Renting and selling locker system based on RFID

Abstract

The invention relates to the technical field of automatic vending, in particular to an RFID-based renting and selling locker system, which comprises a user behavior capture module, a behavior analysis module, a demand prediction module, an article recommendation module, an inventory dynamic management module and a dynamic pricing module. According to the method, information and timestamps are recorded through the RFID reading equipment every time articles are stored and taken, the accuracy and practicability of data collection are improved, activity data are divided into a plurality of time periods through a time sequence decomposition method, comprehensive analysis is carried out, the understanding of user behaviors is enhanced, and the user experience is improved. The prediction model can accurately calculate the use probability of the storage cabinet in a period of time in the future, goods recommendation is performed according to a prediction result and user demands, the personalized demands of users can be accurately met, the user satisfaction degree is increased, market demand fluctuation is effectively dealt with through real-time monitoring and inventory state updating and in combination with a dynamic pricing strategy, and the user experience is improved. And the optimal configuration and utilization efficiency of resources are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic vending, and in particular to an RFID-based locker rental and sales system. Background Art

[0002] The field of vending technology involves systems and methods for automatically distributing and selling goods using machines. These technologies allow consumers to purchase goods without human involvement. Common applications include beverage and snack vending machines, self-service ticket machines, and retail automation solutions. In recent years, vending technology has evolved to include intelligent systems and Internet of Things technologies, which improve user experience and operational efficiency through Internet connectivity, mobile payment solutions, and optimized user interfaces. With the development of artificial intelligence and machine learning, vending machines are able to provide personalized recommendations, further enhancing user interaction.

[0003] Among them, the RFID locker rental and sales system refers to a system that uses radio frequency identification technology (RFID) to manage the rental and sales of lockers. This system allows users to store and access items without human intervention through RFID cards or RFID-enabled devices. The system is used in public places such as swimming pools, gyms, large shopping malls, etc., providing users with safe and convenient storage solutions. The use of RFID technology not only speeds up the transaction process, but also improves the security of the system because the use status of each locker can be accurately tracked and controlled.

[0004] Existing vending technology focuses on simple product distribution and lacks a deep understanding of user behavior and demand changes, limiting the personalization and efficiency of services. In actual applications, traditional vending machines are unable to adjust product supply and prices based on real-time user needs and market changes, resulting in uneven resource allocation, inventory backlogs or out-of-stock phenomena. The lack of effective user interaction and personalized services makes it difficult to form in-depth insights into user behavior, affecting user experience and the overall operational efficiency of the system. For example, during high-demand periods and specific events, inventory and pricing strategies cannot be dynamically adjusted, resulting in decreased customer satisfaction and loss of business opportunities. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a RFID-based locker rental and sales system.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a RFID-based locker rental and sales system includes: The user behavior capture module sets up an RFID reading device at the locker, and records the RFID information and timestamp of the item every time the user deposits or withdraws an item, generating a record of the user's activity; The behavior analysis module extracts the timestamp and user ID from the user activity record, divides the data into multiple time periods using the time series decomposition method, and summarizes the access frequencies in the multiple time periods to obtain access frequency data; The demand prediction module uses the access frequency data, combined with the normal access mode, to calculate the probability of locker rental and sale in the next week through a trend analysis model, and outputs the predicted value to obtain the usage prediction result; The item recommendation module analyzes the matching degree between item attributes and user behavior patterns based on the usage prediction results and the user's storage needs, and identifies potential demand items to obtain a customized recommendation list; The inventory dynamic management module monitors the RFID reading of the locker in real time according to the customized recommendation list and the inventory status of the locker, analyzes the access frequency and duration of the items, updates the inventory status information, and obtains the inventory status record; The dynamic pricing module analyzes market demand fluctuations based on item usage and remaining data in the inventory status record, automatically adjusts the prices of locker items, and obtains pricing update information.

