cameras: used to collect vehicle information at the entrance and exit and surrounding areas of the gas station, including vehicle type, driving trajectory, speed, etc., to provide a basis for traffic data collection; refueling area ( / charging area / gas area) cameras: monitor the vehicle's refueling behavior ( / charging / gas filling behavior), such as refueling time, queuing conditions, etc., to obtain service efficiency-related data; convenience store cameras: set up inside and around the convenience store, combining facial recognition technology and customer behavior analysis to count the number of people entering the store, the number of non-oil consumers, etc., while monitoring product sales.

[0082] Other data collection device interfaces include: gas station ( / charging equipment / gas filling equipment) interface: used to connect to the gas station system (charging equipment / gas filling equipment system) to obtain energy sales data such as fuel quantity and fuel amount; cash register system interface: connected to the cash register system to obtain convenience store sales data, including sales, sales categories, quantity and other information as part of the operating capacity data; customer relationship management system (CRM system) interface: connected to the customer management system to obtain the consumption behavior, satisfaction and other information of the gas station customers as part of the operating capacity data and service efficiency data.

[0083] The data processing layer includes a video AI analysis module and a data fusion and preprocessing module.

[0084] The video AI analysis module includes a target detection and tracking algorithm unit, a behavior analysis algorithm unit, and an image recognition algorithm unit. The target detection and tracking algorithm unit is used to analyze the video streams collected by the entrance and exit and surrounding cameras, accurately identify and distinguish between vehicles entering the station and vehicles outside the station, and classify the vehicle types; the behavior analysis algorithm unit is used to analyze the queuing behavior of vehicles, the shopping behavior and stay time of customers in the convenience store, and the operation process of service personnel based on the video data of the refueling area ( / charging area / gas area) and convenience store cameras; the image recognition algorithm unit is used to use image recognition technology in the convenience store scene to monitor the sales of goods in real time and identify the category, quantity and other information of the goods.

[0085] The data fusion and preprocessing module includes a data fusion unit and a data preprocessing unit. The data fusion unit is used to fuse data from the video AI analysis module, the gas station ( / charging equipment / gas filling equipment) interface, the cash register system interface, and the CRM system interface to form a unified data format to ensure data integrity and consistency. The data preprocessing unit is used to perform preprocessing operations such as cleaning, denoising, and standardization on the fused data to improve data quality and provide a reliable data foundation for subsequent analysis and modeling.

[0086] Furthermore, the steps performed by the data fusion unit include: (1) Data source alignment and standardization. Timestamp synchronization: All devices (cameras, cash register systems, CRM systems) achieve millisecond-level time synchronization through the NTP protocol to ensure the consistency of the time dimension of the data. Spatial mapping: Coordinate calibration of the camera coverage area (such as the refueling area and the entrance of the convenience store), and map the vehicle / customer locations detected in the video to the floor plan of the refueling station to achieve spatial alignment. Data format conversion: Convert the unstructured data of video AI analysis (such as vehicle trajectory and customer stay time) into a structured JSON format to match the database fields of other systems (CRM, cash register system).

[0087] (2) Multi-source data fusion algorithm. Feature-level fusion: Using edge computing framework and multimodal learning framework, features from different data sources are integrated to generate a more comprehensive and informative feature representation. A federated learning framework can also be used to extract features (such as vehicle type, queue length) at local nodes (such as camera edge computing devices), and only upload feature vectors to the central server to avoid leakage of original video data. Decision-level fusion: Use weighted voting to integrate decision results from different data sources. Taking service efficiency data fusion as an example, the confidence level of video AI recognition of "average queue time" is 0.85, and the confidence level of the cash register system recording "checkout time" is 0.75. The fusion formula is final queue time = 0.6×T 视频 +0.4×T 收银系统 . Abnormal data filtering: Detect and remove outliers based on the isolation forest algorithm (such as a sudden drop in the number of vehicles caused by a camera failure).

[0088] The data analysis and modeling layer includes an indicator calculation module, a comprehensive operation index calculation module and a model dynamic adjustment module.

[0089] Among them, the indicator calculation module includes: Traffic flow index calculation unit: According to the pre-processed traffic flow data, calculate the traffic flow related indicators such as the conversion rate of vehicles outside the station, the fluctuation coefficient of the vehicle flow entering the station, and the trend index of the change of the entry rate, and dynamically adjust according to the rules and algorithms set by the system. Service efficiency index calculation unit: Based on the service efficiency data, calculate the intelligent queuing time evaluation index, service process efficiency index, customer satisfaction feedback index and other service efficiency related indicators, while considering dynamic adjustment factors. Business capability index calculation unit: Based on the business capability data, calculate the energy sales efficiency index, non-oil business performance index, customer loyalty and expansion index and other business capability related indicators, and make dynamic adjustments.

[0090] The comprehensive operation index calculation module is used to calculate the comprehensive operation index based on the traffic flow index, service efficiency index and business capacity index and their corresponding weights. The weights can be determined through multi-criteria decision-making methods such as hierarchical analysis method and expert scoring method, and can be adaptively adjusted according to actual operation conditions.

[0091] The model dynamic adjustment module includes: Adaptive weight adjustment unit: using genetic algorithm, according to the collected operation data, with the goal of minimizing the mean square error between the actual operation efficiency and the model prediction value, the weight of each indicator in the comprehensive operation index is adaptively adjusted.

[0092] Continuous learning and optimization unit: Through the Q-learning algorithm, learning and optimization are carried out in the constructed fueling station operation simulation environment. According to market changes and changes in operation data, operation strategies and model parameters are automatically adjusted to ensure the advancement and applicability of the model.

[0093] The visualization and decision support layer includes a visualization module and a decision support module.

