Parking lot management efficiency evaluation method based on parking frequency

Abstract

The invention discloses a parking lot management efficiency evaluation method based on parking frequency, and the method comprises the steps: collecting vehicle access time, license plate number, parking space number, vehicle type and parking lot real-time occupancy rate data through sensors, cameras and intelligent barrier gate systems which are disposed at an entrance, an exit and parking spaces of a parking lot; performing data preprocessing on the acquired data to form a standardized data set; key indexes are extracted from the data set; based on the historical data and the external influence factors, establishing a parking demand prediction model by adopting a time sequence analysis method, and predicting the parking demand in a specified time period in the future; based on the extracted key indexes, corresponding weight factors are set, a management efficiency evaluation model is constructed, and a parking lot management efficiency score is calculated; comparing the calculated efficiency score with historical data to generate an evaluation report; and according to the predicted parking demand and the evaluation report, implementing a dynamic pricing model, and optimizing the utilization rate of the parking lot.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent transportation system and data-driven parking management, and in particular to a parking lot management efficiency evaluation method based on parking frequency. Background Art

[0002] Traditional parking management systems focus primarily on static indicators, such as parking space occupancy and revenue generation, and the technologies used include manual ticketing, basic automatic gate systems, and RFID-based access control. These traditional methods often lack the ability to capture real-time data and provide actionable insights into the dynamic usage patterns of parking facilities, and are therefore limited in supporting strategic decision-making for optimizing parking operations.

[0003] In recent years, advances in sensor technology, Internet of Things (IoT) devices and data analytics have driven the development of more sophisticated parking management systems. These modern systems integrate features such as real-time occupancy monitoring, automatic payment processing, license plate recognition and mobile application interfaces, which have improved user convenience and operational efficiency to a certain extent. In addition, the convergence of big data analytics and cloud computing has opened up new avenues for data storage and processing, enabling parking operators to collect large amounts of data related to parking behaviors and patterns.

[0004] Despite technological advances, existing methods for evaluating parking management efficiency still rely heavily on static indicators such as total occupancy, peak occupancy levels, and accumulated revenue, which can only provide a superficial understanding of the actual operational performance and resource utilization of parking facilities, and often ignore dynamic factors such as parking frequency, vehicle turnover, average parking duration, and temporal changes in occupancy, which are critical for effective resource allocation and maximizing facility utilization; for example, a parking facility may exhibit high occupancy, ostensibly indicating optimal utilization, but this may be due to extended parking durations leading to lower vehicle turnover, limiting use by other users; this situation may lead to increased congestion, decreased user satisfaction, and potential revenue loss due to the inability to accommodate more vehicles.

[0005] Additionally, current systems often lack predictive analytics capabilities to forecast future parking demand based on historical data and external influencing factors such as weather conditions, public events, and traffic flow patterns, hindering proactive management strategies that can improve congestion relief during peak hours, optimize pricing models, and increase overall user satisfaction. Summary of the invention

[0006] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract of the specification and the title of the invention of this application to avoid blurring the purpose of this section, the abstract of the specification and the title of the invention, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0007] In view of the above existing problems, the present invention is proposed.

[0008] Therefore, the technical problem solved by the present invention is: the problem of insufficient efficiency evaluation in the prior art.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions: by installing sensors, cameras and intelligent gate systems at the entrance, exit and parking spaces of the parking lot, the vehicle entry and exit time, license plate number, parking space number, vehicle type and real-time parking lot occupancy rate data are collected;

[0010] Clean, filter and integrate the collected data, remove outliers and duplicate records, and form a standardized data set;

[0011] Extract key indicators of parking frequency, vehicle turnover rate, average parking duration, and peak occupancy rate from the standardized data set;

[0012] Based on historical data and external influencing factors, a parking demand forecasting model is established using time series analysis methods to predict parking demand within a specified time period in the future;

[0013] Based on the extracted key indicators, corresponding weight factors are set, a management efficiency evaluation model is constructed, and a parking lot management efficiency score is calculated;

[0014] Compare the calculated efficiency score with historical data and generate an evaluation report including analysis results and improvement suggestions;

[0015] Dynamically adjust parking lot management strategies based on the predicted parking demand and the evaluation report, including implementing a dynamic pricing model to optimize parking lot utilization.