[0007] As a further solution of the present invention, the user activity record includes the RFID identification of the item, the timestamp of the item access, and the ID of the operating user; the access frequency data includes the number of user visits, the access time points, and the peak time periods of user visits within differentiated time periods; the usage prediction results include the usage probability of the locker every day in the next week, the peak demand period, and the trough demand period; the customized recommendation list includes the types of recommended items, the reasons for the recommendation, and the expected user acceptance; the inventory status record includes the real-time types, quantities, and access frequencies of items in each locker; the pricing update information includes the adjusted item prices, the reasons for the price changes, and the expected market response.

[0008] As a further solution of the present invention, the user behavior capture module includes: The RFID reading submodule installs an RFID reading device at the locker to identify and record the RFID tag information when storing and accessing items, while verifying the item type and status of each RFID tag and generating a tag identification record; The time recording submodule captures the time point of each RFID tag scanning through the tag identification record, stores the timestamp and the tag identification record in correspondence, performs data formatting processing, and generates a timestamp record; The data integration submodule removes duplication and verifies the data based on the timestamp record and the tag identification record, verifies the integrity of the data, and generates a user activity record.

[0009] As a further solution of the present invention, the behavior analysis module includes: The key field extraction submodule extracts key fields from the user activity record, including timestamp and user ID, verifies the integrity of the data, and generates a data extraction record; The time series segmentation submodule extracts records based on the data, applies a time series decomposition method, segments the time data according to preset time intervals, checks that the data of each time period is independent and corresponds to the target user behavior, and generates time period segmentation records; The frequency summary submodule analyzes the data in the time period segmentation records, counts the access frequency of users in multiple time periods, summarizes and analyzes the data, performs iterative behavior analysis and pattern recognition, and generates access frequency data.

[0010] As a further solution of the present invention, the demand forecasting module includes: The consistency verification submodule verifies the consistency and compatibility of the data through the access frequency data combined with the normal access mode, and generates a fused data set by screening and comparing the data modes of different time periods; The periodicity analysis submodule builds a trend analysis model based on the fused data set, performs time series analysis on the data, identifies and extracts key trends, analyzes user behavior changes and periodicity, and generates trend analysis results; The probability calculation submodule uses the trend analysis results to calculate the rental and sale usage probabilities of lockers in differentiated time periods in the next week, verifies the consistency of the calculation process and the reliability of the data, and generates usage prediction results.

[0011] As a further solution of the present invention, the item recommendation module includes: The demand information identification submodule uses the usage prediction results and a clustering algorithm to analyze the user's demand trends and preferences for differentiated items, identify the user's demand changes in the next week, and generate demand analysis results; The matching degree analysis submodule compares the matching degree between the item attributes and the user behavior pattern based on the demand analysis result, identifies the item features that are highly related to the user demand, and generates an attribute matching list; The item recommendation submodule uses the attribute matching list to identify the user's potential required items and generates a customized recommendation list based on the matching degree and user behavior prediction information.

[0012] As a further solution of the present invention, the formula of the clustering algorithm is as follows: ; in, It is the user demand change trend index. Representative Each user’s storage demand data point, Represents the total number of users, Represents the mean value of user storage demand data, represents the standard deviation of the data, and is the weight parameter.

[0013] As a further solution of the present invention, the inventory dynamic management module includes: The inventory monitoring submodule uses the RFID reading device in the locker to monitor the RFID tag status of each item in real time based on the customized recommendation list, and records each access behavior of the item, the real-time and consistency of the data, and generates real-time monitoring records; The access frequency identification submodule uses the real-time monitoring records to analyze the access frequency of the items and the duration of each access, and calculates the usage frequency and average usage duration of the items to generate a frequency analysis result; The status update submodule updates the inventory status in real time according to the frequency analysis result and the current locker inventory status and the item status information, checks the matching degree between the inventory information and the item usage, and generates an inventory status record.

[0014] As a further solution of the present invention, the dynamic pricing module includes: The market demand analysis submodule extracts the usage frequency and remaining quantity data of items from the inventory status record, uses time series analysis technology, combines market trends and consumer behavior analysis information, calculates market demand fluctuations at future time points, and generates market fluctuation analysis results; The price optimization submodule automatically calculates the price of the locker items based on the market fluctuation analysis results, automatically adjusts the price to match the market demand with reference to the market competitiveness and scarcity of the items, and generates a price adjustment record; The label adjustment submodule uses the price adjustment record to update the pricing information of the locker, verify the update of the price label of each item, and synchronize with the market adjustment to generate pricing update information.

