Internet platform-based payment swim ring smart rack rental and return system

Abstract

This invention discloses a rental and return system for a paid swimming ring smart rack based on an internet platform, relating to the field of paid swimming ring rental technology. The system includes the following steps: during the use of the paid swimming ring smart rack, the number of rentals and the number of swimming rings rented in each time period are recorded to obtain core operational data; this data is preprocessed to obtain effective operational reference data, and then divided into time periods to obtain operational time period reference data; time period prediction analysis is performed to obtain operational time period prediction data for the next day; based on the operational time period prediction data for the next day, the pricing of the paid swimming ring smart rack for each time period is dynamically adjusted for the next day. This invention addresses the problem that existing paid swimming ring rental technologies, when setting rental prices based on paid swimming ring smart racks, cannot analyze core rental data to predict swimming ring usage in each time period and reliably adjust rental prices dynamically.

Description

Technical Field

[0001] This invention relates to the field of paid swimming ring rental technology, specifically a paid swimming ring smart rack rental and return system based on an internet platform. Background Technology

[0002] Paid swimming ring rental technology is designed for water recreation scenarios such as swimming pools, water parks, and beaches. It integrates technologies such as the Internet of Things, embedded control, wireless communication, data acquisition and intelligent analysis, self-service interaction, and electronic payment to realize an intelligent, unmanned, self-service rental system that enables users to pick up and return swimming rings themselves and charge according to regulations, as well as monitor equipment status, collect rental behavior data, and operate the rental system in a refined manner.

[0003] Existing paid swim ring rental technologies, when setting rental prices based on smart pool racks, often employ a uniform fixed price for all time periods, or a fixed price for different time periods. A uniform fixed price is the most basic static pricing model, failing to consider the supply and demand differences based on time periods and scenarios, completely wasting the rental-related data that the smart racks can collect. A uniform fixed price cannot match resource scarcity through price premiums during high-demand periods, and the resource value cannot be fully realized. During low-demand periods, it fails to stimulate user demand, impacting equipment revenue. Fixed prices for different time periods are often determined by human experience, without matching objective rental-related data, potentially leading to discrepancies between pricing and supply and demand. In cases of mismatched pricing, such as lower prices during high-demand periods and higher prices during low-demand periods, and the inability of fixed-time pricing to adapt to normal supply and demand fluctuations, while analyzing multi-dimensional rental data and other relevant data to adjust prices in real time can accurately match the supply and demand of swimming rings during specific periods, it requires powerful backend computing power, making the computational and analytical costs too high for individual paid swimming ring smart racks. Therefore, existing paid swimming ring rental technologies, when setting rental prices based on paid swimming ring smart racks, cannot analyze core rental data, predict swimming ring usage during different time periods, or reliably adjust rental prices dynamically. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the prior art. By analyzing the number of rentals and rentals of paid swimming rings during each time period in the use of the smart paid swimming ring system, core operational data is obtained. This data is then preprocessed to obtain effective operational reference data, and further divided into time periods to obtain operational time period reference data. Time period prediction analysis is then performed to obtain operational time period prediction data for the next day. Finally, the pricing of the paid swimming ring smart system for each time period on the next day is dynamically adjusted. This addresses the problem that existing paid swimming ring rental technologies, when setting rental prices based on the smart paid swimming ring system, cannot analyze core rental data to predict swimming ring usage in each time period and reliably adjust rental prices dynamically.

[0005] To achieve the above objectives, this application provides a rental and return system for paid smart swimming ring racks based on an internet platform, including an operation data collection module, an operation analysis module, a time period prediction module, and a pricing adjustment module;

[0006] The operation data collection module is used to collect the number of times and the number of paid swimming rings rented in each time period during the use of the paid swimming ring smart rack, and obtain the core operation data of the swimming rings.

[0007] The operation analysis module includes a preprocessing unit and a time period analysis unit. The preprocessing unit is used to preprocess the core operation data of the swimming ring to obtain effective operation reference data, and the time period analysis unit is used to perform time period division processing to obtain operation time period reference data.

[0008] The time period prediction module performs time period prediction analysis based on the operating time period reference data to obtain the operating time period prediction data for the next day.

[0009] The pricing adjustment module dynamically adjusts the pricing of the paid smart swimming ring rack for each time period of the next day based on the predicted operating hours data for the next day.

[0010] Furthermore, the operations data collection module is configured with operations data collection strategies, which include:

[0011] Designate any paid smart pool ring rack as the first smart rack, obtain the daily opening hours of the first smart rack, and record them as the rental opening hours;

[0012] The location of the first smart shelf used to store paid swimming rings is recorded as a swimming ring slot. The total number of swimming ring slots owned by the first smart shelf is recorded as the total number of slots.

[0013] The first duration is set to t1, each duration of t1 is a time period, and each time period is evenly divided into k1 sub-time periods, where k1 is the number of sub-time periods set.

[0014] Furthermore, the operational data collection strategy also includes:

[0015] Any sub-period within the rental opening time is designated as the first sub-period, and any period is designated as the first period; the number of times the first smart rack rents a paid swimming ring during the first sub-period is collected and recorded as the rental count of the first sub-period;

[0016] In the first time period, the number of paid swimming rings rented by the first smart rack is collected at the first time interval and recorded as the real-time rental number. The average of all real-time rental numbers is calculated and recorded as the rental number in the first sub-time period, where the first time interval is t2.

[0017] Every day during the rental opening hours, the number of rentals and the number of rentals for each segment are collected and sorted by date and time, which is recorded as the core operational data of the first smart rack.

[0018] Furthermore, the preprocessing unit is configured with a preprocessing strategy, which includes:

[0019] Extract the portion collected in the last k2 days from the core operation data of the swimming ring and record it as the recent operation data. Combine the number of rentals and the number of rentals in the first sub-period of each day in the recent operation data into a data pair, denoted as (x, y), to obtain the data pair set, where k2 is the set number of days, x represents the number of rentals, and y represents the number of rentals.

