Digital warehousing method and system based on task behaviors

Abstract

The invention provides a digital warehousing method and system based on task behaviors, and the method comprises the steps: analyzing historical data through a machine learning algorithm, predicting the arrival and departure modes of a vehicle in future time, and distributing a parking space for the vehicle in advance according to the prediction information; when the vehicle enters the garage, scanning a vehicle body two-dimensional code through a bar code scanning device to obtain a vehicle frame number, matching whether the vehicle frame number information is consistent with the predicted vehicle arrival mode information or not, if yes, directly distributing a parking space, and if not, matching vehicle information in a storage database based on the vehicle frame number to obtain vehicle related information; and allocating a parking space to the vehicle based on the vehicle-related information. Through the method and the corresponding system, the intelligent level of parking space warehouse management can be remarkably improved, and resource use is optimized.

Description

Technical Field

[0001] The present invention proposes a digital warehousing method and system based on task behavior, belonging to the technical field of intelligent warehousing. Background Art

[0002] With the development of social economy and the continuous growth of logistics demand, the vehicle warehousing industry is facing unprecedented challenges. Traditional vehicle warehousing management relies on a large amount of manpower, and has problems such as low efficiency, high cost, and high error rate. At the same time, modern vehicle warehousing management requires that warehouses not only have storage functions, but also form an organic whole with transportation, logistics information, logistics operations, etc. to optimize the entire supply chain. Therefore, the vehicle warehousing industry needs a more efficient and intelligent management method. Summary of the invention

[0003] The present invention provides a digital warehousing method and system based on task behavior to solve the above-mentioned problems: The present invention proposes a digital warehousing method based on task behavior, the method comprising: Analyze historical data through machine learning algorithms to predict the arrival and departure patterns of vehicles in the future, and allocate parking spaces to vehicles in advance based on the predicted information; When a vehicle enters the warehouse, the barcode scanning device is used to scan the QR code on the vehicle body to obtain the frame number, and the frame number information is matched with the predicted vehicle arrival mode information to see if they are consistent. If they are consistent, a parking space is directly allocated. If they are inconsistent, the vehicle information in the storage database is matched based on the frame number to obtain vehicle-related information, and a parking space is allocated to the vehicle based on the vehicle-related information.

[0004] Furthermore, a digital warehousing method based on task behavior analyzes historical data through machine learning algorithms, predicts the arrival and departure patterns of vehicles in the future, and allocates parking spaces to vehicles in advance based on the predicted information, including: Collecting historical vehicle travel data from the monitoring channel of the warehouse management system, wherein the historical vehicle travel data includes the arrival timestamp of the vehicle, the departure timestamp of the vehicle, and the parking time of the vehicle; Removing duplicate data, erroneous data and outliers from historical vehicle travel data, converting arrival timestamps and departure timestamps of vehicles into a unified date and time format, and extracting useful features from the converted data, wherein the useful features include the distribution of arrival and departure times, vehicle types and parking space usage; Divide the historical driving data of vehicles from which historical features have been extracted into a training set and a test set; Using training set data to train a time series prediction model, and adjusting parameters of the time series prediction model to optimize prediction performance; Verify the accuracy and generalization ability of the model through test set data; The trained machine learning model is used to predict the vehicle arrival and departure patterns in the future, generate prediction results, and calculate the parking space demand in each future time period based on the predicted results of vehicle arrival and departure.

[0005] Furthermore, a digital warehousing method based on task behavior, when a vehicle enters the warehouse, a barcode scanning device is used to scan a QR code on the vehicle body to obtain a frame number, and vehicle related information is obtained based on the frame number matching data entry in the storage database, and a parking space is allocated to the vehicle based on the vehicle related information, including: Scan the QR code on the vehicle body with a barcode scanner to obtain the vehicle frame number; Automatically input the frame number into a storage database, and the storage database searches for the frame number information to obtain vehicle related information matching the frame number information, the vehicle related information including: frame number, vehicle model, source order number and customer; The vehicle model in the vehicle-related information is obtained, the vehicle size is obtained based on the vehicle model, and a parking space is allocated to the vehicle in an idle parking space whose size is greater than or equal to the vehicle size by using a parking space allocation model.

[0006] Furthermore, a digital warehousing method based on task behavior, specifically, the parking space allocation model is: in, represents the priority of parking space i, represents the size fit of parking space i, Indicates the driving distance from the parking space to the garage entrance. represents the distance constant, Indicates the estimated parking time. represents the time constant, represents the functional matching factor, represents the seasonal impact factor.