[0015] As a further solution of the present invention, the formula of the time series analysis technology is as follows: ; in, For future time points Market demand fluctuations, Representing time point The frequency of use of items, Representing time point The remaining quantity of items, is the influence coefficient of market trend, represents the coefficient of the changing trend of consumer behavior, is the historical demand average, is the base offset of the demand, is a natural constant.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by upgrading the vending technology, especially by applying RFID technology in the rental and sale locker system for innovative optimization, significant beneficial effects are brought. In actual operation, the RFID reading device is used to record information and timestamps each time an item is stored or retrieved, which can effectively generate detailed user activity records. Such data collection improves the accuracy and practicality of the data. Through the time series decomposition method, the activity data is divided into multiple time periods and a comprehensive analysis is performed, which enhances the understanding of user behavior and enables the prediction model to accurately calculate the probability of using the locker in the future. According to the prediction results and user needs, item recommendations can accurately meet the personalized needs of users and increase user satisfaction. By real-time monitoring and updating of inventory status, combined with dynamic pricing strategies, the system can effectively respond to market demand fluctuations and ensure the optimal allocation and utilization efficiency of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a system flow chart of the present invention; Figure 2 It is a schematic diagram of the system framework of the present invention; Figure 3 It is a flow chart of the user behavior capture module of the present invention; Figure 4 is a flow chart of the behavior analysis module of the present invention; Figure 5 It is a flow chart of the demand forecasting module of the present invention; Figure 6 This is a flow chart of the item recommendation module of the present invention; Figure 7 It is a flow chart of the inventory dynamic management module of the present invention; Figure 8 This is a flow chart of the dynamic pricing module of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0020] See also Figure 1 to Figure 2 , a RFID-based locker rental and sales system includes: The user behavior capture module sets up an RFID reading device at the locker, and records the RFID information and timestamp of the item every time the user deposits or withdraws an item, generating a record of the user's activity; The behavior analysis module extracts timestamps and user IDs from user activity records, divides the data into multiple time periods using the time series decomposition method, and summarizes the access frequencies within multiple time periods to obtain access frequency data; The demand forecasting module uses access frequency data, combined with normal access patterns, to calculate the probability of locker rental and sale usage in the next week through a trend analysis model, and outputs the predicted value to obtain the usage forecast result; The item recommendation module analyzes the matching degree between item attributes and user behavior patterns based on the usage prediction results and the user's storage needs, and identifies potential items to obtain a customized recommendation list; The inventory dynamic management module monitors the RFID readings of the lockers in real time based on the customized recommendation list and the inventory status of the lockers, analyzes the access frequency and duration of items, updates the inventory status information, and obtains the inventory status record; The dynamic pricing module analyzes market demand fluctuations based on item usage and remaining data in inventory status records, automatically adjusts the prices of locker items, and obtains pricing update information.

[0021] User activity records include the RFID identification of items, timestamps of item access, and the ID of the operating user. Access frequency data include the number of user visits, access time points, and user access peak periods within differentiated time periods. Usage prediction results include the probability of daily use of lockers in the next week, peak demand periods, and trough demand periods. The customized recommendation list includes the types of recommended items, reasons for recommendation, and expected user acceptance. Inventory status records include the real-time types, quantities, and access frequencies of items in each locker. Pricing update information includes adjusted item prices, reasons for price changes, and expected market response.

[0022] See also Figure 2 , 3 , the user behavior capture module includes: The RFID reading submodule installs an RFID reading device at the locker to identify and record the RFID tag information when storing and accessing items, and at the same time verifies the item type and status of each RFID tag. The execution process of generating tag identification records is as follows; RFID reading equipment is installed at the locker. Its main function is to use RFID technology to identify and record RFID tag information when storing and accessing items, and verify the item type and status of each RFID tag. The process includes reading the RFID tag information, matching and verifying the scanned data with the tag data stored in the database, confirming whether the item type is consistent with the system record, and checking whether the current status of the item allows access. This operation is completed through an item identification process. In actual implementation, the process includes using special software to decode the received RFID signal. The decoded data is further processed to match the pre-stored information in the database. The entire process ensures the correct classification and status update of the items and generates tag identification records.