[0020] Calculate the Euclidean distance between any two data pairs in the dataset to obtain the distance set; obtain the median ME of the distance set; use ME as the neighborhood radius and set e1 as the minimum number of cluster points, where e1 is the set number of points;

[0021] Perform density clustering on the data set and obtain all core points, boundary points, and isolated points. Record all core points as a core cluster.

[0022] Furthermore, the preprocessing strategy also includes:

[0023] Let x / y be the coupling ratio, and let x / y = 1 when y = 0; calculate the coupling ratio of all data pairs in the core cluster, and calculate the mean AP and standard deviation AB of all coupling ratios; denote [AP - k3 × AB, AP + k3 × AB] and 1 as the normal coupling range, where k3 is the set scaling factor;

[0024] Sort the number of leases for all data pairs in the core cluster in ascending order, and obtain the e2 percentile EX1 and e3 percentile EX2. Record [EX1, EX2] as the normal fluctuation range corresponding to the number of leases; and repeatedly obtain the normal fluctuation range corresponding to the number of leases, recorded as [EY1, EY2], where e2 and e3 are the set percentiles.

[0025] Furthermore, the preprocessing strategy also includes:

[0026] Denote any data pair corresponding to a boundary point or an isolated point as the first data pair. If the y of the first data pair is zero and the x is not zero, then determine the x and y of the first data as outliers.

[0027] If the y of the first data pair is not zero and the coupling ratio of the first data pair does not fall within the normal coupling range, then determine the x and y of the first data as outliers.

[0028] If the coupling ratio of the first data pair is within the normal coupling range, but the corresponding x does not fall within [EX1, EX2], and the corresponding y does not fall within [EY1, EY2], and the first data pair is an isolated point, then determine the x and y of the first data as outliers.

[0029] If the x and y of the first data are outliers, then obtain the x and y of the first sub-period that are not outliers in the most recent k4 days, calculate the corresponding average values respectively, and perform replacement, where k4 is the set number of days.

[0030] Repeat the screening of all outliers in the data pair set and perform replacement to obtain the standard operation data for the first sub-period. Obtain the standard operation data for all sub-periods based on the recent operation data to obtain the effective operation reference data.

[0031] Furthermore, the time period analysis unit is configured with a time period analysis strategy, and the time period analysis strategy includes:

[0032] Based on the effective operation reference data, for the first time period of any day, calculate the ratio of the sum of the rental times of the included first sub-periods to the total number of rental times of the day, denoted as the rental volume ratio of the first time period; and calculate the ratio of the maximum value of the number of rental units of the included first sub-periods to the total number of positions, denoted as the time period rental rate. Repeat to obtain the rental volume ratio and the time period rental rate for all time periods.

[0033] Combine the rental volume ratio and the time period rental rate of each time period into a data point, denoted as (VX, VY), where VX represents the rental volume ratio and VY represents the time period rental rate.

[0034] Obtain the 33rd percentile FV1 and the 66th percentile FV2 of the rental volume ratio for all time periods, and obtain the 33rd percentile GV1 and the 66th percentile GV2 of the time period rental rate.

[0035] For any data point, denoted as (AX, AY), if AX ≤ FV1 and AY ≤ GV1, then classify it as a low-segment point. If FV1 < AX < FV2 and GV1 < AY < GV2, then classify it as a flat-segment point. If FV2 ≤ AX and GV2 ≤ AY, then classify it as a high-segment point, otherwise do not classify.

[0036] Furthermore, time-period analysis strategies also include:

[0037] Divide all data points into segments, calculate the mean value LX of VX and the mean value LY of VY for low-segment points, and obtain the center point corresponding to the low-segment point, denoted as the low-peak center (LX, LY); repeatedly obtain the flat-peak center (MX, MY) corresponding to the flat-segment point and the peak center (HX, HY) corresponding to the high-segment point.

[0038] For any data point, calculate the Euclidean distances to (LX, LY), (MX, MY) and (HX, HY) respectively, and mark the data point as a low peak, flat peak or high peak based on the minimum Euclidean distance;

[0039] All data points are marked, and the corresponding time periods in the valid operational reference data are marked as off-peak, off-peak, or peak periods. After completion, the operational time period reference data is obtained.

[0040] Furthermore, the time-period prediction module is configured with a time-period prediction strategy, which includes:

[0041] Retrieve the date type for the second day, including rest days and workdays; retrieve data for all dates of the same type as the second day from the operational period reference data, and record them as valid reference data;

[0042] For the first time period, the low-peak, off-peak, and peak periods corresponding to the first time period in the valid reference data are sorted from oldest to newest date and recorded as the date type sequence of the first time period;

[0043] Get the sum of the position indices of the low-peak periods in the date type sequence of the first period, and record it as the probability score of the low-peak period. Repeat the acquisition of the probability scores of the off-peak and peak periods. Based on the highest probability score, predict the first period of the next day as the corresponding low-peak, off-peak, or peak period.

[0044] Repeat the analysis and forecasting for all time periods on the second day to obtain operational time period forecast data.

[0045] Furthermore, the pricing adjustment module is configured with pricing adjustment strategies, which include:

[0046] Set separate prices for paid swimming rings during peak, off-peak, and low-peak hours; adjust the usage price of paid swimming rings for all time periods of the next day based on the operational period forecast data, and record it as the swimming ring usage price for the next day;

[0047] When the user returns the paid swimming ring on the second day, a corresponding bill is generated based on the user's usage time and the price of the swimming ring, and the user is charged accordingly.