[0016] As a preferred solution of the parking lot management efficiency evaluation method based on parking frequency described in the present invention, the collected data also includes data of external influencing factors, and the external influencing factors include weather conditions, holiday information, surrounding activity information and traffic flow data.

[0017] As a preferred solution of the parking lot management efficiency evaluation method based on parking frequency described in the present invention, before forming the standardized data set, it is also necessary to fill in missing values ​​and perform time synchronization processing on the collected data.

[0018] As a preferred solution of the parking lot management efficiency evaluation method based on parking frequency described in the present invention, the data in the data set is summarized and calculated according to time periods by using database query and statistical functions to extract the key indicators, including:

[0019] Stop frequency:

[0020]

[0021] Vehicle turnover rate:

[0022]

[0023] Average parking time:

[0024]

[0025] Peak occupancy:

[0026]

[0027] Where N is the total number of vehicles in the specified time period, T is the length of the time period, S is the total number of parking spaces, and l i is the parking time of the i-th vehicle, U(t) is the parking lot occupancy rate at time t, are the entry and departure times of the ith vehicle respectively.

[0028] As a preferred solution of the parking lot management efficiency evaluation method based on parking frequency described in the present invention, historical parking demand data Y is selected. t As a time series, the external influencing factor X t As an exogenous variable of the model, the parking demand prediction model is constructed using the autoregressive integrated moving average model, and its mathematical expression is:

[0029] Y t =φ1Y t-1 +φ2Y t-2 +…+φ p Y t-p +θ1∈ t-1 +θ2∈ t-2 +…+θ q ∈ t-q +β1X 1t +β2X 2t +…+β k X kt +∈ t

[0030] Among them, Y t is the parking demand at time t, φ p is the autoregressive coefficient, θ q is the moving average coefficient, ∈t is the random error term, and β k is the exogenous variable X k 's coefficient, and X kt is the value of the k-th external influencing factor at time t.

[0031] As a preferred solution of the parking lot management efficiency evaluation method based on parking frequency described in the present invention, the calculation formula of the management efficiency score E is:

[0032]

[0033] where w1, w2, w3, and w4 are weight factors, satisfying w1 + w2 + w3 + w4 = 1, and F max , R max , L max , U max are the benchmark values of each index.

[0034] As a preferred solution of the parking lot management efficiency evaluation method based on parking frequency described in the present invention, the comparison method of the efficiency score includes using the significance test method in statistics to compare the difference between the current efficiency score and the historical average value or industry standard.

[0035] As a preferred solution of the parking lot management efficiency evaluation method based on parking frequency described in the present invention, the mathematical expression formula of the dynamic pricing model:

[0036]

[0037] where P(t) is the adjusted parking price at time t, P0 is the base price, that is, the initially set price,

[0038] β is the price sensitivity coefficient, controlling the amplitude of price adjustment, is the predicted parking demand, and D avg is the average parking demand;

[0039] If then P(t) > P0, which is a high-demand period, and the price needs to be increased to regulate the demand;

[0040] If then P(t) < P0, which is a low-demand period, and the price needs to be reduced to attract vehicles.

[0041] Advantages of the present invention:

[0042] 1. A high-precision and high-timeliness basic data set is established, ensuring the accuracy and integrity of subsequent analysis, achieving the beneficial effects of real-time monitoring of the operation status of the parking lot and laying a foundation for data-driven decision-making;

[0043] 2. By cleaning, filtering and integrating the original data, outliers and duplicate records are eliminated, data consistency and standardization are achieved, data noise and redundant information are eliminated, high standardization of data quality is ensured, the credibility and availability of data are improved, and the beneficial effect of providing an accurate and reliable data basis for subsequent analysis is achieved;

[0044] 3. It realizes the quantitative description of the parking lot operation status, provides the necessary parameter input for the evaluation model, and achieves the beneficial effect of providing multi-dimensional and refined indicator support for management efficiency evaluation by extracting the dynamic characteristics of parking lot use;

[0045] 4. Identify peak and trough periods in advance, realize the prediction of future parking demand, provide managers with forward-looking decision-making basis, and achieve the beneficial effect of optimizing resource allocation and improving response capabilities;

[0046] 5. It realizes the comprehensive quantitative evaluation of multi-dimensional indicators, avoids the bias that may be caused by a single indicator, comprehensively evaluates the operational efficiency of the parking lot, and achieves the beneficial effect of providing accurate and quantitative basis for improving management strategies;