[0094] The visualization display module includes a dynamic report generation unit and a visualization prediction and simulation unit. The dynamic report generation unit is used to automatically generate operation reports according to the set time period (such as day, week, month), and display the detailed data, change trends and comparison with historical data of various indicators in the form of bar charts, line charts, pie charts and other visualization forms, and also includes the dynamic adjustment information of the model; the visualization prediction and simulation unit is used to visualize the prediction trends of indicators such as traffic flow, service efficiency and business capacity based on the prediction results of future operations of the model; it provides strategy simulation visualization function, and after entering different strategy parameters, it displays the changes of operation indicators after simulation implementation to assist managers in making decisions.

[0095] The decision support module includes a strategy suggestion generation unit and a resource allocation suggestion unit. The strategy suggestion generation unit is used to dynamically adjust information based on the indicator analysis results and the model to generate strategy suggestions, such as proposing promotional activities or setting up guide signs for the situation of low traffic flow index, and proposing suggestions for adding temporary refueling equipment or optimizing the refueling process for the problem of long queuing time in the service efficiency index. The resource allocation suggestion unit is used to combine traffic flow and service efficiency data to arrange the appropriate number of service personnel such as gas station attendants and convenience store cashiers for different time periods; according to the business capacity index, the inventory of oil and non-oil products is reasonably allocated, such as determining the replenishment time and replenishment quantity based on the energy sales efficiency index and the non-oil product business performance index.

[0096] The method for analyzing the operating capacity of a fueling station in this embodiment includes the following steps: 1. Collecting data using video AI analysis technology: Traffic flow data collection: Install high-definition smart cameras at the entrances and exits of the energy station and key locations around it, and use the target detection and tracking algorithms in video AI analysis technology to accurately identify and count vehicle information. It can not only distinguish between vehicles entering the station and vehicles outside the station, but also classify and identify vehicle types (such as small cars, large trucks, new energy vehicles, etc.). Through real-time analysis of video streams, obtain information such as vehicle driving trajectories and speeds, and calculate traffic flow-related data, such as the conversion rate of vehicles outside the station (the ratio of the actual number of vehicles entering the station to the number of monitored vehicles outside the station), the fluctuation coefficient of vehicle flow in the station (the ratio of the number of vehicles entering the station in the current time period to the average number of vehicles entering the station, taking into account the time fluctuation weight), etc.

[0097] Service efficiency data collection: Cameras are set up in service areas such as refueling areas, charging areas, and convenience stores. The behavior analysis algorithm in video AI analysis technology is used to monitor the queue situation of vehicles, refueling time ( / charging time), and the time customers stay in convenience stores. For example, the actual average refueling time ( / charging time) can be determined by analyzing the time a vehicle stays at a refueling station ( / charging station); the average length of a customer's stay can be calculated by identifying the customer's shopping behavior and checkout time at a convenience store. At the same time, video AI analysis technology can also be used to monitor the operating procedures of service personnel to ensure the implementation of service standards and provide accurate data support for the calculation of the service efficiency index.

[0098] Business capacity data collection: Monitor the customer flow of convenience stores through cameras, combine facial recognition technology and customer behavior analysis to count the number of people entering the store and the number of non-oil consumption. Use image recognition technology to monitor the sales of goods in real time and obtain non-oil sales data, such as sales categories, quantities, and amounts. In addition, video AI analysis technology can also connect with data from equipment such as gas pumps and cash register systems to obtain energy sales data such as fuel volume and fuel amount, thereby comprehensively evaluating the business capacity of the gas station.

[0099] (II) Constructing a comprehensive operational capability analysis model: 1. Traffic flow index (1) Index composition and calculation method ① The conversion rate of off-site vehicles = (the actual number of vehicles entering the station ÷ the number of off-site vehicle monitoring) × 100%. The number of off-site vehicle monitoring is calculated by using multiple smart cameras reasonably set up around the energy station, combined with video analysis technology, to count the number of vehicles that pass by but do not enter the station within a certain period of time (such as 1 hour or 1 day). The actual number of vehicles entering the station is accurately recorded by the vehicle identification system at the entrance of the station.

[0100] The system automatically adjusts the off-station vehicle monitoring area and time range according to different time periods (peak, flat, and trough), different date types (weekdays, weekends, holidays), and real-time changes in surrounding traffic conditions to more accurately reflect the potential customer flow. At the same time, the reference value is set more refined. In addition to the average off-station vehicle conversion rate data of similar refueling stations in the area where the refueling station is located and the historical operation data of this station, factors such as surrounding business district activities and large-scale events are also considered. For example, at a refueling station next to a city's main road, if there are large shopping malls holding promotional activities nearby, the initial reference value during the peak hours on weekdays can be dynamically adjusted according to the scale of the event and the expected number of participants, and may be set to 12%-15%. The system will continuously optimize and adjust the reference value based on the accumulation and analysis of actual operating data.

[0101] ② Fluctuation coefficient of vehicle flow entering the station = (number of vehicles entering the station in the current time period ÷ average number of vehicles entering the station) ÷ time fluctuation weight.

[0102] Introduce a dynamic time weight adjustment mechanism and a vehicle flow trend prediction algorithm. The average number of vehicles entering the station is the average number of vehicles entering the station in the same time period in the past (such as a week or a month). After calculating the ratio of the number of vehicles entering the station in the current time period to the average number of vehicles entering the station, divide it by a time fluctuation weight that comprehensively considers the importance and volatility of the time period and the trend of the vehicle flow entering the station. The time fluctuation weight is initially set according to the importance and volatility of different time periods (the weight of peak time periods is larger, and the weight of trough time periods is smaller, for example, the time fluctuation weight of peak time periods is 0.8, 0.5 for flat peak time periods, and 0.2 for trough time periods), and is dynamically adjusted in combination with the vehicle flow trend prediction model for entering the station based on machine learning. If it is predicted that the number of vehicles entering the station will show an upward trend in the future, the time fluctuation weight of the time period should be appropriately reduced, otherwise it should be appropriately increased.