[0048] The beneficial effects of this invention are as follows: This invention obtains core operational data for the paid swimming ring smart racks by tracking the number of rentals and rentals of swimming rings during each time period. This core operational data is preprocessed to obtain effective operational reference data, and then divided into time periods to obtain operational time period reference data. Based on the operational time period reference data, time period prediction analysis is performed to obtain operational time period prediction data for the next day. Based on the operational time period prediction data for the next day, the pricing of the paid swimming ring smart racks for each time period is dynamically adjusted. When setting rental prices for paid swimming rings based on the paid swimming ring smart racks, the core rental data can be analyzed to predict the usage of swimming rings in each time period, and the rental prices can be reliably and dynamically adjusted, thereby increasing the revenue of the paid swimming ring smart racks.

[0049] This invention constructs data pairs from recent dates and uses density clustering to identify core points, boundary points, and isolated points, separating random noise and outlier data to prevent abnormal samples from affecting the division of subsequent time periods. It uses coupling ratio to obtain the relationship between rental frequency and rental quantity, constructs a normal coupling range, and cross-judges it with the normal fluctuation range, enabling more accurate differentiation between true anomalies and normal fluctuations. By sorting the time period types of the same date type within a given time period and obtaining a probability score through position sequence and summation, it can easily and reliably predict the usage of swimming rings in each time period of the next day. Based on the predicted time period situation, it can quickly adjust prices to obtain higher revenue during high-demand periods and promote sales during low-demand periods to increase profitability. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of the system of the present invention;

[0051] Figure 2 This is a flowchart of the steps of the method of the present invention;

[0052] Figure 3 This is a flowchart of the outlier screening process of the present invention;

[0053] Figure 4 This is a schematic diagram of the electronic device of the present invention. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] Example 1, please refer to Figure 1As shown, this application presents a paid smart swimming ring rental and return system based on an internet platform, which includes an operation data collection module, an operation analysis module, a time period prediction module, and a pricing adjustment module.

[0056] The operation data collection module is used to collect the number of times and the number of paid swimming rings rented in each time period during the use of the paid swimming ring smart rack, thereby obtaining the core operation data of the swimming rings;

[0057] The operation data collection module is configured with an operation data collection strategy, which includes: designating any paid swimming ring smart rack as the first smart rack, obtaining the daily opening time of the first smart rack, and recording it as the rental opening time, that is, the time period during which the first smart rack provides paid swimming ring rental services each day;

[0058] The location of the first smart shelf used to store paid swimming rings is recorded as a swimming ring slot. The total number of swimming ring slots owned by the first smart shelf is recorded as the total number of slots.

[0059] The first duration is set as t1, and each t1 duration is a time period. Each time period is evenly divided into k1 sub-time periods, where k1 is the number of sub-time periods. In this embodiment, t1 = 1 hour, that is, each hour is a time period. t1 can be flexibly set according to actual needs, generally from 30 minutes to 2 hours.

[0060] Any sub-period within the rental opening time is designated as the first sub-period, and any period is designated as the first period. The number of times paid swimming rings are rented from the first smart rack during the first sub-period is recorded as the rental count for the first sub-period. That is, the total number of valid rental orders completed by the first smart rack within the sub-period is an absolute value indicator that directly reflects the activity level of user rental behavior during that period.

[0061] In the first time period, the number of paid swimming rings rented by the first smart rack is collected at the first time interval and recorded as the real-time rental number. The average of all real-time rental numbers is calculated and recorded as the rental number in the first sub-time period. The first time interval is t2. In this embodiment, t2 = 1 minute, which can be flexibly set, generally [1 minute, 5 minutes]. The rental number represents the number of swimming rings actually occupied by the smart rack in the same period within the sub-time period. It is an absolute value indicator that directly reflects the actual load on the supply side of the smart rack in that period.

[0062] Every day during the rental opening hours, the number of rentals and the number of rentals for each sub-segment are collected and sorted by time and date, and recorded as the core operational data of the first smart rack.

[0063] In the specific implementation process, the number of rentals can intuitively identify the fluctuations in instantaneous rental demand: the number of rentals reflects the actual usage status of the smart rack and distinguishes between false demand and real demand: if the number of rentals is high in a certain period, but the actual number of swimming rings rented is low, it means that the same batch of swimming rings is being repeatedly picked up and returned, and it is not a high demand period; if the number of rentals is high and the actual number of swimming rings rented is also high, it means that a large number of users are renting at the same time, and it is a high demand period.

[0064] The operation analysis module includes a preprocessing unit and a time period analysis unit. The preprocessing unit is used to preprocess the core operation data of the swimming ring to obtain effective operation reference data, and the time period analysis unit is used to perform time period division processing to obtain operation time period reference data.

[0065] The preprocessing unit is configured with a preprocessing strategy, which includes: (See details) Figure 3 As shown, the data collected in the last k2 days is extracted from the core operation data of the swimming pool and recorded as recent operation data. The number of rentals and the number of rentals in the first sub-period of each day in the recent operation data are combined into data pairs, denoted as (x, y), to obtain the data pair set. The analysis focuses on a single 15-minute sub-period to avoid the interference of the characteristic differences of different time periods on the judgment of outliers and to ensure the relevance of the analysis. Here, k2 is the set number of days, x represents the number of rentals, and y represents the number of rentals. In this embodiment, k2 is 30 days, that is, the data of the most recent 30 days is used as a reference. It can be set flexibly, but it is generally no less than 20 days.