[0023] The time recording submodule captures the time point of each RFID tag scanning through tag identification records, stores the timestamp and tag identification records in correspondence, and performs data formatting. The execution process of generating timestamp records is as follows; By capturing the time point of each RFID tag scan, the submodule uses an internal clock to work with the RFID reading device to record the specific time point of each item access. The process includes obtaining timestamps from the RFID reading device and associating these timestamps with the tag identification records received from the RFID device. Through a formatting process, including using database technology to store and index the timestamp data, it is ensured that the records of any specific time point can be quickly retrieved to provide support for further data processing and analysis. The formatting process ensures the consistency and traceability of the data and generates a timestamp record.

[0024] The data integration submodule deduplicates and verifies the data based on timestamp records and tag identification records, verifies the integrity of the data, and generates user activity records. The execution process is as follows; Combine the tag identification record and the timestamp record, according to the formula , calculate the user activity record. In the formula, Represents the calculated user activity record, Representative Tag identification records, Represents the corresponding timestamp record. Detailed explanation of the formula and the process of formula calculation: There are five item access events, and the tag identification records of each event are , the timestamp records are . Then user activity records The calculation process is as follows: The results show that the data were correctly combined and calculated during the integration process, ensuring the accuracy and completeness of user activity records.

[0025] See also Figure 2 , 4 , the behavior analysis module includes: The key field extraction submodule extracts key fields from user activity records, including timestamps and user IDs, verifies the integrity of the data, and generates data extraction records. The execution process is as follows; Extract timestamp and user ID from user activity records according to the formula , generate data extraction records. In the formula, represents the quantitative score of the data extraction record, Represents the number of key fields extracted, Represents the number of consistency verifications of the data in the field. Formula explanation and formula calculation derivation process: Set the number of key fields to be extracted (timestamp and user ID), and verify the data consistency of each field 3 times, that is, The calculation process of data extraction record is as follows: .

[0026] The results show that the integrity and reliability of data extraction records are ensured through effective extraction and consistency verification of key fields.

[0027] The time series segmentation submodule is based on data extraction records and applies the time series decomposition method to segment the time data according to the preset time intervals, and checks that the data in each time period is independent and corresponds to the target user behavior. The execution process of generating time period segmentation records is as follows; The time series decomposition method is used to segment the time data in the data extraction record according to the preset time interval. The process includes reading the timestamps in the data extraction record, assigning these timestamps to different time periods according to the preset time intervals, and dynamically adjusting the time period to optimize the accuracy of data segmentation, ensuring that the data in each time period can independently reflect the corresponding user behavior and generate time period segmentation records.

[0028] The frequency summary submodule analyzes the data in the time segmentation records, counts the user's access frequency in multiple time periods, summarizes and analyzes the data, performs iterative behavior analysis and pattern recognition, and generates the access frequency data. The execution process is as follows; The time period segmentation records are analyzed and the access frequency of users in multiple time periods is counted. In this process, the access data of each time period is extracted from the time period segmentation records, and these data are classified and summarized. By comparing and analyzing the access frequency of different time periods, the data is iteratively analyzed using statistical methods. The analysis process includes the identification of behavioral patterns and the evaluation of the changing trend of access frequency. Through the iterative analysis method, user behavior patterns can be accurately identified and access frequency data can be generated.

[0029] See also Figure 2 , 5 , the demand forecasting module includes: The consistency check submodule verifies the consistency and compatibility of data by accessing frequency data and combining normal access patterns. The execution process of generating a fused data set by screening and comparing data patterns in differentiated time periods is as follows; Consistency and compatibility verification is performed through access frequency data. This process specifically includes analyzing the central trend and dispersion of access frequency data, taking the normal access pattern as a reference benchmark, and achieving the purpose of data consistency verification by comparing the data patterns of differentiated time periods, and screening out abnormal data that deviates from the normal pattern. This process uses statistical analysis methods, such as variance analysis and skewness analysis, to evaluate and classify the data in each time period to ensure that only data that conforms to the normal pattern can be further analyzed and used to generate a fused data set.