[0103] The system monitors the changes in the number of vehicles entering the station in real time. When the number of vehicles entering the station in the same time period deviates from the average number of vehicles entering the station by a certain percentage (such as ±20%), it automatically triggers the re-evaluation and adjustment of the time fluctuation weight. At the same time, the reference value is dynamically set according to seasonal changes in the station's historical data, the opening of new commercial facilities in the surrounding area, road construction and other factors. For example, in the first month after a new shopping mall opens in the surrounding area, the system appropriately increases the reference value of the fluctuation coefficient of the vehicle flow in the station during that time period to adapt to the new market environment changes.

[0104] ③Entry rate change trend index = (current entry rate - historical average entry rate) ÷ entry rate standard deviation × trend adjustment factor.

[0105] Among them, a multi-level trend analysis method can be used. First, the station entry rate is calculated, that is, the ratio of the number of vehicles entering the station to the sum of the number of vehicles outside the station. Then, the fluctuation of the station entry rate is measured by calculating the standard deviation of the station entry rate. On this basis, a trend analysis model based on deep learning is used to analyze the changing trend of the station entry rate on different time scales (short-term, medium-term, and long-term), and different trend adjustment factors are set according to the trends of different time scales (greater than 1 for an upward trend, less than 1 for a downward trend, and 1 for a stable trend).

[0106] The system automatically calculates the standard deviation and historical average entry rate of the station entry rate through long-term accumulated historical data. The trend adjustment factor is automatically adjusted according to the changes in the station entry rate for multiple consecutive time periods. For example, if the station entry rate has been on an upward trend for three consecutive working days and the increase exceeds 5%, the trend adjustment factor is set to 1.2. The reference value is set based on the industry average station entry rate level and the market positioning and brand influence of the station. For refueling stations located in highway service areas, the reference value can be set according to the station entry rate change trend index.

[0107] (2) Calculation of comprehensive traffic flow index Traffic flow index = conversion rate of vehicles outside the station × w1 + fluctuation coefficient of vehicle flow entering the station × w2 + entry rate change trend index × w3.

[0108] The initial values ​​of weights are w1=0.4, w2=0.3, w3=0.3.

[0109] The weight of the off-station vehicle conversion rate is 0.4, because it reflects the ability of the refueling station to attract potential customers to a certain extent. A higher off-station vehicle conversion rate means that more passing vehicles can be converted into customers who enter the station for consumption, which is more critical for the refueling station to expand its market share.

[0110] The weight of the fluctuation coefficient of the inbound vehicle flow is 0.3, which reflects its impact on the stability of the station's operation. Although its importance is slightly lower than the conversion rate of vehicles outside the station, a stable inbound vehicle flow helps to reasonably arrange resources such as personnel, materials and equipment.

[0111] The weight of the entry rate change trend index is 0.3. This index reflects the changing trend of the number of vehicles entering the station. It is of great significance to understand the development direction of the refueling station business and can provide a basis for long-term strategic decision-making.

[0112] However, the specific weight distribution may vary depending on different application scenarios, the characteristics of the charging station, and the focus of the operator. In actual applications, operators can adjust the weight distribution based on their own needs and actual data performance through the analytic hierarchy process (AHP), expert scoring method, or other multi-criteria decision-making methods to better reflect the impact of each indicator on the comprehensive traffic flow index.

[0113] 2. Service efficiency index (1) Index composition and calculation method ① Intelligent queuing time evaluation index = (maximum tolerable queuing time ÷ actual average queuing time) × queuing time weight.

[0114] AI analysis technology can be used to process the video images collected by cameras at the entrance and exit of the refueling area ( / charging area) and the queuing area. The specific process is as follows: The target detection and tracking algorithm can use deep learning target detection algorithms (such as YOLO, SSD, etc.) to identify and track vehicles entering the queue in real time.

[0115] Vehicle trajectory analysis can track each vehicle from the time it enters the queue to the start of refueling ( / charging) through the collaborative work of multiple cameras, and record the vehicle's driving trajectory.

[0116] The timestamp record can be that when a vehicle enters the queue, the system automatically records a timestamp; when the vehicle arrives at the refueling station and starts refueling (or charging), another timestamp is recorded. The actual queue time of each vehicle is calculated through these two timestamps.

[0117] The actual average queue time can be obtained in the following ways: Data collection: The system continuously collects the queue time data of each vehicle. Data processing: The collected data is aggregated to calculate the actual average queue time within a certain period of time. For example, the average value is calculated once every hour or every day to obtain the actual average queue time.

[0118] The maximum tolerable queue time can be determined based on customer satisfaction surveys and industry standards.

[0119] The queue time weight can be dynamically adjusted according to customer flow and service demand in different time periods.

[0120] The system automatically adjusts the queue time weight based on real-time customer traffic and the use of refueling equipment ( / charging equipment). For example, during peak hours, the queue time weight is increased to 0.6 to emphasize the impact of queue time on service efficiency. The maximum tolerable queue time reference value is set according to the characteristics of different types of refueling stations (such as urban gas stations, highway gas stations) and customer groups, and is regularly adjusted based on customer feedback and market competition. For example, the maximum tolerable queue time reference value for urban gas stations is 5 minutes, and for highway gas stations is 8 minutes.

[0121] ② Service process efficiency index = (number of standard service process steps ÷ actual number of completed service process steps) × service process weight + (standard service time ÷ actual average service time) × service time weight.