[0047] 6. Achieved in-depth diagnosis of parking lot operation status, guided managers to take targeted improvement measures, and achieved the beneficial effect of continuously improving management efficiency and optimizing operation strategies;

[0048] 7. The flexibility and adaptability of management strategies have been achieved, which can respond to changes in market demand in a timely manner, maximize the utilization rate and revenue of parking lots, and achieve the beneficial effects of improving customer satisfaction, optimizing resource utilization and enhancing economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0050] Figure 1 It is a flow chart of a parking lot management efficiency evaluation method based on parking frequency shown in the present invention;

[0051] Figure 2 This is a schematic diagram of the comparison of parking lot occupancy rates according to Article 7 of the present invention;

[0052] Figure 3 This is a schematic diagram showing the comparison of peak occupancy rates from the first day to the seventh day shown in the present invention. DETAILED DESCRIPTION

[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0054] Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without making any creative work should fall within the scope of protection of the present invention.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0056] According to an embodiment of the present invention, Figure 1 The flowchart shown in the figure is a parking lot management efficiency evaluation method based on parking frequency, which specifically includes the following steps:

[0057] S1. Collect vehicle entry and exit time, license plate number, parking space number, vehicle type and real-time parking lot occupancy data through sensors, cameras and intelligent gate systems installed at the parking lot entrance, exit and parking spaces. Among them, the following points need to be explained in this step:

[0058] In and out time t in and t out : Record vehicle entry and departure times through sensors and cameras at the entrance and exit;

[0059] License plate number V plate : Obtained through the license plate recognition system to uniquely identify the vehicle;

[0060] Parking Space No. V slo t: The parking space number where the vehicle is parked is recorded through the parking space sensor;

[0061] Vehicle Type V type : Such as small cars, large cars, electric cars, for classification analysis;

[0062] Total number of parking spaces S: the total number of parking spaces in the parking lot;

[0063] Real-time occupancy rate U(t): real-time monitoring by parking sensors, the calculation formula is:

[0064]

[0065] Among them, S occupied (t) is the number of occupied parking spaces at time t;

[0066] As an example, external influencing factor data:

[0067] Weather conditions W(t): obtained through the meteorological data interface and marked as sunny, rainy, or snowy;

[0068] Holiday information H(t): Mark working days or holidays according to calendar information;

[0069] Surrounding event information A(t): obtain the time and scale of nearby large-scale events;

[0070] Traffic flow data T(t): obtain real-time road traffic information through the traffic department’s API.

[0071] In an optional implementation, data is collected by installing high-precision sensors, cameras, and intelligent gate systems, and wireless or wired networks are used to transmit the data in real time to a central database to ensure the accuracy and real-time nature of the data.

[0072] S2. Clean, filter and integrate the collected data, remove outliers and duplicate records, and form a standardized data set. Among them, the following points need to be explained in this step:

[0073] Missing value processing: If the proportion of missing values ​​to the total data volume is small (such as less than 5%), directly delete the records with missing values; for important features, use the mean, median or mode to fill the missing values. For example, the missing value of parking time can be filled with the average parking time;

[0074] Outlier processing: Use the box plot method or the 3σ principle to identify outliers. For example, if the parking time exceeds the normal range (such as more than 24 hours), it is considered an outlier. Verify the outliers to confirm whether they are data input errors. If necessary, delete or correct the outliers.

[0075] Data standardization: normalize data of different dimensions to meet the input requirements of subsequent models;

[0076] Time synchronization: align data from different sources according to a unified timestamp to ensure data consistency;

[0077] Time feature extraction: extract date, weekday, and hour features from timestamps, and mark weekdays, weekends, and holidays;

[0078] External factor encoding: Use One-Hot Encoding to convert weather conditions into numerical features. For example, sunny days, rainy days, and snowy days are encoded as (1,0,0), (0,1,0), and (0,0,1) respectively.