[0066] Calculate the Euclidean distance between any two data pairs in the data set to obtain the distance set; obtain the median ME of the distance set; use ME as the neighborhood radius and set e1 as the minimum number of cluster points, where e1 is the set number of points; in this embodiment, e1 is 6, which can be flexibly set, generally k1 / 5; using the median distance instead of the mean as the neighborhood radius can take advantage of the median's resistance to extreme values ​​and avoid a few outliers from raising the average distance;

[0067] Density clustering was performed on the dataset to obtain all core points, boundary points, and isolated points. All core points were denoted as core clusters. A core point is a point whose neighborhood contains more than or equal to the minimum cluster size, and the neighborhood is a circular area with a radius equal to that point. A boundary point is a point whose neighborhood contains fewer than the minimum cluster size, but falls within the neighborhood of a core point. Isolated points are neither core points nor within the neighborhood of any core point. Boundary points are slightly discrete points around core clusters, mostly representing real data. Isolated points are single points or a few points with no neighborhood data, possibly indicating a data collection error.

[0068] In reality, x and y are strongly coupled. Even with fluctuations, data points will form continuous clusters due to similar features. However, points that are collected incorrectly are isolated because the coupling is broken, and their features are completely disconnected from other points.

[0069] Let x / y be the coupling ratio, and let x / y = 1 when y = 0; calculate the coupling ratio of all data pairs in the core cluster, and calculate the mean AP and standard deviation AB of all coupling ratios; let [AP - k3 × AB, AP + k3 × AB] and 1 be the normal coupling range, where k3 is the set proportional coefficient. In this embodiment, k3 = 2, which can be set flexibly, generally [1.5, 3]; when y = 0, x will also be equal to 0, which is the real business state, and the coupling is valid. Therefore, the coupling ratio is defined as 1.

[0070] The coupling ratio is the direct ratio of x / y and is the core measure of the coupling of the two indicators. The mean of all coupling ratios represents the average coupling level of normal business in that sub-period; the standard deviation reflects the dispersion of the coupling ratio within the core cluster; only when the coupling ratio is within the normal coupling range does it indicate that the data point conforms to the actual situation, that is, the data is normal.

[0071] Sort the number of leases for all data pairs within the core cluster in ascending order, and obtain the e2 percentile EX1 and e3 percentile EX2. Record [EX1, EX2] as the normal fluctuation range corresponding to the number of leases; and repeatedly obtain the normal fluctuation range corresponding to the number of leases, recorded as [EY1, EY2]; where e2 and e3 are set percentiles. In this embodiment, e2=5 and e3=95, which can be flexibly set. e2 is generally [2, 5] and e3 is generally [95, 98].

[0072] Normal coupling only means that the x / y ratio is reasonable. It is also necessary to ensure that the absolute values ​​of x and y conform to the business characteristics of the sub-period to avoid collection errors where the ratio is reasonable but the values ​​are abnormal. Using the 5th percentile and 95th percentile instead of the minimum and maximum values ​​can remove extreme small and extreme large values ​​within the core cluster, ensuring that the interval is the mainstream value range of the sub-period and is more representative.

[0073] Any data pair corresponding to a boundary point or an isolated point is denoted as the first data pair. If the y of the first data pair is zero and the x is not zero, then the x and y of the first data pair are determined to be outliers.

[0074] If the first data pair y is not zero and the coupling ratio of the first data pair is not within the normal coupling range, i.e. the coupling ratio is abnormal, then the x and y of the first data are determined to be abnormal values.

[0075] If the coupling ratio of the first data pair is within the normal coupling range, but the corresponding x is not located in [EX1, EX2] and the corresponding y is not located in [EY1, EY2], the coupling ratio is normal but the value is abnormal, and the first data pair is an isolated point, then the x and y of the first data are judged as outliers; if the first data pair meets the above conditions, but is a boundary point, it may be a real data fluctuation, so it is not judged as an outlier.

[0076] For example, given the data set {(9,8), (9,8), (10,10), (11,11), (12,12), (8,8), (9,9), (10,10), (11,11), (12,12), (8,8), (9,9), (10,10), (11,11), (12,12), (8,8), (9,9), (10,10), (11,11), (12,12), (8,8), (9,9), (10,10), (11,11), (12,12), (8,8), (9,9), (7,7), (13,13), (6,6), (99,2), (100,99), (0,0), (5,10), (11,10)}, density clustering yields the core cluster {(9,8), (9,8), (10,10), (11,11)}. (12, 12), (8, 8), (9, 9), (10, 10), (11, 11), (12, 12), (8, 8), (9, 9), (10, 10), (11, 11), (12, 12), (8, 8), (9, 9), (10, 10), (11, 11), (12, 12), (8, 8), (9, 9), (0, 0)}; Boundary points are {(7, 7), (13, 13), (6, 6), (11, 10)}; Isolated points are {(99, 2), (100, 99), (5, 10)}; The coupling ratio of the core cluster is {1.125, 1.125, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1};

[0077] If AP = 1.01 and AB = 0.04, then the normal coupling range is [0.97, 1.05]; then [EX1, EX2] is [8, 12] and [EY1, EY2] is [8, 12]; then the coupling ratios of (99, 2) and (5, 10) are not within the normal coupling range, the x of (100, 99) is not within [EX1, EX2], and the corresponding y is not within [EY1, EY2] and is an isolated point, then the corresponding x and y are determined to be outliers;

[0078] If x and y of the first data are outliers, then obtain x and y of the first sub-period of the most recent k4 days that are not outliers, calculate the corresponding average values, and replace them, where k4 is the set number of days; in this embodiment, k4=4, which can be set flexibly, generally [2, 6]. Replacing outliers with non-outliers collected based on the most recent date can ensure the integrity of the data;

[0079] Repeatedly filter the data to find all outliers in the collection and replace them to obtain the standard operating data for the first sub-period. Based on the recent operating data, obtain the standard operating data for all sub-periods to obtain effective operating reference data.