[0007] The present invention proposes a task behavior-based digital warehousing system, the system comprising: The prediction allocation module is used to analyze historical data through machine learning algorithms to predict the arrival and departure patterns of vehicles in the future and allocate parking spaces to vehicles in advance based on the predicted information; The warehousing allocation module is used to obtain the frame number by scanning the QR code on the vehicle body through a barcode scanning device when the vehicle enters the warehouse, and to match the frame number information with the predicted vehicle arrival mode information to see if they are consistent. If they are consistent, a parking space is directly allocated. If they are inconsistent, the vehicle information in the storage database is matched based on the frame number to obtain vehicle-related information, and a parking space is allocated to the vehicle based on the vehicle-related information.

[0008] Furthermore, in a digital warehousing system based on task behavior, the prediction allocation module includes: A module for acquiring vehicle historical driving data is used to collect vehicle historical driving data from the monitoring channel of the warehouse management system, wherein the vehicle historical driving data includes the vehicle arrival timestamp, the vehicle departure timestamp and the vehicle parking time; A preprocessing module, used to remove duplicate data, erroneous data and outliers in the historical driving data of the vehicle, convert the arrival timestamp and departure timestamp of the vehicle into a unified date and time format, and extract useful features from the converted data, wherein the useful features include the distribution of arrival and departure times, vehicle type and parking space usage; A data set partitioning module is used to partition the historical driving data of vehicles from which historical features have been extracted into a training set and a test set; A training module is used to train a time series prediction model using training set data and adjust parameters of the time series prediction model to optimize prediction performance; The testing module is used to verify the accuracy and generalization ability of the model through the test set data; The prediction-based parking allocation module is used to use the trained machine learning model to predict the vehicle arrival and departure patterns in the future, generate prediction results, and calculate the parking space demand in each future time period based on the prediction results of vehicle arrival and departure.

[0009] Furthermore, in a digital warehousing system based on task behavior, the warehousing allocation module includes: The module for obtaining the vehicle frame number is used to obtain the vehicle frame number by scanning the vehicle body QR code through a barcode scanning device; A module for searching vehicle related information, used for automatically entering the frame number into a storage database, wherein the storage database searches for the frame number information to obtain vehicle related information matching the frame number information, wherein the vehicle related information includes: frame number, vehicle model, source order number and customer; The parking space allocation module is used to obtain the vehicle model in the vehicle-related information, obtain the vehicle size based on the vehicle model, and allocate a parking space to the vehicle in an idle parking space whose size is greater than or equal to the vehicle size by using the parking space allocation model.

[0010] Furthermore, in a digital warehousing system based on task behavior, the parking space allocation module includes: The storage allocation parking space model module, specifically, the parking space allocation model is: in, represents the priority of parking space i, represents the size fit of parking space i, Indicates the driving distance from the parking space to the garage entrance. represents the distance constant, Indicates the estimated parking time. represents the time constant, represents the functional matching factor, represents the seasonal impact factor.

[0011] The beneficial effects of the present invention are as follows: a large amount of historical parking data is collected, including vehicle arrival time, parking duration, parking space usage, etc.; machine learning algorithms are used to analyze these data to understand and predict future vehicle arrival and departure patterns; based on the trained model, the arrival and departure of vehicles in a certain time period in the future are predicted; the prediction results are used as the basis for pre-allocation of parking spaces. When a vehicle enters the warehouse, the QR code corresponding to the frame number is scanned by a barcode scanning device to quickly obtain the vehicle's identity information; a database search is performed to check whether the scanned frame number is consistent with the predicted arriving vehicle; if the scanned frame number is consistent with the prediction, a reserved parking space is directly allocated to it to reduce waiting time; if not, the vehicle detailed information (such as size, electric or fuel, priority, etc.) is searched in the storage database by the frame number, and parking spaces are allocated according to the set rules and parameters (which may include vehicle type, parking time estimate, parking space characteristics, etc.). Improve parking space utilization. Through prediction and automated allocation, maximize the efficiency of each parking space, reduce the idle rate and uneven occupancy of parking spaces; shorten waiting time and provide predictive parking space allocation solutions, which can significantly reduce the waiting and management time of vehicles entering the warehouse and improve user experience; adapt to changes in vehicle arrival patterns over a long period of time through machine learning models, and the system can dynamically adjust allocation strategies to adapt to changes in demand in different periods; provide more accurate resource management methods through analysis and decision-making using actual data, and continuously accumulate the ability to improve models; improve management efficiency. Automated solutions reduce manual intervention, improve parking space management efficiency, and respond to actual vehicle needs more quickly; solutions using machine learning and automated data collection systems can significantly improve the level of intelligent parking management, optimize resource use, and improve user satisfaction. Through continuous data feedback and model improvement, prediction accuracy and allocation efficiency can also be continuously improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a schematic diagram of a task behavior-based digital warehousing method described in the present invention. DETAILED DESCRIPTION