[0122] The actual number of service process steps completed can be calculated through camera + AI analysis: install high-definition cameras in the refueling area ( / charging area) and service area, and use behavioral analysis algorithms to identify and record the completion status of each service step of the staff; automatic timing system: combined with the camera's timestamp record, automatically count the completion time of each step and the total service time.

[0123] The actual average service time can be calculated through camera + AI analysis: the camera records the time from the customer's entry to the customer's departure, and calculates the actual average service time; data processing: the system automatically summarizes and calculates the average service time within a certain period of time.

[0124] The standard number of service process steps and standard service time can be determined based on the service specifications of the refueling station and industry best practices.

[0125] The system automatically updates the number of standard service process steps and standard service time based on the large amount of data obtained from video AI analysis and the feedback from service personnel. For example, if the video AI analysis finds that a certain service process step can be optimized and merged, the system will adjust the number of standard service process steps accordingly.

[0126] The service process weight and service time weight are adjusted according to the customer's attention to service quality and the operation strategy of the charging station. At the same time, combined with the data mining results of video AI analysis, if it is found that customers are highly sensitive to the time of a certain service link, the service time weight can be increased accordingly. The reference value is set according to the advanced level of the same industry and the historical data of this station, encouraging service personnel to continuously improve service efficiency and approach or exceed the advanced level of the industry.

[0127] ③Customer satisfaction feedback index = (number of satisfied customers ÷ total number of customers) × customer satisfaction weight + (customer complaint resolution rate × complaint resolution quality score) ÷ (1 + customer complaint rate) × complaint handling weight.

[0128] The number of satisfied customers and the total number of customers can be collected with the assistance of video AI analysis technology. Cameras are set up in service areas and customer feedback areas (such as cashier counters, rest areas, etc.), and the video AI analysis system is used to identify customers' expressions (such as positive behaviors such as smiles and nods), body language and other emotional clues. The number of satisfied customers and the total number of customers collected by online evaluation systems and on-site questionnaires can be used to more accurately judge customer satisfaction.

[0129] The customer complaint resolution rate is the ratio of the number of resolved complaints to the total number of complaints. The complaint resolution quality score is determined based on the customer's satisfaction evaluation of the complaint resolution results. The customer complaint rate is the ratio of the number of complaining customers to the total number of customers.

[0130] The system collects customer satisfaction feedback and complaint data in real time, including emotional clues data mined by video AI analysis, and automatically adjusts customer satisfaction weights and complaint handling weights according to the changing trends of customer feedback. For example, if video AI analysis finds that customer dissatisfaction is increasing, the complaint handling weight will be increased to 0.4 to emphasize the impact of complaint handling on service efficiency.

[0131] The reference value is set according to the customer satisfaction level and market competition situation of industry benchmark enterprises, and is adjusted regularly. For example, if the customer satisfaction of the industry-leading refueling station reaches 95%, this value will be used as a reference value to motivate the station to continuously improve service quality and customer satisfaction. Through the application of video AI analysis technology, the calculation of service process efficiency index and customer satisfaction feedback index is more accurate and objective, and it also provides a stronger basis for the service optimization of refueling stations.

[0132] (2) Calculation of comprehensive service efficiency index Service efficiency index = intelligent queuing time evaluation index × w1 + service process efficiency index × w2 + customer satisfaction feedback index × w3.

[0133] The initial values ​​of weights are w1=0.4, w2=0.3, w3=0.3.

[0134] The weight of the intelligent queue time evaluation index is 0.4, because the queue time is the key to customer experience. Too long queue time will lead to decreased satisfaction and customer loss, and it reflects the smoothness of the operation process. The weight of the service process efficiency index is 0.3. It is related to service quality and speed and is the focus of operation management. Although the impact is not as direct as the queue time, it is still an important part. The weight of the customer satisfaction feedback index is 0.3. It is the ultimate measure of service effectiveness. Regardless of the queue and process, if the customer is dissatisfied, the service efficiency is low. It can reflect the needs and provide a basis for continuous improvement. However, the specific weight distribution may vary depending on different application scenarios, the characteristics of the charging station, and the focus of the operator. In actual applications, operators can adjust according to their own needs and actual data performance through the analytic hierarchy process (AHP), expert scoring method or other multi-criteria decision-making methods to better reflect the impact of each indicator on the comprehensive service efficiency index.

[0135] 3. Business Capability Index (1) Index composition and calculation method ① Energy sales efficiency index = (actual average added energy ÷ target average added energy) × added energy weight + (sales proportion of high value-added energy products ÷ target sales proportion of high value-added products) × high value-added weight + (energy sales profit margin ÷ target sales profit margin) × profit margin weight.

[0136] Video AI analysis technology is used to assist in monitoring the refueling area (or charging area, gas filling area), automatically identifying the refueling (or charging, gas filling) behavior of oil vehicles and new energy vehicles, including the amount of refueling, refueling time, etc., to achieve statistics on energy sales data of different types of vehicles; it can also monitor the sales of high-value-added energy products in real time, such as high-quality gasoline, special-purpose diesel, etc., and automatically count sales and sales volume. In addition, it can also connect to the gas station (or charging pile) to collect refueling data.

[0137] The system automatically updates the target value based on market price fluctuations, changes in demand for energy products, and adjustments to competitors' marketing strategies. For example, when the demand for high-quality gasoline increases in the market, the system increases the target sales share of high value-added products. The reference value is set based on the advanced level of similar refueling stations in the same region and industry development trends. For refueling stations located in high-end commercial areas, the target average refueling volume and target sales profit margin may be relatively high to reflect their higher market positioning and customer spending power.