[0080] The time period analysis unit is configured with a time period analysis strategy, which includes: based on effective operational reference data, for the first time period of any day, calculate the ratio of the sum of the number of rentals in the first sub-time period to the total number of rentals on that day, and record it as the rental volume ratio of the first time period; and calculate the ratio of the maximum number of rentals in the first sub-time period to the total number of positions, and record it as the time period rental rate. Repeat the acquisition of the rental volume ratio and time period rental rate for all time periods.

[0081] The rental volume ratio is used to quantify the weight of rental demand in a certain period of time in the overall rental demand of the day. It reflects the degree of time concentration of rental demand and is a key indicator for judging the demand heat of a certain period. The higher the ratio, the more concentrated most of the rental demand of the day is in that period, and the more obvious the peak characteristics of demand are.

[0082] The time-based rental rate is used to quantify the proportion of the number of smart racks actually used during a certain time period to the total number of racks. It primarily reflects the saturation level of the rental equipment's rack usage and is a core indicator for judging the supply load during a certain time period. The higher the rental rate, the greater the proportion of racks being used and the tighter the equipment resources.

[0083] Combine the rental volume percentage and rental rate of each time period into a data point, denoted as (VX, VY), where VX represents the rental volume percentage and VY represents the rental rate of the time period.

[0084] Obtain the 33rd percentile FV1 and 66th percentile FV2 of the rental volume share for all time periods, and obtain the 33rd percentile GV1 and 66th percentile GV2 of the rental rate for each time period.

[0085] For any data point, denoted as (AX, AY), if AX ≤ FV1 and AY ≤ GV1, it is classified as a low-segment point; if FV1 < AX < FV2 and GV1 < AY < GV2, it is classified as a flat-segment point; if FV2 ≤ AX and GV2 ≤ AY, it is classified as a high-segment point; otherwise, it is not classified. The continuous VX and VY are divided into three grades of high, medium, and low using the 33rd percentile and the 66th percentile. Then, according to the combined effect of VX and VY, the time period is initially divided into low-segment points, flat-segment points, high-segment points, or not classified, and the percentiles can be flexibly set.

[0086] Fully classify all data points, calculate the mean value LX of VX and the mean value LY of VY for the low-segment points, obtain the central point corresponding to the low-segment points, denoted as the low-peak center (LX, LY); repeat to obtain the flat-peak center (MX, MY) corresponding to the flat-segment points and the high-peak center (HX, HY) corresponding to the high-segment points.

[0087] For any data point, calculate the Euclidean distances to (LX, LY), (MX, MY), and (HX, HY) respectively, and mark the data point as a low-peak point, flat-peak point, or high-peak point according to the minimum Euclidean distance; for example, if the distance of a certain data point to (LX, LY) is the smallest, it is marked as a low-peak point, and if the Euclidean distances are the same, it is marked in the order of high-peak point, flat-peak point, and low-peak point.

[0088] Mark all data points, and mark the corresponding time periods in the effective operation reference data as low-peak time periods, flat-peak time periods, or high-peak time periods. After completion, the operation time period reference data is obtained.

[0089] In the specific implementation process, the corresponding central point is the mean value of each type of point in the (VX, VY) space, and the minimum distance allocation ensures that each time period belongs to the closest category; finally, the operation time period reference data is obtained, that is, the category label of each time period, which is the basis for subsequent prediction and pricing.

[0090] The time period prediction module performs time period prediction analysis based on the operation time period reference data to obtain the operation time period prediction data for the next day.

[0091] The time period prediction module is configured with a time period prediction strategy, and the time period prediction strategy includes: obtaining the date type of the next day, and the date types include rest days and working days; obtaining all the data of the same type as the next day in the operation time period reference data, denoted as the effective reference data; grouping the historical samples by working days and rest days because the user behavior patterns on the same working day or rest day are usually more similar; using only the historical data of the same type as a reference can improve the reliability of the prediction.

[0092] For the first time period, the low-peak, off-peak, and peak periods corresponding to the first time period in the valid reference data are sorted from oldest to newest date and recorded as the date type sequence of the first time period;

[0093] The sum of the position indices of the off-peak periods in the date type sequence of the first time period is recorded as the probability score of the off-peak period. The probability scores of the off-peak and peak periods are obtained repeatedly. Based on the highest probability score, the first time period of the next day is predicted as the corresponding off-peak, off-peak, or peak period. For example, if the date type sequence of the first time period is {off-peak period, off-peak period, off-peak period, off-peak period, off-peak period, peak period, off-peak period}, then the probability scores of the off-peak, off-peak, and peak periods are 15, 31, and 9, respectively. Therefore, the first time period of the next day is predicted as the off-peak period.

[0094] Repeat the analysis and forecasting for all time periods on the second day to obtain operational time period forecast data;

[0095] In practice, the sum of position numbers is used as the score, which includes a strategy of giving higher weight to more recent samples; this reflects recent trends better than simply the number of occurrences.

[0096] The pricing adjustment module dynamically adjusts the pricing of the paid smart swimming ring for each time period of the next day based on the predicted operating hours data for the next day.

[0097] The pricing adjustment module is configured with pricing adjustment strategies, which include: setting prices for paid swimming rings during peak, off-peak, and low-peak hours respectively; adjusting the usage price of paid swimming rings for all hours of the next day based on operational period forecast data, and recording it as the swimming ring usage price for the next day; the prices for peak, off-peak, and low-peak hours can be set manually.

[0098] When the user returns the paid swimming ring on the second day, a corresponding bill is generated based on the user's usage time and the price of the swimming ring, and the user is charged accordingly.

[0099] In practice, during peak hours, the resource value of swimming rings is fully realized through premium pricing; during off-peak hours, sales volume and revenue are balanced by flat pricing; and during low-peak hours, demand is stimulated through low prices, thereby increasing overall revenue and revenue per unit time.