[0013] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0014] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. The embodiments described are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0015] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0016] One embodiment of the present invention provides a digital warehousing method based on task behavior, the method comprising: Analyze historical data through machine learning algorithms to predict the arrival and departure patterns of vehicles in the future, and allocate parking spaces to vehicles in advance based on the predicted information; When a vehicle enters the warehouse, the barcode scanning device is used to scan the QR code on the vehicle body to obtain the frame number, and the frame number information is matched with the predicted vehicle arrival mode information to see if they are consistent. If they are consistent, a parking space is directly allocated. If they are inconsistent, the vehicle information in the storage database is matched based on the frame number to obtain vehicle-related information, and a parking space is allocated to the vehicle based on the vehicle-related information.

[0017] The working principle and effect of the above technical solution are as follows: Data collection and processing: Collect historical vehicle arrival, stay and departure data, record each vehicle's frame number, entry time, departure time and other relevant information; Prediction model construction, use machine learning algorithms (time series analysis) to analyze historical data, predict future vehicle arrival and departure patterns, the model considers time factors, seasonal changes, weather conditions, special events and other influencing factors; Advance parking space allocation, according to the predicted vehicle arrival pattern, plan the general arrival time period of the vehicle in advance, and allocate potential parking spaces in advance according to the vehicle's historical behavior pattern; Entry verification and parking space allocation, when the vehicle arrives, Scan the QR code on the vehicle body with a barcode scanning device, extract the frame number, match the frame number with the arrival mode information in the prediction model, if the frame number is consistent with the arrival mode information in the prediction model, guide the vehicle into the warehouse directly according to the pre-allocated parking space, if the frame number is inconsistent with the arrival mode information in the prediction model, retrieve the vehicle information (such as vehicle model, frequency of use, historical stay duration, etc.) from the warehouse database based on the frame number, and re-implement parking space allocation based on this information; Dynamic adjustment and optimization: real-time monitoring of vehicle entry and exit, adjust the prediction model to cope with changes, and continuously calibrate and optimize the machine learning model to improve prediction accuracy. Improve efficiency. Predicting and allocating parking spaces in advance reduces waiting time when vehicles arrive, improves the operational efficiency of parking lots, reduces human intervention, and speeds up parking space allocation through automated processes. Enhance user experience. Vehicles do not need to search for parking spaces for a long time, which improves driver satisfaction. Systematic management makes the parking process smoother. Improve parking space utilization. Through accurate prediction and allocation optimization, the turnover rate and utilization rate of parking spaces are improved, and the distribution of parking spaces can be reasonably planned to reduce idle parking spaces, especially during peak vehicle traffic. The system can adapt to changes and dynamically adjust the allocation logic to deal with temporary events or unexpected situations. The solution can be expanded to multiple parking lots for collaborative management to improve overall efficiency. Data-driven decision-making helps managers understand the usage patterns and trends of parking lots, thereby optimizing strategies and resource allocation, and providing a clear basis for future development and layout by collecting and analyzing data. It has good results in improving management efficiency, improving user experience, and optimizing resource utilization, reflecting the value of machine learning and automation technology in practical applications.

[0018] One embodiment of the present invention uses a machine learning algorithm to analyze historical data, predict the arrival and departure patterns of vehicles in the future, and allocate parking spaces to vehicles in advance based on the predicted information, including: Collecting historical vehicle travel data from the monitoring channel of the warehouse management system, wherein the historical vehicle travel data includes the arrival timestamp of the vehicle, the departure timestamp of the vehicle, and the parking time of the vehicle; Removing duplicate data, erroneous data and outliers from historical vehicle travel data, converting arrival timestamps and departure timestamps of vehicles into a unified date and time format, and extracting useful features from the converted data, wherein the useful features include the distribution of arrival and departure times, vehicle types and parking space usage; Divide the historical driving data of vehicles from which historical features have been extracted into a training set and a test set; Using training set data to train a time series prediction model, and adjusting parameters of the time series prediction model to optimize prediction performance; Verify the accuracy and generalization ability of the model through test set data; The trained machine learning model is used to predict the vehicle arrival and departure patterns in the future, generate prediction results, and calculate the parking space demand in each future time period based on the predicted results of vehicle arrival and departure.