[0079] Activity information: mark whether there is a large-scale activity, binary processing, 0 means no activity, 1 means there is an activity;

[0080] Data Integration:

[0081] Time alignment: Summarize vehicle entry and exit data, parking space occupancy data, and external factor data at a unified time interval (such as every hour) to ensure that records from different data sources are aligned at the same time point; the final data set is expressed in a structured data table, where:

[0082] Row (sample): data record for each time interval (such as every hour);

[0083] Column (feature): includes all processed features, such as time features, vehicle entry and exit times, average parking time, occupancy rate, weather code, and activity information;

[0084] Table 1. Example dataset

[0085]

[0086] S3. Extract key indicators of parking frequency, vehicle turnover rate, average parking duration, and peak occupancy rate from the standardized data set. It should be noted that this step uses database query and statistical functions to summarize and calculate the data in the data set according to time periods to extract key indicators, including:

[0087] Stop frequency:

[0088]

[0089] Vehicle turnover rate:

[0090]

[0091] Average parking time:

[0092]

[0093] Peak occupancy:

[0094]

[0095] Where N is the total number of vehicles in the specified time period, T is the length of the time period, S is the total number of parking spaces, and l i is the parking time of the i-th vehicle, U(t) is the parking lot occupancy rate at time t, are the entry and departure times of the ith vehicle, and R represents the number of times a unit parking space is used within a specified time period.

[0096] S4. Based on historical data and external influencing factors, a parking demand forecasting model is established using a time series analysis method to forecast parking demand in a future specified time period. The following points need to be noted in this step:

[0097] Select historical parking demand data Y t As a time series;

[0098] The external influencing factor X t (such as W(t), H(t), A(t), T(t)) as exogenous variables of the model;

[0099] The parking demand prediction model is constructed using the autoregressive integrated moving average model, and its mathematical expression is:

[0100] Y t =φ1Y t-1 +φ2Y t-2 +…+φ p Y t-p +θ1∈ t-1 +θ2∈ t-2 +…+θ q ∈t- q +β1

[0101] X 1t +β2X 2t +…+β x X kt +∈ t

[0102] Among them, Y t is the parking demand at time t, φ p is the autoregressive coefficient, θ q is the moving average coefficient, ∈ t is the random error term, β k is the exogenous variable X k The coefficient of X kt is the value of the kth external influencing factor at time t;

[0103] Using historical data, estimate the model parameters φ by least squares or maximum likelihood estimation i ,θ j ,β k ;

[0104] As an example, evaluate the model's fit and select the best model using the AIC (Akaike Information Criterion) metric;

[0105] Use the trained model to predict parking demand at future times

[0106] Calculate the confidence interval for the predicted value, ranging from Where Z is the quantile of the standard normal distribution and σ is the standard deviation of the prediction error;

[0107] Prediction rule judgment conditions:

[0108] High Demand Warning: When the predicted parking demand When more than 90% of the total number of parking spaces are occupied, an early warning is triggered;

[0109] Low Demand Opportunities: When it is below 80% of the average demand, consider taking promotional measures.

[0110] In an optional implementation, the prediction model improves prediction accuracy through learning and training, and the training process steps are as follows:

[0111] Training set: Use the first 70% of historical data for model training;

[0112] Validation set: uses the middle 15% of historical data for model tuning;

[0113] Test set: uses the last 15% of historical data for model testing;

[0114] It uses ARIMAX (autoregressive integrated moving average model with exogenous variables), which is suitable for time series forecasting and can include external factors;

[0115] Determine the number of differences d: Use the ADF (Augmented Dickey-Fuller) test. If the time series is non-stationary, perform differences until the series becomes stationary.

[0116] Determine the autoregressive order p and the moving average order q: Use the autocorrelation function (ACF) and partial autocorrelation function (PACF) graphs to observe the tailing and truncation conditions and determine the values ​​of p and q;

[0117] Evaluate the model performance on the validation set:

[0118] Evaluation indicators:

[0119] Mean Squared Error (MSE):

[0120]

[0121] Mean Absolute Percent Error (MAPE):

[0122]

[0123] Parameter tuning:

[0124] According to the evaluation results, adjust p, d, q and exogenous variables and repeat the training until the model performance is optimal;

[0125] Model testing: Verify model performance on the test set to ensure the generalization ability of the model;

[0126] Model application: Use the trained model to predict parking demand in a specified time period in the future;

[0127] As an example, the predicted future specified time period is a short-term prediction, that is, the hourly parking demand in the next 24 hours, so as to adjust the parking lot management strategy in real time.