[0138] ② Non-oil product operating performance index = (non-oil product sales ÷ target non-oil product sales) × sales weight + (non-oil product gross profit margin ÷ target non-oil product gross profit margin) × gross profit margin weight + (non-oil product inventory turnover rate ÷ target inventory turnover rate) × inventory turnover rate weight.

[0139] Use video AI analysis technology to monitor customer traffic and shopping behavior in convenience stores, automatically count the number of people entering the store, the number of non-oil consumer goods, etc.; combine image recognition technology to monitor the sales of goods in real time and obtain non-oil sales data, including sales categories, quantities, amounts, etc. In addition, non-oil sales and gross profit margins can also be counted through the convenience store sales system. The non-oil inventory turnover rate is the ratio of the non-oil sales cost to the average inventory cost over a certain period of time. The target non-oil sales, target non-oil gross profit margin, and target inventory turnover rate are set according to the non-oil business strategy and market analysis of the gas station.

[0140] The system dynamically adjusts the target value and reference value based on seasonal changes, the effectiveness of promotional activities, and changes in the surrounding business environment. For example, in summer, the peak sales season for non-oil products such as beverages, the system increases the target non-oil sales and inventory turnover reference values. The reference value is set based on the operating data of excellent convenience stores in the same industry and the historical sales performance of this site to encourage the continuous optimization and development of non-oil business.

[0141] ③ Customer loyalty and expansion index = (old customer repeat consumption rate ÷ target old customer repeat consumption rate) × old customer weight + (new customer growth rate ÷ target new customer growth rate) × new customer weight + (member consumption share ÷ target member consumption share) × member weight.

[0142] The repeat consumption rate of old customers is the ratio of the number of old customers who consume again in a certain period to the total number of old customers, the growth rate of new customers is the ratio of the number of new customers to the number of original customers, and the proportion of member consumption is the ratio of the member consumption amount to the total sales. The target value is set according to the customer relationship management goals and market development plan of the energy station.

[0143] Dynamic adjustment and reference value setting: The system analyzes customer consumption behavior and market competition through the customer relationship management system (CRM) and automatically adjusts the target value. For example, when new competitors appear in the surrounding area, the system increases the target new customer growth rate to strengthen market expansion. The reference value is set based on the data of energy stations with high customer loyalty in the industry and market research results, and is committed to improving customer loyalty and expanding the customer base.

[0144] (2) Calculation of comprehensive business capability index Business capability index = energy sales efficiency index × w1 + non-oil product business performance index × w2 + customer loyalty and expansion index × w3.

[0145] The initial values ​​of weights are w1=0.4, w2=0.3, w3=0.3.

[0146] The energy sales efficiency index has a weight of 0.4, because energy sales is a core business, related to profitability and development, and is the focus of market competition. The non-oil business performance index has a weight of 0.3, and non-oil business is an important part of diversification, with an increasing share of revenue and can improve customer experience. The customer loyalty and expansion index has a weight of 0.3, which is the key to long-term stable development. Having a stable customer base and attracting new customers can ensure sustainable operations and is also a manifestation of competitive advantage.

[0147] 4. Calculation of comprehensive operation index: Comprehensive operation index = traffic flow index × weight 1 + service efficiency index × weight 2 + business capacity index × weight 3. Among them, weight 1, weight 2 and weight 3 can be determined according to the actual operation characteristics and strategic goals of the charging station through the analytic hierarchy process (AHP), expert scoring method and other methods to reflect the importance of each index in the comprehensive evaluation.

[0148] Weighted average method: assign different weights to different indicators according to their importance, then multiply the score of each indicator by its corresponding weight and sum them up to get a comprehensive score.

[0149] Standardization: Since different indicators may have different units and scales, each indicator needs to be standardized (e.g., minimum-maximum normalization) so that they can be directly compared and aggregated.

[0150] Analytical Hierarchy Process (AHP): It is used to determine the relative importance of various indicators and build an evaluation system based on this.

[0151] Furthermore, the computational model can be continuously optimized using dynamic adjustment.

[0152] 3. Dynamic adjustment of the model: 1. Adaptive weight adjustment Data preparation: Collect operational data from the past year, including daily traffic volume, number of vehicles entering the station, entry rate, average waiting time, average stay time, number of refueling vehicles, refueling rate, average refueling amount, number of people entering the store, number of non-oil product consumers, non-oil product consumption rate, average unit price per customer for non-oil products, etc.

[0153] Initialization weights: The initial weights are set to w1=0.4, w2=0.3, w3=0.3.

[0154] The objective function is to minimize the prediction error, which is defined as the mean square error (MSE) between the actual operating efficiency (such as the comprehensive operating index) and the model prediction value.

[0155]

[0156] in, is the actual value, is the model prediction value, and N is the number of samples.

[0157] Genetic Algorithm Implementation: Genetic algorithm is implemented using Python's DEAP library.

[0158] The initial population size is 50 and the number of iterations is 100.

[0159] Population initialization: The initial weights are set to w1=0.4, w2=0.3, w3=0.3.

[0160] The selection operation adopts the tournament selection method, which randomly selects 5 individuals from the population and retains the 2 with the highest fitness.

[0161] The crossover operation uses a single-point crossover to exchange a part of the parent weight vector (such as exchanging w1 and w2).

[0162] Mutation operation: Gaussian noise (μ=0, σ=0.05) is applied to the weights to ensure that the sum of the weights is still 1.

[0163] The genetic algorithm optimization is run once a week to update the optimal weights to the comprehensive operation index calculation formula: Comprehensive operation index = traffic flow index × weight 1 + service efficiency index × weight 2 + business capacity index × weight 3. Constraints: Each weight wi ≥ 0.1 to avoid a single indicator being completely ignored.