[0030] The periodicity analysis submodule builds a trend analysis model based on the fused data set, performs time series analysis on the data, identifies and extracts key trends, analyzes user behavior changes and periodicity, and generates trend analysis results. The execution process is as follows; Based on the fused data set, a trend analysis model is constructed. The construction process of this model includes selecting appropriate time series analysis tools, such as ARIMA or seasonally decomposed time series, analyzing the time data in the fused data set, and identifying key trends and cyclical changes through the model. This analysis process helps to reveal the cyclical laws and changing trends of user behavior, and describes in detail the main trends, changing cycles and future directions in the data, providing a key perspective and basis for understanding user behavior and generating trend analysis results.

[0031] The probability calculation submodule uses the trend analysis results to calculate the probability of locker rental and sale usage in different time periods in the next week, verifies the consistency of the calculation process and the reliability of the data, and generates the execution process of the usage prediction results as follows; Using the trend analysis results, according to the formula , calculate the probability of renting and selling lockers in different time periods in the next week. represents the probability of use, Represents the average rate of events occurring within a time period, Represents the time period, Represents the total time. Formula explanation and formula calculation derivation process: Set the average rate of events occurring within the time period times / day, time period Days, total time days. Then use the probability The calculation process is as follows: This result shows that the generation of prediction results is based on scientific computing, ensuring the consistency of the calculation process and the reliability of the data.

[0032] See also Figure 2 , 6 , the item recommendation module includes: The demand information identification submodule uses the usage prediction results and adopts clustering algorithms to analyze the user's demand trends and preferences for differentiated items, identify the user's demand changes in the next week, and generate the demand analysis results. The execution process is as follows; The formula of the clustering algorithm is as follows: ; in, It is the user demand change trend index. Representative Each user’s storage demand data point, Represents the total number of users, Represents the mean value of user storage demand data, represents the standard deviation of the data, and is the weight parameter.

[0033] Detailed explanation of the formula and the process of formula calculation and derivation: set up The first For example, the storage data of 20 users are collected through user behavior monitoring software. is the total number of users, here .

[0034] is the mean of all users’ storage demand data, is the standard deviation of these data. units, These values ​​reflect the average level and variation of users' storage needs.

[0035] and is a weight parameter, which is set based on historical data analysis and expert opinions. Here, and , reflecting the influence of mean and standard deviation on the results.

[0036] Next, enter the specific data into the formula for calculation. Set the collected user data points for .

[0037] Calculate the absolute value of the difference between each data point and the mean, take the square root, and multiply by : For each

[0038] The calculations are as follows: (Calculate the remaining data points in the same way) Add up all the results and divide by : Sum ;

[0039] Calculate the standard deviation part: ; Combining the two parts results in: ; The results show that according to the defined weights and the collected data, the calculated user demand change trend index is 1.04, which reflects the degree of change in user storage demand after weighting. The weight parameters are set based on the analysis of the variability of historical data to ensure that the calculation results are in line with the actual situation and can flexibly reflect the slight changes in user demand.

[0040] The matching degree analysis submodule compares the matching degree between item attributes and user behavior patterns based on the demand analysis results, identifies item features that are highly correlated with user needs, and generates an attribute matching list. The execution process is as follows; Compare item attributes with user behavior patterns, according to the formula , generate attribute matching list. In the formula, Represents the degree of matching, Represents the weight of the item's attributes, The score representing the user's behavior pattern. Detailed explanation of the formula and the process of formula calculation: Suppose there are three item attributes, and their weights are , the score corresponding to the user behavior pattern is . Then the attribute matching degree The calculation process is as follows: This result shows that by quantifying the matching degree between item attributes and user behavior patterns, the generated attribute matching list can accurately reflect the correlation degree between user needs and item characteristics.