[0128] S5. Based on the extracted key indicators, set the corresponding weight factors, build a management efficiency evaluation model, and calculate the parking lot management efficiency score. The following points need to be explained in this step:

[0129] The calculation formula of management efficiency score E is:

[0130]

[0131] Among them, w1, w2, w3, and w4 are weight factors, satisfying w1+w2+w3+w4=1, F max , R max , L max , U max is the benchmark value of each indicator.

[0132] As an example, the weight factor is set according to the operation target to satisfy w1+w2+w3+w4=1; for example, if the vehicle turnover rate is emphasized, w2 can be set larger.

[0133] As an example, the benchmark value is determined as an industry standard or a historical best value, such as F max The highest parking frequency in history, R max The highest vehicle turnover rate in history, L max is the longest acceptable average parking time, U max is the theoretical maximum occupancy rate (generally 100%), E is the management efficiency score, ranging from 0 to 1, and the higher the F, R and U peak It is beneficial to improve E. A lower average parking time L is beneficial to improve E.

[0134] S6. Compare the calculated efficiency score with historical data to generate an evaluation report containing analysis results and improvement suggestions. It should be noted that the efficiency score comparison method includes using a statistical significance test method to compare the difference between the current efficiency score and the historical average or industry standard, where:

[0135] Calculate the current efficiency score E current : Use the efficiency score formula in step S5 to calculate the efficiency score of the current period;

[0136] Get the benchmark efficiency score E baseline : Calculate the average efficiency score under the same conditions in the past and obtain the standard efficiency score in the industry;

[0137] Setting up hypothesis testing:

[0138] Null hypothesis H0: E current =E baseline , that is, there is no significant difference;

[0139] Alternative hypothesis H1: E current ≠E baseline , that is, there is a significant difference;

[0140] t-test: Compare the current score E current With the benchmark score E baseline ;

[0141] Calculate the t value:

[0142]

[0143] Among them, s current 、s baseline is the standard deviation between the current efficiency score and the benchmark efficiency score, n current 、n baseline is the sample size;

[0144] Determine the critical value and degrees of freedom:

[0145] Degrees of freedom df:

[0146]

[0147] Critical value: According to the significance level α (such as 0.05) and the degrees of freedom df, find the critical value t from the t distribution table a / 2,df ;

[0148] Making Statistical Decisions: Comparing t-values ​​and critical values:

[0149] If |t|>t α / 2,df , then reject the original hypothesis and think that the difference is significant;

[0150] If |t|≤t α / 2,df , we cannot reject the null hypothesis and believe that the difference is not significant.

[0151] As an example, the difference is significant: if the efficiency score drops, improvement measures need to be taken; if it increases, it should be maintained or optimized.

[0152] As an example, the difference is not significant: indicating that the current efficiency level is consistent with the baseline and the existing management strategy can be maintained.

[0153] In an alternative embodiment, the content of the evaluation report includes:

[0154] An abstract for summarizing the evaluation results, giving the overall score and main conclusions;

[0155] Analysis of key indicators, such as charts showing the current values, historical trends, and benchmark values of each key indicator, or textual explanations for interpreting the changes in the indicators;

[0156] Evaluation of the efficiency score, detailing the calculation process and results of the score; comparing historical data and analyzing the reasons for changes;

[0157] Suggestions for improvement, providing specific improvement measures based on the deficiencies of the indicators, such as adjusting pricing or optimizing parking space allocation.

[0158] S7. Dynamically adjust the parking lot management strategy according to the predicted parking demand and the evaluation report, including implementing a dynamic pricing model to optimize the utilization rate of the parking lot.

[0159]

[0160] Where P(t) is the adjusted parking price at time t, P0 is the base price, i.e., the initially set price, β is the price sensitivity coefficient controlling the amplitude of price adjustment, is the predicted parking demand, D avg is the average parking demand;

[0161] If then P(t) > P0, it is a high-demand period, and the price needs to be increased to regulate demand;

[0162] If then P(t) < P0, it is a low-demand period, and the price needs to be decreased to attract vehicles.

[0163] The following describes the implementation process and / or effects of certain examples of the present invention in more detail in conjunction with some preferred or alternative examples of the present invention.