[0164] For example, after optimization by genetic algorithm, the optimal weight combination w1=0.35, w2=0.35, w3=0.3 is obtained. The calculation formula for updating the comprehensive operation index is: comprehensive operation index=0.35×traffic flow index+0.35×service efficiency index+0.3×business capacity index.

[0165] 2. Continuous learning and optimization ①Environmental modeling: Use Python's SimPy library to build a fueling station operation simulation environment. The state space includes indicators such as vehicle flow, service efficiency, and business capacity. The action space includes adjusting the weight of each indicator and changing the operation strategy.

[0166] ② Reward function: The reward function is defined as positive rewards for improving operational efficiency and negative rewards for reducing efficiency.

[0167] ③Q-learning algorithm implementation: Use Python's Gym library to implement Q-learning, including: (1) State space: discretized into three dimensions: Traffic flow level (low / medium / high, with thresholds determined by historical data quantiles).

[0168] Service efficiency rating (poor / good / excellent, based on queue time and service process score).

[0169] Business capability level (low / medium / high, based on sales and inventory turnover).

[0170] (2) Action space: includes 6 operations: Increase / decrease the number of refuelers (±1 person).

[0171] Start / stop promotions.

[0172] Adjust convenience store merchandise inventory (±10%).

[0173] (3) Reward function design Positive rewards: When the comprehensive operation index increases, the reward value is the index change (for example, +10 for every 1% increase).

[0174] Negative rewards: If the index decreases or an abnormal event is triggered (such as customer complaint rate > 5%), the reward value is -20.

[0175] The formula is R(s,a)=α×Δcomprehensive operation index+β×event penalty. Among them, positive reward α=10, negative reward β=-20.

[0176] (4) Exploration and Exploitation Strategy Using the ε-greedy strategy, the initial exploration rate =0.5, decaying by 5% each round until =0.1.

[0177] The Q table update formula is as follows:

[0178] Among them, the learning rate η=0.1 and the discount factor γ=0.9.

[0179] (5) Strategy deployment and update The trained Q-table optimal strategy is deployed to the actual system every week. For example, if the status is "high traffic volume + poor service efficiency", then "increase gas station attendants + start promotions" will be executed.

[0180] The Q table is reset every month and retrained to adapt to changes in the market environment.

[0181] 4. Visualization and decision support 1. Visual display: (1) Dynamic report generation Regular reports: Automatically generate operation reports according to the set time period (such as daily, weekly, and monthly). The report content includes detailed data of various indicators, analysis of indicator change trends, comparison with historical data, and dynamic adjustment information of the model. The report is presented in a visual form, such as using bar charts, line charts, pie charts, etc., which is convenient for managers to view and analyze.

[0182] Event-driven reporting: When the system detects a specific operational event (such as abnormal fluctuations in traffic volume, an increase in service complaints, a sudden increase or decrease in sales of a certain type of product, etc.), it automatically generates a targeted report to analyze the cause of the event and make preliminary response suggestions. The report is sent to relevant managers through various means such as email or in-site notification.

[0183] (2) Visual prediction and simulation Trend forecast visualization: Based on the model's forecast results for future operations, the forecast trends of indicators such as vehicle flow, service efficiency, and business capacity are displayed in a visual manner. For example, a line chart can be used to display the forecast value of the number of vehicles entering the station, the average refueling time forecast value, and the sales forecast value in the next week or month, helping managers to plan and prepare in advance.

[0184] Strategy simulation visualization: When managers consider adjusting operational strategies (such as changing promotional activities, adjusting staffing, etc.), the system provides a visualized simulation function. By inputting different strategy parameters, the system displays the changes in operational indicators after the simulation is implemented to help managers evaluate the potential impact of the strategy and choose the best decision-making plan. For example, after simulating the increase in the promotion of a certain type of oil products, the changing trends of indicators such as traffic volume and sales are presented to managers in the form of intuitive charts.

[0185] 2. Decision support: (1) Strategy Recommendation Generation Strategy based on index analysis: Generate strategy recommendations based on the analysis results of the traffic flow index, service efficiency index, business capacity index and their sub-indexes. For example, when the conversion rate of vehicles outside the station is low, it is recommended to set up more obvious guide signs around the refueling station or carry out targeted promotional activities to attract vehicles to enter the station. If the intelligent queue time evaluation index in the service efficiency index is too low, it is suggested to add temporary refueling equipment or optimize the refueling process.

[0186] Dynamic adjustment strategy: According to the dynamic adjustment information of the model, such as weight changes, reference value adjustments, etc., corresponding operational strategy adjustments are proposed. For example, when it is found that the weight of the fluctuation coefficient of the inbound vehicle flow increases in a specific time period, it is recommended to strengthen the staffing and material preparation during this period to cope with the changes in vehicle flow.

[0187] (2) Resource allocation suggestions Staffing: Combine traffic flow and service efficiency data to arrange the appropriate number of gas station attendants, convenience store cashiers, and other service personnel for different time periods. During peak hours, increase staffing appropriately based on predicted traffic flow and expected service time; during off-peak hours, reduce staffing appropriately to improve the efficiency of human resource utilization.

[0188] Material allocation: According to the energy sales efficiency index and non-oil product operating performance index in the business capability index, the inventory of oil and non-oil products is reasonably allocated. For example, when the demand for a certain high-value-added energy product is predicted to increase, replenishment is arranged in advance; according to seasonal changes and non-oil product sales trends, the inventory types and quantities of convenience store goods are adjusted.

[0189] (3) Marketing activity planning Personalized marketing: Develop personalized marketing activities based on customer loyalty and expansion index and customer portraits (obtained through data collection and analysis by video AI). Provide exclusive discounts and services to old customers with high loyalty; design attractive first-time consumer discounts for new customers. Push product and service information that meets the needs of different types of customers (such as long-distance drivers, surrounding residents, etc.).