[0100] Example 2, please refer to Figure 2 As shown, this application provides a method for renting and returning a paid smart swim ring holder based on an internet platform, including the following steps:

[0101] Step S1 involves recording the number of rentals and the number of pool rings rented during each time period during the use of the paid pool ring smart rack, thus obtaining core operational data for the pool rings. Step S1 includes the following sub-steps:

[0102] Step S101: Record any paid swimming ring smart rack as the first smart rack, obtain the daily opening time of the first smart rack, and record it as the rental opening time;

[0103] Step S102: Record the location of the first smart shelf used to store paid swimming rings as a swimming ring compartment, and obtain the total number of swimming ring compartments owned by the first smart shelf, which is recorded as the total number of compartments;

[0104] Step S103: Set the first duration to t1, each duration t1 is a time period, and divide each time period evenly into k1 sub-time periods, where k1 is the set number.

[0105] Step S104: Record any sub-period of the rental opening time as the first sub-period, and record any period as the first period; collect the number of times the first smart rack rents paid swimming rings during the first sub-period, and record it as the number of rentals during the first sub-period;

[0106] Step S105: Collect the number of paid swimming rings rented out by the first smart frame at the first time interval during the first time period, record it as the real-time rental number, and calculate the average of all real-time rental numbers, record it as the rental number in the first sub-time period, where the first time interval is t2.

[0107] Step S106: Collect the number of rentals and the number of rentals for each sub-segment during the daily rental opening hours, and sort them by date and time, recording them as the core operation data of the first smart frame's swimming rings.

[0108] Step S2 involves preprocessing the core operational data of the swimming ring to obtain effective operational reference data, and then dividing the data into time periods to obtain operational time period reference data. Step S2 includes the following sub-steps:

[0109] Step S201: Extract the portion collected in the most recent k2 days from the core operation data of the swimming ring and record it as recent operation data. Combine the number of rentals and the number of rentals in the first sub-period of each day in the recent operation data into a data pair, denoted as (x, y), to obtain the data pair set, where k2 is the set number of days, x represents the number of rentals, and y represents the number of rentals.

[0110] Step S202: Calculate the Euclidean distance between any two data pairs in the data pair set to obtain the distance set; obtain the median ME of the distance set; use ME as the neighborhood radius and set e1 as the minimum number of cluster points, where e1 is the set number of points;

[0111] Step S203: Perform density clustering on the data pair set and obtain all core points, boundary points and isolated points, and record all core points as core clusters.

[0112] Step S204: Denote x / y as the coupling ratio, and set x / y=1 when y=0; calculate the coupling ratio of all data pairs in the core cluster, and calculate the mean AP and standard deviation AB of all coupling ratios; denote [AP-k3×AB, AP+k3×AB] and 1 as the normal coupling range, where k3 is the set proportional coefficient;

[0113] Step S205: Sort the number of leases of all data pairs in the core cluster in ascending order, and obtain the e2 percentile EX1 and the e3 percentile EX2. Record [EX1, EX2] as the normal fluctuation range corresponding to the number of leases; and repeatedly obtain the normal fluctuation range corresponding to the number of leases, recorded as [EY1, EY2], where e2 and e3 are the set percentiles.

[0114] Step S206: Denote any data pair corresponding to the boundary point or isolated point as the first data pair. If the y of the first data pair is zero and the x is not zero, then the x and y of the first data pair are determined to be outliers.

[0115] Step S207: If the first data pair y is not zero and the coupling ratio of the first data pair is not within the normal coupling range, then the x and y of the first data are determined to be outliers.

[0116] Step S208: If the coupling ratio of the first data pair is within the normal coupling range, but the corresponding x is not located in [EX1, EX2] and the corresponding y is not located in [EY1, EY2], and the first data pair is an isolated point, then the x and y of the first data are determined to be outliers.

[0117] Step S209: If x and y of the first data are outliers, obtain x and y of the first sub-period of the most recent k4 days that are not outliers, calculate the corresponding average values, and replace them, where k4 is the set number of days;

[0118] Step S210: Repeatedly filter all outliers in the data set and replace them to obtain the standard operating data for the first sub-period. Obtain the standard operating data for all sub-periods based on the recent operating data to obtain effective operating reference data.

[0119] Step S211: Based on effective operational reference data, for the first time period of any day, calculate the ratio of the sum of the number of rentals in the first sub-time period to the total number of rentals on that day, and record it as the rental volume ratio of the first time period; and calculate the ratio of the maximum number of rentals in the first sub-time period to the total number of positions, and record it as the time period rental rate. Repeat the acquisition of the rental volume ratio and time period rental rate for all time periods.

[0120] Step S212: Combine the rental volume ratio and the time period rental rate of each time period into a data point, denoted as (VX, VY), where VX represents the rental volume ratio and VY represents the time period rental rate;

[0121] Step S213: Obtain the 33rd percentile FV1 and the 66th percentile FV2 of the rental volume ratio of all time periods, and obtain the 33rd percentile GV1 and the 66th percentile GV2 of the time period rental rate;

[0122] Step S214: For any data point, denoted as (AX, AY), if AX ≤ FV1 and AY ≤ GV1, it is classified as a low segment point; if FV1 < AX < FV2 and GV1 < AY < GV2, it is classified as a flat segment point; if FV2 ≤ AX and GV2 ≤ AY, it is classified as a high segment point; otherwise, it is not classified.

[0123] Step S215: Fully classify all data points, calculate the mean LX of VX and the mean LY of VY of the low segment points to obtain the center point corresponding to the low segment points, denoted as the low peak center (LX, LY); repeatedly obtain the flat peak center (MX, MY) corresponding to the flat segment points and the high peak center (HX, HY) corresponding to the high segment points;

[0124] Step S216: For any data point, calculate the Euclidean distances to (LX, LY), (MX, MY), and (HX, HY) respectively, and mark the data point as a low peak point, a flat peak point, or a high peak point according to the minimum Euclidean distance;

[0125] Step S217: Mark all data points and mark the corresponding time periods in the effective operation reference data as low peak time periods, flat peak time periods, or high peak time periods. After completion, obtain the operation time period reference data.