[0164] Type of parking lot: underground parking lot in a commercial center;

[0165] Total number of parking spaces S = 500;

[0166] Operating hours: 8:00 - 22:00 every day;

[0167] Base parking fee P0 = 10 yuan / hour;

[0168] Data collection period: data for the most recent month;

[0169] The implementation process steps according to the above steps S1 to S7 include:

[0170] (1) Data collection

[0171] Vehicle data: Collect the entry time t of each vehicle in the last month in , departure time t out 、License plate number V plate 、Parking space number V slot ;

[0172] Parking space occupancy data: The parking space sensor records the occupancy status U(t) of each parking space;

[0173] External factor data: obtain daily weather conditions W(t), holiday information H(t), and surrounding activity information A(t);

[0174] (2) Data preprocessing

[0175] Missing values: If some license plate numbers are found to be missing, they are marked as "unknown";

[0176] Outliers: Parking records exceeding 24 hours will be deleted after verification;

[0177] Data standardization and normalization: standardize numerical features (such as parking time, vehicle entry and exit times);

[0178] Time features: extract date, day of the week, hour;

[0179] Weather encoding: Use one-hot encoding to encode sunny days, rainy days, and snowy days separately;

[0180] Activity information: whether there is activity or not, binary processing;

[0181] Data integration: Aggregate every hour to form the final data set;

[0182] (3) Extraction of key indicators

[0183] Take the data of a certain day as an example (2023-10-15):

[0184] Stop frequency F: The total number of vehicles on the day N = 1200, the operating time T = 14 hours, then:

[0185]

[0186] Vehicle turnover rate R:

[0187]

[0188] Average parking time L: total parking time ∑l i =1800 hours, then:

[0189]

[0190] Peak occupancy U peak : Peak occupancy U peak =95%;

[0191] (4) Prediction model construction

[0192] Dataset division:

[0193] Training set: data from the first 20 days;

[0194] Validation set: data from the middle 5 days;

[0195] Test set: data from the last 5 days;

[0196] Model training: difference times d = 1, model order p = 2, q = 1,

[0197] Exogenous variables: weather, holidays, and event information;

[0198] Model evaluation: MAPE on the validation set = 8%, the model performance is good and enters the testing phase;

[0199] Model testing: MAPE on the test set = 9%. The model has good generalization ability and predicts the parking demand for the next 24 hours, obtaining the predicted demand per hour.

[0200] (5) Establishment of evaluation model

[0201] Set weight factors: w1 = 0.3, w2 = 0.3, w3 = 0.2, w4 = 0.2;

[0202] Base value: F max =90 times / hour, R max = 2.5 times / parking space, L max =2,U max =1;

[0203] Calculate the current efficiency score E current :

[0204]

[0205] (6) Efficiency score comparison and evaluation report generation

[0206] Get the benchmark efficiency score: historical average efficiency score E baseline =0.78, standard deviation s baseline =0.05, sample size n baseline =30;

[0207] Calculate the t value: Since the current efficiency score sample size is 1, a single-sample t test is used, with a standard error of:

[0208]

[0209] t value:

[0210]

[0211] Significance test: degrees of freedom df = n baseline -1=29;

[0212] The significance level is α = 0.05, and the critical value is t 0.05,29 ≈2.045;

[0213] Judgment: Because t=3.69>2.045, the original hypothesis is rejected, and it is believed that the current efficiency score is significantly higher than the benchmark efficiency score;

[0214] Contents of the assessment report:

[0215] Abstract: The current parking lot management efficiency score is 0.8136, which is significantly higher than the historical average of 0.78, and the management efficiency has been significantly improved;

[0216] Indicator analysis: Parking frequency and vehicle turnover rate are close to the maximum value, parking space utilization is high, the average parking time is 1.5 hours, which is lower than the benchmark value, which is conducive to improving turnover rate. The peak occupancy rate reaches 95%, and attention should be paid to the regulation during peak hours;

[0217] Improvement suggestions: During peak hours (such as 18:00-20:00), appropriately increase parking fees to ease parking pressure, increase staff guidance, and improve vehicle entry and exit efficiency.

[0218] (7) Management strategy optimization

[0219] Implement dynamic pricing model: price sensitivity coefficient β = 0.5, average parking demand D avg =85.71 times / hour;

[0220] Adjusting parking fees: predicting demand during peak hours

[0221]

[0222] Forecast demand during off-peak hours

[0223]

[0224] Strategy implementation: Parking fees during peak hours (18:00-20:00) are adjusted to RMB 11.05 per hour, and parking fees during off-peak hours (10:00-16:00) are adjusted to RMB 9.12 per hour;

[0225] Personnel deployment: Add 3 guides during peak hours to optimize the vehicle entry and exit process.