[0190] Promotion timing selection: Combine the changing trends of the traffic flow index and business capacity index to select the best promotion timing. For example, conduct promotion activities during the low traffic period to stimulate consumption; during holidays or when there are large-scale events in the surrounding area, adjust the promotion strategy and product mix according to the expected traffic flow and customer types.

[0191] A method for analyzing the operating capacity of a fueling station provided in this embodiment also includes a dynamic adjustment trigger mechanism, including: 1. Trigger conditions and threshold settings (1) Abnormal fluctuations in indicators Traffic flow fluctuation: If the number of vehicles entering the station for three consecutive time periods (such as hours) deviates from the historical average by ±20%, the weight adjustment will be triggered.

[0192] Decreased service efficiency: When the queuing time exceeds the maximum tolerance value (e.g. >5 minutes at a city station) for 2 hours, service process optimization is triggered.

[0193] (2) External event driven Surrounding commercial activities: Access local traffic / commercial activity data through API, and adjust the traffic flow prediction model in advance when large-scale activities are detected within 1km.

[0194] Abnormal weather: Access to meteorological data, heavy rain / heavy snow weather will automatically increase the inventory turnover rate reference value by 10%.

[0195] (3) Model performance decay Prediction error exceeds the limit: When the MSE exceeds the threshold (such as 0.05) for three consecutive days, the genetic algorithm is triggered to re-optimize the weights.

[0196] Strategy failure: The Q-learning strategy does not improve the overall index within one week after deployment, triggering strategy retraining.

[0197] 2. Real-time response process (1) Anomaly Detection Module Monitor key indicators based on control charts (such as X-bar charts) and set upper and lower limits (±3σ).

[0198] Use the LSTM model to predict the indicators for the next period and compare them with the actual values. If the deviation is greater than 15%, an alarm is triggered.

[0199] (2) Dynamically adjust execution Immediate response: For emergency events (such as equipment failure), skip model optimization and directly execute preset strategies (such as activating a backup refueling machine).

[0200] Delayed confirmation: For non-emergency events (such as traffic flow fluctuations), it is necessary to confirm that there are abnormalities at three consecutive time points before starting the genetic algorithm or Q-learning optimization.

[0201] (3) Feedback loop The actual effect is recorded after each adjustment. If it does not meet expectations (such as MSE does not decrease), the search space is expanded (such as increasing the mutation intensity of the genetic algorithm to σ = 0.1).

[0202] The present embodiment provides a method for analyzing the operational capacity of a refueling station. By utilizing video AI analysis technology and combining a variety of algorithms, the method realizes the real-time comprehensive collection of multi-dimensional data such as vehicle information, customer behavior, and service personnel operations, thereby avoiding the one-sidedness and backwardness of data caused by a single data source and manual statistics, thereby achieving the effect of improving the real-time and accuracy of data collection. Compared with the single indicator of the traditional system, which focuses more on a few key indicators such as sales and refueling volume, the present embodiment constructs a comprehensive indicator system including a vehicle flow index, a service efficiency index, and a business capacity index. Each index is further subdivided into multiple indicators, such as the off-station vehicle conversion rate and the in-station vehicle flow fluctuation coefficient in the vehicle flow index. The operation of the refueling station is comprehensively evaluated from multiple dimensions, which makes up for the shortcomings of the traditional indicator system, thereby achieving the effect of improving the comprehensiveness of the operation evaluation. Compared with traditional indicators, which are mostly static and cannot be adjusted according to market changes and actual operations, the present embodiment constructs a comprehensive indicator system including a vehicle flow index, a service efficiency index, and a business capacity index. Each index is further subdivided into multiple indicators, such as the off-station vehicle conversion rate and the in-station vehicle flow fluctuation coefficient in the vehicle flow index. The indicator system of this embodiment can dynamically adjust the calculation method, reference value and weight of the indicator according to factors such as different time periods, date types, surrounding traffic and business environment, so that the indicator is closer to the actual operating conditions, improves the accuracy and effectiveness of the evaluation, and can achieve the effect of enhancing environmental adaptability and the scientific nature of operational decisions; compared with the traditional system, which lacks self-learning and adaptive capabilities and cannot automatically adjust model parameters according to historical data and market changes, the analysis model of this embodiment has self-learning and dynamic adjustment capabilities, and realizes adaptive weight adjustment through genetic algorithms, optimizing weights with the goal of minimizing the mean square error between actual operating efficiency and model prediction values; uses Q-learning algorithm for continuous learning and optimization, automatically adjusts operating strategies according to market changes, and realizes the granting of self-learning and adaptive capabilities, so that the model can continuously adapt to changes in the market environment and the development needs of the energy station itself.

[0203] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0204] Based on the same inventive concept, the embodiment of the present application also provides a system for analyzing the operating capacity of a fueling station for implementing the above-mentioned method for analyzing the operating capacity of a fueling station. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above-mentioned method, so the specific limitations in one or more embodiments of the system for analyzing the operating capacity of a fueling station provided below can refer to the limitations of the method for analyzing the operating capacity of a fueling station above, and will not be repeated here.

[0205] In the second aspect, the present application provides a system for analyzing the operating capacity of a fueling station, such as Figure 4 As shown, the energy station operation capability analysis system includes: The data collection module 100 is used to obtain the operation data in the preset area; the operation data at least includes vehicle flow data, passenger flow data and customer interaction data; The data analysis module 200 is used to calculate a local operation index based on the operation data; the local operation index includes at least a vehicle flow index, a service efficiency index and an operating capacity index; and calculate a comprehensive operation index of the energy station based on a preset weight combination and the local operation index; The data display module 300 is used to generate an operation report based on the operation data, the local operation index and the comprehensive operation index, and to display the operation report.