[0126] Step S3: Perform time period prediction analysis based on the operation time period reference data to obtain the operation time period prediction data for the next day; Step S3 includes the following sub-steps:

[0127] Step S301: Obtain the date type of the next day, where the date type includes rest days and working days; obtain the data of all dates of the same type as the next day in the operation time period reference data, denoted as the effective reference data;

[0128] Step S302: For the first time period, sort the low peak time period, flat peak time period, and high peak time period corresponding to the first time period in the effective reference data in ascending order of date, denoted as the date type sequence of the first time period;

[0129] Step S303: Obtain the sum of the position indices of the low-peak periods in the date type sequence of the first period, and record it as the probability score of the low-peak period. Repeatedly obtain the probability scores of the off-peak and peak periods. Based on the highest probability score, predict the first period of the next day as the corresponding low-peak, off-peak, or peak period.

[0130] Repeat the analysis and forecasting for all time periods on the second day to obtain operational time period forecast data.

[0131] Step S4: Based on the operational time forecast data for the next day, dynamically adjust the pricing of the paid smart swimming ring racks for each time slot on the following day; Step S4 includes the following sub-steps:

[0132] Step S401: Set the prices for paid swimming rings during peak, off-peak, and low-peak hours respectively; adjust the usage prices of paid swimming rings for all hours of the next day based on the operation period forecast data, and record them as the swimming ring usage prices for the next day.

[0133] In step S402, when the user returns the paid swimming ring on the second day, a corresponding bill is generated based on the user's usage period and the price of the swimming ring, and the user is charged.

[0134] Example 3, please refer to Figure 4 As shown, Figure 4 A schematic diagram of an electronic device is provided, which may include a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores computer-readable instructions, and the processor can call these instructions. When the processor executes a computer-readable instruction, it performs steps similar to those in a paid smart swimming ring rental and return method based on an internet platform, to achieve the following functions: During the use of the paid smart swimming ring, the number of rentals and the number of swimming rings rented in each time period are recorded to obtain core operational data; the core operational data is preprocessed to obtain effective operational reference data, and then divided into time periods to obtain operational time period reference data; based on the operational time period reference data, time period prediction analysis is performed to obtain operational time period prediction data for the next day; based on the operational time period prediction data for the next day, the pricing of the paid smart swimming ring for each time period is dynamically adjusted for the next day.

[0135] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0136] Example 4: This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program performs steps such as those in the rental and return method of a paid smart swimming ring rack based on an internet platform to achieve the following functions: during the use of the paid smart swimming ring rack, the number of rentals and rentals of paid swimming rings in each time period are purchased to obtain core operating data of the swimming rings; the core operating data of the swimming rings is preprocessed to obtain effective operating reference data, and time period segmentation is performed to obtain operating time period reference data; time period prediction analysis is performed based on the operating time period reference data to obtain operating time period prediction data for the next day; and the pricing of the paid smart swimming ring rack for each time period on the next day is dynamically adjusted based on the operating time period prediction data for the next day.

[0137] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0138] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An internet platform based rental and return system for a paid swim ring smart rack, characterized in that, It includes an operations data collection module, an operations analysis module, a time period forecasting module, and a pricing adjustment module; The operation data collection module is used to collect the number of times and the number of paid swimming rings rented in each time period during the use of the paid swimming ring smart rack, and obtain the core operation data of the swimming rings. The operation analysis module includes a preprocessing unit and a time period analysis unit. The preprocessing unit is used to preprocess the core operation data of the swimming ring to obtain effective operation reference data, and the time period analysis unit is used to perform time period division processing to obtain operation time period reference data. The time period prediction module performs time period prediction analysis based on the operating time period reference data to obtain the operating time period prediction data for the next day. The pricing adjustment module dynamically adjusts the pricing of the paid smart swimming ring rack for each time period of the next day based on the predicted operating hours data for the next day.

2. The internet platform based pay per use swim ring smart locker rental and return system as claimed in claim 1 wherein, The operations data collection module is configured with operations data collection strategies, which include: Designate any paid smart pool ring rack as the first smart rack, obtain the daily opening hours of the first smart rack, and record them as the rental opening hours; The location of the first smart shelf used to store paid swimming rings is recorded as a swimming ring slot. The total number of swimming ring slots owned by the first smart shelf is recorded as the total number of slots. The first duration is set to t1, each duration of t1 is a time period, and each time period is evenly divided into k1 sub-time periods, where k1 is the number of sub-time periods set.

3. The internet platform based pay per use swim ring smart locker rental and return system as claimed in claim 2 wherein, Operational data collection strategies also include: Any sub-period within the rental opening time is designated as the first sub-period, and any period is designated as the first period; the number of times the first smart rack rents a paid swimming ring during the first sub-period is collected and recorded as the rental count of the first sub-period; In the first time period, the number of paid swimming rings rented by the first smart rack is collected at the first time interval and recorded as the real-time rental number. The average of all real-time rental numbers is calculated and recorded as the rental number in the first sub-time period, where the first time interval is t2. Every day during the rental opening hours, the number of rentals and the number of rentals for each segment are collected and sorted by date and time, which is recorded as the core operational data of the first smart rack.