[0041] The item recommendation submodule uses the attribute matching list to identify the user's potential required items, and generates a customized recommendation list based on the matching degree and user behavior prediction information. The execution process is as follows; The operation is performed using the attribute matching list. According to the matching degree between the item attributes and the user behavior pattern and the user behavior prediction information, the process includes screening out items from the attribute matching list that are highly consistent with the user behavior prediction, evaluating the matching degree of each item, and sorting them according to the matching degree. It not only reflects the user's current behavior trend, but also predicts the user's potential needs. By comprehensively analyzing the user's historical behavior data and potential needs, it provides personalized item recommendations, ensures the accuracy of the recommendations and user satisfaction, and generates a customized recommendation list.

[0042] See also Figure 2 , 7 , the inventory dynamic management module includes: The inventory monitoring submodule uses the RFID reading device in the locker to monitor the RFID tag status of each item in real time based on the customized recommendation list, and records the access behavior of each item, the real-time and consistency of the data, and generates the execution process of real-time monitoring records as follows; Use customized recommendation lists to monitor items in real time. Specific operations include using RFID readers in lockers to monitor the status of each item’s RFID tag in real time. During this process, each access behavior of the item is recorded to ensure the real-time and consistency of the data. The status changes of each item’s tag are captured through RFID technology and updated synchronously to the monitoring system. The monitoring mechanism continuously tracks the status of items, provides real-time data streams, and generates real-time monitoring records.

[0043] The access frequency identification submodule uses real-time monitoring records to analyze the access frequency of items and the duration of each access, and calculates the usage frequency and average usage duration of items. The execution process of generating frequency analysis results is as follows; Using real-time monitoring records, according to the formula , calculate the usage frequency and average usage time of the item. In the formula, Represents the average usage time, Representative The duration of the access, Represents the number of accesses. Detailed explanation of the formula and the process of formula calculation: There are five access actions, and the durations are minutes. The average usage time The calculation process is as follows: The results show that the accuracy of the frequency analysis results and their practicality for inventory management can be ensured through detailed analysis and calculation of the access frequency.

[0044] The status update submodule updates the inventory status in real time based on the frequency analysis results and the current locker inventory status, according to the item status information, checks the matching degree between the inventory information and the item usage, and generates the inventory status record. The execution process is as follows; The inventory status is updated in real time based on the frequency analysis results and the current locker inventory status, and the degree of match between inventory information and item usage is compared. Through regular or real-time status checks, the inventory status records are updated to ensure that the inventory information always reflects the latest item status and availability. This update mechanism improves the efficiency and accuracy of inventory management by dynamically monitoring and adjusting inventory, providing accurate and timely data support for inventory management, ensuring the efficient operation of the inventory system and the continuity of item supply, and generating inventory status records.

[0045] See also Figure 2 , 8 , the dynamic pricing module includes: The market demand analysis submodule extracts the usage frequency and remaining quantity data of items from the inventory status record, uses time series analysis technology, combines market trends and consumer behavior analysis information, calculates the market demand fluctuations at future time points, and generates the execution process of market fluctuation analysis results as follows; The formula for the time series analysis technique is as follows: ; in, For future time points Market demand fluctuations, Representing time point The frequency of use of items, Representing time point The remaining quantity of items, is the influence coefficient of market trend, represents the coefficient of the changing trend of consumer behavior, is the historical demand average, is the base offset of the demand, is a natural constant.

[0046] Detailed explanation of the formula and the process of formula calculation: This formula calculates the market demand fluctuation, where Represents the predicted market demand at time point t. The specific calculation and data sources of the parameters are as follows: : Time point The frequency of item usage is determined through historical data analysis and is set at 500 times per day.

[0047] : Time point The remaining quantity of the item is obtained from the inventory management system and is set to 150 pieces.

[0048] : Used to adjust the weight of the time coefficient, calculated based on the rate of change of historical data and set to 0.03.

[0049] : The coefficient of the changing trend of consumer behavior is calculated based on the consumer behavior analysis tool and is set to 1.2.

[0050] : The influence coefficient of market trend is set according to market research data and the value is 0.5.

[0051] : The historical demand average, calculated from sales records, is set to 300 times.

[0052] : The baseline offset of demand is set to 50 to balance the error between historical data and the forecast model.

[0053] Substitute specific values ​​for calculation: calculate .

[0054] calculate .