[0206] Each module in the above-mentioned energy station operation capacity analysis system can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0207] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0208] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0209] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for analyzing the operating capacity of a fueling station, characterized in that: The energy station operation capacity analysis method comprises: Acquiring operation data within a preset area; the operation data at least includes vehicle flow data, passenger flow data, and customer interaction data; Based on the operation data, a local operation index is calculated; the local operation index at least includes a traffic flow index, a service efficiency index and an operating capacity index; Calculating a comprehensive operation index of the energy station based on a preset weight combination and the local operation index; An operation report is generated based on the operation data, local operation index and comprehensive operation index, and the operation report is displayed.

2. The method for analyzing the operating capacity of a fueling station according to claim 1, characterized in that: The obtaining of operation data in a preset area includes: Acquire video data collected by a plurality of video acquisition devices in a preset area, and customer interaction data collected by at least one customer interaction system; Based on the video data, vehicle information recognition and customer information recognition are performed to obtain vehicle flow data and passenger flow data; Based on the vehicle flow data, the pedestrian flow data and the customer interaction data, operation data within a preset area is determined.

3. The method for analyzing the operating capacity of a fueling station according to claim 2, characterized in that: The determining of the operation data in the preset area based on the vehicle flow data, the pedestrian flow data and the customer interaction data includes: Performing data source alignment and standardization on the vehicle flow data, the pedestrian flow data, and the customer interaction data to obtain standardized operation data; Extracting features from the standardized operation data to obtain a plurality of operation features corresponding to each collection device; Based on the weighted voting method, the same operation characteristics of multiple collection devices are fused at the decision level to obtain fused operation data; Based on the isolation forest algorithm, abnormal data is detected and filtered on the fused operation data to obtain the operation data in the preset area.

4. The method for analyzing the operating capacity of a fueling station according to claim 1, characterized in that: The calculating of the local operation index based on the operation data comprises: Based on the traffic flow data, a plurality of traffic flow sub-indexes are calculated; the traffic flow sub-indexes include one or more of the following: the conversion rate of vehicles outside the station, the fluctuation coefficient of vehicle flow entering the station, and the trend index of vehicle flow entering the station; A traffic flow index is determined based on the plurality of traffic flow sub-indices and preset traffic flow index weights.

5. The method for analyzing the operating capacity of a fueling station according to claim 1, characterized in that: The calculating of the local operation index based on the operation data comprises: Based on the passenger flow data and the customer interaction data, a plurality of service efficiency sub-indices are calculated; the service efficiency sub-indices include one or more of an intelligent queuing time evaluation index, a service process efficiency index, and a customer satisfaction feedback index; A service efficiency index is determined based on the plurality of service efficiency sub-indices and preset service efficiency index weights.

6. The method for analyzing the operating capacity of a fueling station according to claim 1, characterized in that: The calculating of the local operation index based on the operation data comprises: Based on the customer interaction data, a plurality of business capability sub-indexes are calculated; the business capability sub-indexes include one or more of an energy sales efficiency index, a non-oil product business performance index, and a customer loyalty and expansion index; Based on the plurality of business capability sub-indices and preset business capability index weights, a business capability index is determined.

7. The method for analyzing the operating capacity of a fueling station according to claim 1, characterized in that: The operation data includes historical operation data and real-time operation data; the preset weight combination is a dynamic weight combination; and the calculation of the comprehensive operation index of the energy station based on the preset weight combination and the local operation index includes: Generate current forecasted operating data based on historical operating data; Based on the predicted operation data and the real-time operation data, a dynamic weight combination is determined; the dynamic weight combination includes a dynamic weight corresponding to each type of the real-time operation data.

8. The method for analyzing the operating capacity of a fueling station according to claim 7, characterized in that: The determining of a dynamic weight combination based on the predicted operation data and the real-time operation data comprises: Acquire an initial weight combination; the initial weight combination includes an initial weight corresponding to each of the local operation indexes; Calculating a mean square error between the predicted operation data and the real-time operation data; Based on the mean square error and the genetic algorithm, a tournament selection method is used to select at least two initial weight combinations as parent weight combinations; Performing a single-point crossover operation on the at least two initial weight combinations and applying Gaussian noise to obtain an updated weight combination; Based on multiple iterative optimizations, the weight combination with the smallest mean square error is selected as the dynamic weight combination.

9. The method for analyzing the operating capacity of a fueling station according to claim 7, characterized in that: The generating of the operation report based on the operation data, the local operation index and the comprehensive operation index, and displaying the operation report includes: Based on the reinforcement learning algorithm, the operation data, the local operation index, the comprehensive operation index and the dynamic weight combination, training is performed to obtain a strategy mapping relationship; the strategy mapping relationship includes a mapping relationship between the operation status corresponding to the real-time operation data and the strategy recommendation; Based on the real-time operation data and the strategy mapping relationship, obtaining a strategy suggestion corresponding to the real-time operation data; An operation report is generated based on the operation data, local operation index, comprehensive operation index and the strategy recommendation, and the operation report is displayed.

10. A fueling station operation capacity analysis system, characterized in that: The energy station operation capability analysis system comprises: A data collection module, used to obtain operation data in a preset area; the operation data at least includes vehicle flow data, passenger flow data and customer interaction data; A data analysis module, for calculating a local operation index based on the operation data; the local operation index at least includes a traffic flow index, a service efficiency index and an operating capacity index; and calculating a comprehensive operation index of the energy station based on a preset weight combination and the local operation index; The data display module is used to generate an operation report based on the operation data, local operation index and comprehensive operation index, and to display the operation report.

Images