4. The rental and return system for a paid swimming ring smart frame based on an internet platform as described in claim 3, characterized in that, The preprocessing unit is configured with a preprocessing strategy, which includes: Extract the portion collected in the last k2 days from the core operation data of the swimming ring and record it as the recent operation data. Combine the number of rentals and the number of rentals in the first sub-period of each day in the recent operation data into a data pair, denoted as (x, y), to obtain the data pair set, where k2 is the set number of days, x represents the number of rentals, and y represents the number of rentals. Calculate the Euclidean distance between any two data pairs in the dataset to obtain the distance set; obtain the median ME of the distance set; use ME as the neighborhood radius and set e1 as the minimum number of cluster points, where e1 is the set number of points; Perform density clustering on the data set and obtain all core points, boundary points, and isolated points. Record all core points as a core cluster.

5. The rental and return system for a paid smart swimming ring holder based on an internet platform as described in claim 4, characterized in that, Preprocessing strategies also include: Denote x / y as the coupling ratio, and when y = 0, let x / y = 1; calculate the coupling ratios of all data pairs within the core cluster, and calculate the mean AP and standard deviation AB of all coupling ratios; denote [AP - k3×AB, AP + k3×AB] and 1 as the normal coupling range, where k3 is the set proportionality coefficient; Sort the lease times of all data pairs within the core cluster in ascending order, and obtain the e2-th percentile EX1 and the e3-th percentile EX2, and denote [EX1, EX2] as the corresponding normal fluctuation range of the lease times; and repeatedly obtain the normal fluctuation range corresponding to the lease count, denoted as [EY1, EY2], where e2 and e3 are the set percentiles.

6. The rental and return system for a paid smart swimming ring holder based on an internet platform as described in claim 5, characterized in that, The preprocessing strategy also includes: Denote any one data pair corresponding to a boundary point or an isolated point as the first data pair. If y of the first data pair is zero and x is not zero, then determine x and y of the first data as outliers; If y of the first data pair is not zero and the coupling ratio of the first data pair is not within the normal coupling range, then determine x and y of the first data as outliers; If the coupling ratio of the first data pair is within the normal coupling range, but the corresponding x is not within [EX1, EX2], and the corresponding y is not within [EY1, EY2], and the first data pair is an isolated point, then determine x and y of the first data as outliers; If x and y of the first data are outliers, then obtain the x and y that are not outliers in the first sub-period of the most recent k4 days, calculate the corresponding averages respectively, and perform replacement, where k4 is the set number of days; Repeatedly screen all outliers in the data pair set and perform replacement to obtain the standard operation data for the first sub-period. Obtain the standard operation data for all sub-periods based on the recent operation data to obtain the effective operation reference data.

7. The rental and return system for a paid swimming ring smart frame based on an internet platform as described in claim 6, characterized in that, The time period analysis unit is configured with a time period analysis strategy, and the time period analysis strategy includes: Based on the effective operation reference data, for the first time period of any day, calculate the proportion of the sum of the lease times of the included first sub-periods to the total lease times of the day, denoted as the lease volume ratio of the first time period; and calculate the proportion of the maximum value of the lease count of the included first sub-periods to the total number of positions, denoted as the time period lease rate, and repeatedly obtain the lease volume ratio and the time period lease rate for all time periods; Combine the lease volume ratio and the time period lease rate of each time period into a data point, denoted as (VX, VY), where VX represents the lease volume ratio and VY represents the time period lease rate; Obtain the 33rd percentile FV1 and the 66th percentile FV2 of the lease volume ratios of all time periods, and obtain the 33rd percentile GV1 and the 66th percentile GV2 of the time period lease rates; For any data point, denoted as (AX, AY), if AX ≤ FV1 and AY ≤ GV1, then classify it as a low-segment point. If FV1 < AX < FV2 and GV1 < AY < GV2, then classify it as a flat-segment point. If FV2 ≤ AX and GV2 ≤ AY, then classify it as a high-segment point, otherwise do not classify.

8. The rental and return system for a paid smart swimming ring holder based on an internet platform as described in claim 7, characterized in that, The time period analysis strategy also includes: Divide all data points into segments, calculate the mean value LX of VX and the mean value LY of VY for low-segment points, and obtain the center point corresponding to the low-segment point, denoted as the low-peak center (LX, LY); repeatedly obtain the flat-peak center (MX, MY) corresponding to the flat-segment point and the peak center (HX, HY) corresponding to the high-segment point. For any data point, calculate the Euclidean distances to (LX, LY), (MX, MY) and (HX, HY) respectively, and mark the data point as a low peak, flat peak or high peak based on the minimum Euclidean distance; All data points are marked, and the corresponding time periods in the valid operational reference data are marked as off-peak, off-peak, or peak periods. After completion, the operational time period reference data is obtained.

9. The rental and return system for a paid swimming ring smart frame based on an internet platform as described in claim 8, characterized in that, The time-period forecasting module is configured with time-period forecasting strategies, which include: Retrieve the date type for the second day, including rest days and workdays; retrieve data for all dates of the same type as the second day from the operational period reference data, and record them as valid reference data; For the first time period, the low-peak, off-peak, and peak periods corresponding to the first time period in the valid reference data are sorted from oldest to newest date and recorded as the date type sequence of the first time period; Get the sum of the position indices of the low-peak periods in the date type sequence of the first period, and record it as the probability score of the low-peak period. Repeat the acquisition of the probability scores of the off-peak and peak periods. Based on the highest probability score, predict the first period of the next day as the corresponding low-peak, off-peak, or peak period. Repeat the analysis and forecasting for all time periods on the second day to obtain operational time period forecast data.

10. The rental and return system for a paid smart swimming ring holder based on an internet platform as described in claim 9, characterized in that, The pricing adjustment module is configured with pricing adjustment strategies, which include: Set separate prices for paid swimming rings during peak, off-peak, and low-peak hours; adjust the usage price of paid swimming rings for all time periods of the next day based on the operational period forecast data, and record it as the swimming ring usage price for the next day; When the user returns the paid swimming ring on the second day, a corresponding bill is generated based on the user's usage time and the price of the swimming ring, and the user is charged accordingly.

Images