[0226] To further verify the effectiveness of a parking lot management efficiency evaluation method based on parking frequency provided by an embodiment of the present invention, this embodiment selected an underground parking lot of a large shopping mall as an experimental scenario. The parking lot has a total of 200 parking spaces and operates from 8 am to 10 pm every day. The experimental period is seven consecutive days, of which the traditional static indicator management method is used for the first six days, and the method of the present invention is used for comparative analysis on the seventh day.

[0227] High-precision sensors, high-definition cameras and intelligent barrier systems are installed at the entrance, exit and each parking space of the parking lot. The sensors and cameras are used to collect real-time information on vehicle entry and exit time, license plate number, parking space number and vehicle type. The intelligent barrier system cooperates with vehicle detection equipment to ensure the accuracy and completeness of data collection. In addition, parking space sensors are used to monitor the occupancy rate of the parking lot in real time. In order to take external influencing factors into consideration, additional data on weather conditions, surrounding activities and traffic flow are collected.

[0228] During the data collection process, the above information was collected 24 hours a day for seven consecutive days. The collected raw data included vehicle entry and exit timestamps, license plate numbers, vehicle types, parking space numbers, and parking lot occupancy rates at each time point. To ensure the reliability of the data, the collected data was strictly preprocessed and cleaned to remove outliers, such as records with logical errors in entry and exit times and duplicate license plate numbers. The data was filtered to exclude records that were incomplete or missing key fields. The cleaned data was integrated to form a standardized data set in preparation for subsequent analysis.

[0229] Key indicators were extracted from the standardized dataset: parking frequency (total number of parking times per day), vehicle turnover rate (ratio of total parking times to total number of parking spaces), average parking time (average parking time of all vehicles) and peak occupancy rate (highest parking lot occupancy rate per day). A parking demand forecasting model was established using a time series analysis method (ARIMA model) based on historical data from the previous six days and external influencing factors. The model predicts parking demand in each time period on the seventh day by learning historical trends and cyclical changes.

[0230] When constructing the management efficiency evaluation model, corresponding weight factors were set for the extracted key indicators, with the parking frequency weight being 0.4, the vehicle turnover rate weight being 0.3, the average parking time weight being 0.2, and the peak occupancy rate weight being 0.1. Using these weights, the parking lot management efficiency score for each day was calculated, and the efficiency score calculated on the seventh day using the method of the present invention was compared with the efficiency score of the previous six days using the traditional method. An evaluation report containing detailed analysis results and improvement suggestions was generated, and based on the predicted parking demand and the suggestions in the evaluation report, the parking lot management strategy was dynamically adjusted on the seventh day, including the implementation of a dynamic pricing model to appropriately increase parking fees during peak hours and reduce fees during off-peak hours, so as to optimize the utilization rate of the parking lot.

[0231] Reference Figure 2 It can be seen intuitively that during the peak period from 12:00 to 18:00, the occupancy rate of the traditional method is stable at 80% to 81%, while the occupancy rate of the method of the present invention is increased to 85% to 90%, which shows that the method of the present invention makes more effective use of parking resources during peak hours and attracts more vehicles to enter the parking lot; the occupancy rate curve of the method of the present invention is relatively smooth and has small fluctuations, indicating that the utilization rate of the parking lot is more stable. In contrast, the occupancy rate of the traditional method fluctuates greatly before and after the peak period, and the resource utilization is not balanced enough.

[0232] During non-peak hours (8:00 to 12:00, 18:00 to 22:00), the occupancy rate of the method of the present invention is always higher than that of the traditional method, indicating that through dynamic pricing and management strategy adjustment, more vehicles are successfully attracted to enter the parking lot during the off-peak period, thereby improving the overall utilization rate; the increase in occupancy rate and the reduction in fluctuation reflect that the parking lot is more fully utilized in different time periods, avoiding waste of resources.

[0233] Reference Figure 3 It can be seen intuitively that from the 1st day to the 6th day: the peak occupancy rate of the traditional method fluctuates between 80% and 85%, and the peak occupancy rate of the method of the present invention is between 82% and 87%, which is always higher than the traditional method by about 2% to 2%; this shows that even in daily operations, the method of the present invention can improve the peak utilization rate of the parking lot.