[0055] Substitute into the formula .

[0056] Result explanation: The result shows that the predicted market demand at time point t is 668.14 times. This value reflects the comprehensive forecast of market demand based on the analysis of the current frequency of item use, remaining quantity, and market and consumer behavior trends. This calculation is based on actual monitored data and market research, rather than settings, to ensure the accuracy and reliability of the forecast.

[0057] The price optimization submodule automatically calculates the price of locker items based on the market fluctuation analysis results, and automatically adjusts the price to match the market demand with reference to the market competitiveness and item scarcity. The execution process of generating price adjustment records is as follows; Automatically calculate the price of locker items according to the formula , automatically adjusts the price to match market demand. represents the adjusted price, Represents the base price, represents the market competitiveness adjustment factor, Represents market demand, Represents the item scarcity adjustment factor, Represents the current scarcity of the item. Detailed explanation of the formula and the process of formula calculation: Setting the base price Yuan, market competitiveness adjustment factor , market demand , item scarcity adjustment factor , the current scarcity of the item The adjusted price is The calculation process is as follows: The results show that through the quantitative analysis of market demand and item scarcity, prices can be effectively adjusted to match market dynamics, ensuring the rationality of price adjustments and market competitiveness.

[0058] The label adjustment submodule uses the price adjustment record to update the locker pricing information, verify the update of each item's price label, and synchronize with the market adjustment to generate the execution flow of pricing update information as follows; Use price adjustment records to update the pricing information of lockers. The process includes extracting the adjusted price of each item from the price adjustment record, verifying and replacing the price tags displayed in the lockers to ensure that the price information is synchronized with the market adjustment. The update process is carried out through an automated system to ensure accurate transmission and real-time updates of information, providing customers and managers with instant and accurate pricing data, supporting the commercial operations and market adaptability of lockers, and generating pricing update information.

[0059] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A RFID-based locker rental and sales system, characterized in that: The system comprises: The user behavior capture module sets up an RFID reading device at the locker, and records the RFID information and timestamp of the item every time the user deposits or withdraws an item, generating a record of the user's activity; The behavior analysis module extracts the timestamp and user ID from the user activity record, divides the data into multiple time periods using the time series decomposition method, and summarizes the access frequencies in the multiple time periods to obtain access frequency data; The demand prediction module uses the access frequency data, combined with the normal access mode, to calculate the probability of locker rental and sale in the next week through a trend analysis model, and outputs the predicted value to obtain the usage prediction result; The item recommendation module analyzes the matching degree between item attributes and user behavior patterns based on the usage prediction results and the user's storage needs, and identifies potential demand items to obtain a customized recommendation list; The inventory dynamic management module monitors the RFID reading of the locker in real time according to the customized recommendation list and the inventory status of the locker, analyzes the access frequency and duration of the items, updates the inventory status information, and obtains the inventory status record; The dynamic pricing module analyzes market demand fluctuations based on item usage and remaining data in the inventory status record, automatically adjusts the prices of locker items, and obtains pricing update information.

2. The RFID-based locker rental and sales system according to claim 1 is characterized in that: The user activity record includes the RFID identification of the item, the timestamp of the item access, and the ID of the operating user; the access frequency data includes the number of user visits, access time points, and user access peak time periods within differentiated time periods; the usage prediction results include the usage probability of the locker every day in the next week, the peak demand period, and the trough demand period; the customized recommendation list includes the types of recommended items, the reasons for the recommendation, and the expected user acceptance; the inventory status record includes the real-time types and quantities of items in each locker, and the access frequency of items; the pricing update information includes the adjusted item prices, the reasons for the price changes, and the expected market response.

3. The RFID-based locker rental and sales system according to claim 1 is characterized in that: The user behavior capture module includes: The RFID reading submodule installs an RFID reading device at the locker to identify and record the RFID tag information when storing and accessing items, while verifying the item type and status of each RFID tag and generating a tag identification record; The time recording submodule captures the time point of each RFID tag scanning through the tag identification record, stores the timestamp and the tag identification record in correspondence, performs data formatting processing, and generates a timestamp record; The data integration submodule removes duplication and verifies the data based on the timestamp record and the tag identification record, verifies the integrity of the data, and generates a user activity record.