[0234] On the 7th day, the peak occupancy rate of the traditional method dropped to 81%, which may be affected by certain external factors. The peak occupancy rate of the method of the present invention was significantly increased to 90%, which was 9% higher than that of the traditional method. This shows that the method of the present invention has a significant effect in improving the peak occupancy rate of the parking lot and makes full use of parking resources.

[0235] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A parking lot management efficiency evaluation method based on parking frequency, characterized in that: include: The sensors, cameras and intelligent gate systems installed at the parking lot entrances, exits and parking spaces collect data on vehicle entry and exit times, license plate numbers, parking space numbers, vehicle types and real-time parking lot occupancy rates; Clean, filter and integrate the collected data, remove outliers and duplicate records, and form a standardized data set; Extract key indicators of parking frequency, vehicle turnover rate, average parking duration, and peak occupancy rate from the standardized data set; Based on historical data and external influencing factors, a parking demand forecasting model is established using time series analysis methods to predict parking demand within a specified time period in the future; Based on the extracted key indicators, corresponding weight factors are set, a management efficiency evaluation model is constructed, and a parking lot management efficiency score is calculated; Compare the calculated efficiency score with historical data and generate an evaluation report including analysis results and improvement suggestions; Dynamically adjust parking lot management strategies based on the predicted parking demand and the evaluation report, including implementing a dynamic pricing model to optimize parking lot utilization.

2. The parking lot management efficiency evaluation method based on parking frequency according to claim 1 is characterized in that: The collected data also includes data of external influencing factors, and the external influencing factors include weather conditions, holiday information, surrounding activity information and traffic flow data.

3. The parking lot management efficiency evaluation method based on parking frequency according to claim 1 is characterized in that: Before forming the standardized data set, the collected data also needs to be filled with missing values ​​and the data needs to be synchronized in time.

4. The parking lot management efficiency evaluation method based on parking frequency according to claim 1 is characterized in that: Using database query and statistical functions, the data in the data set is aggregated and calculated according to time periods to extract the key indicators, including: Stop frequency: Vehicle turnover rate: Average parking time: Peak occupancy: Where N is the total number of vehicles in the specified time period, T is the length of the time period, S is the total number of parking spaces, and l i is the parking time of the i-th vehicle, U(t) is the parking lot occupancy rate at time t, are the entry and departure times of the ith vehicle respectively.

5. The parking lot management efficiency evaluation method based on parking frequency according to claim 1 is characterized in that: Select historical parking demand data Y t As a time series, the external influencing factor X t As an exogenous variable of the model, the parking demand prediction model is constructed using the autoregressive integrated moving average model, and its mathematical expression is: Y t =φ1Y t-1 +φ2Y t-2 +…+φ p Y t-p +θ1∈ t-1 +θ2∈ t-2 +…+θ q ∈ t-q +β1 X 1t +β2X 2t +…+b k X kt +∈ t Among them, Y t is the parking demand at time t, φ p is the autoregressive coefficient, θ q is the moving average coefficient, ∈ t is the random error term, β k is the exogenous variable X k The coefficient of X kt is the value of the kth external influencing factor at time t.

6. The parking lot management efficiency evaluation method based on parking frequency according to claim 1 is characterized in that: The calculation formula of management efficiency score E is: Among them, w1, w2, w3, and w4 are weight factors, satisfying w1+w2+w3+w4=1, F max , R max , L max , U max is the benchmark value of each indicator.

7. The parking lot management efficiency evaluation method based on parking frequency according to claim 1 is characterized in that: The efficiency score comparison method includes using a statistical significance test method to compare the difference between the current efficiency score and the historical average or industry standard.

8. The parking lot management efficiency evaluation method based on parking frequency according to claim 1 is characterized in that: Mathematical expression formula of the dynamic pricing model: Among them, P(t) is the adjusted parking price at time t, P0 is the basic price, that is, the initial price, and β is the price sensitivity coefficient, which controls the amplitude of price adjustment. is the predicted parking demand, D avg is the average parking demand; like Then P(t)>P0, which is a high demand period, and the price needs to be increased to regulate demand; If then P(t) < P0, which is a low-demand period, and the price needs to be reduced to attract vehicles.

Images