4. The RFID-based locker rental and sales system according to claim 1 is characterized in that: The behavior analysis module includes: The key field extraction submodule extracts key fields from the user activity record, including timestamp and user ID, verifies the integrity of the data, and generates a data extraction record; The time series segmentation submodule extracts records based on the data, applies a time series decomposition method, segments the time data according to preset time intervals, checks that the data of each time period is independent and corresponds to the target user behavior, and generates time period segmentation records; The frequency summary submodule analyzes the data in the time period segmentation records, counts the access frequency of users in multiple time periods, summarizes and analyzes the data, performs iterative behavior analysis and pattern recognition, and generates access frequency data.

5. The RFID-based locker rental and sales system according to claim 1 is characterized in that: The demand forecasting module comprises: The consistency verification submodule verifies the consistency and compatibility of the data through the access frequency data combined with the normal access mode, and generates a fused data set by screening and comparing the data modes of different time periods; The periodicity analysis submodule builds a trend analysis model based on the fused data set, performs time series analysis on the data, identifies and extracts key trends, analyzes user behavior changes and periodicity, and generates trend analysis results; The probability calculation submodule uses the trend analysis results to calculate the rental and sale usage probabilities of lockers in differentiated time periods in the next week, verifies the consistency of the calculation process and the reliability of the data, and generates usage prediction results.

6. The RFID-based locker rental and sales system according to claim 1, characterized in that: The item recommendation module includes: The demand information identification submodule uses the usage prediction results and a clustering algorithm to analyze the user's demand trends and preferences for differentiated items, identify the user's demand changes in the next week, and generate demand analysis results; The matching degree analysis submodule compares the matching degree between the item attributes and the user behavior pattern based on the demand analysis result, identifies the item features that are highly related to the user demand, and generates an attribute matching list; The item recommendation submodule uses the attribute matching list to identify the user's potential required items and generates a customized recommendation list based on the matching degree and user behavior prediction information.

7. The RFID-based locker rental and sales system according to claim 6, characterized in that: The formula of the clustering algorithm is as follows: ； in, It is the user demand change trend index. Representative Each user’s storage demand data point, Represents the total number of users, Represents the mean value of user storage demand data, represents the standard deviation of the data, and is the weight parameter.

8. The RFID-based locker rental and sales system according to claim 1, characterized in that: The inventory dynamic management module includes: The inventory monitoring submodule uses the RFID reading device in the locker to monitor the RFID tag status of each item in real time based on the customized recommendation list, and records each access behavior of the item, the real-time and consistency of the data, and generates real-time monitoring records; The access frequency identification submodule uses the real-time monitoring records to analyze the access frequency of the items and the duration of each access, and calculates the usage frequency and average usage duration of the items to generate a frequency analysis result; The status update submodule updates the inventory status in real time according to the frequency analysis result and the current locker inventory status and the item status information, checks the matching degree between the inventory information and the item usage, and generates an inventory status record.

9. The RFID-based locker rental and sales system according to claim 1, characterized in that: The dynamic pricing module includes: The market demand analysis submodule extracts the usage frequency and remaining quantity data of items from the inventory status record, uses time series analysis technology, combines market trends and consumer behavior analysis information, calculates market demand fluctuations at future time points, and generates market fluctuation analysis results; The price optimization submodule automatically calculates the price of the locker items based on the market fluctuation analysis results, automatically adjusts the price to match the market demand with reference to the market competitiveness and scarcity of the items, and generates a price adjustment record; The label adjustment submodule uses the price adjustment record to update the pricing information of the locker, verify the update of the price label of each item, and synchronize with the market adjustment to generate pricing update information.

10. The RFID-based locker rental and sales system according to claim 9, characterized in that: The formula for the time series analysis technique is as follows: ； in, For future time points Market demand fluctuations, Representing time point The frequency of use of items, Representing time point The remaining quantity of items, is the influence coefficient of market trend, represents the coefficient of the changing trend of consumer behavior, is the historical demand average, is the base offset of the demand, is a natural constant.

Images