[0019] Data collection, collect vehicle historical driving data, including vehicle arrival timestamp, departure timestamp, vehicle frame number (for vehicle identification), and vehicle category, such as car, truck, etc.; remove duplicate records, check and correct obviously erroneous data (such as negative parking time), identify and remove outliers through statistical methods (such as box plot analysis), and convert timestamps to a unified date and time format; extract useful features from the original data, such as: time period features: such as morning peak, noon, and evening; parking time: calculate the parking time of each vehicle, arrival and departure hours: extract and analyze arrival and departure hours, and find daily and weekly patterns; feature conversion, convert timestamps into periodic features such as day of the week, month, hour, etc.; divide the data into training sets and test sets, for example, it can be divided by time: use the first three months of data as the training set and the last month of data as the test set, and use time series prediction models such as LSTM, model training: use the training set to train the selected model. Tune the model parameters to optimize its prediction performance. Parameters can be adjusted using methods such as grid search or Bayesian optimization, model performance can be evaluated using test set data, prediction accuracy can be evaluated using metrics such as RMSE and MAE, and model accuracy and generalization can be further verified using visualization methods (such as a comparison chart of predicted values ​​versus true values). Future vehicle arrival and departure patterns can be predicted, and the trained model can be used to predict vehicle arrival and departure patterns for future time periods (for example, next month). Based on the prediction results, the demand for parking lots in different time periods can be calculated, and parking can be arranged in advance based on the predicted parking demand to optimize parking allocation and utilization.

[0020] The working principle and effect of the above technical solution are as follows: collect historical vehicle driving data from the warehouse management system, including arrival timestamp, departure timestamp and parking duration, etc.; preprocess data to remove duplicate data, erroneous data and outliers, improve data quality, and convert timestamp data to a unified date and time format to ensure data consistency; extract features, Extract useful features from the cleaned data, such as: the distribution of arrival and departure times (peak time period), vehicle type (such as small cars, large cars), parking space usage (such as daily usage rate, average stay time); data set division, divide the processed data set into training set and test set to ensure the independence of model training and evaluation, model training and optimization, use the training set data to train the time series prediction model, and optimize the model's prediction performance through parameter tuning and cross-validation; model verification, use the test set data to verify the accuracy and generalization ability of the model, and evaluate its prediction effect in different time periods; Perform forecasting and demand calculations, apply trained and verified machine learning models to predict vehicle arrival and departure patterns in future time periods, calculate parking space demand in future time periods based on predicted vehicle flow information, and help parking lot managers plan resources in advance. Improved prediction accuracy: Through high-quality data preprocessing and model optimization, the accuracy of parking demand prediction is improved, and prediction errors caused by data noise and outliers are reduced; optimized parking space management: accurate parking demand prediction is provided, so that parking lot managers can allocate and manage parking space resources more efficiently, plan parking space configuration and allocation in advance, reduce parking lot congestion, and improve parking space utilization; reduced operating costs: through accurate demand prediction, the parking lot’s manpower and equipment scheduling is optimized, operating costs are reduced, and economic losses caused by parking space shortages or improper allocation are reduced; improved user experience: accurate parking space allocation reduces vehicle waiting time, improves user satisfaction, and helps drivers meet parking needs while achieving a smooth parking experience; practicality and scalability: the system can be expanded to multi-location parking lots, improving the efficiency of the overall parking network, with wide adaptability, and can integrate more external data sources (such as weather, urban events) to improve the comprehensiveness of predictions; data-driven decision support: detailed historical features and predictive analysis are provided to support data-based decision making, better understand vehicle flow patterns and parking behaviors, and optimize long-term parking lot development plans. Through the application of machine learning, the intelligent technology potential is demonstrated in improving parking resource management efficiency and user satisfaction.

[0021] In one embodiment of the present invention, a digital warehousing method based on task behavior is provided. When a vehicle enters the warehouse, a barcode scanning device is used to scan a QR code on the vehicle body to obtain a frame number, and vehicle related information is obtained based on the frame number matching data entry in a storage database. A parking space is allocated to the vehicle based on the vehicle related information, including: Scan the QR code on the vehicle body with a barcode scanner to obtain the vehicle frame number; Automatically input the frame number into a storage database, and the storage database searches for the frame number information to obtain vehicle related information matching the frame number information, the vehicle related information including: frame number, vehicle model, source order number and customer; The vehicle model in the vehicle-related information is obtained, the vehicle size is obtained based on the vehicle model, and a parking space is allocated to the vehicle in an idle parking space whose size is greater than or equal to the vehicle size by using a parking space allocation model.

[0022] The working principle and effect of the above technical solution are as follows: QR code scanning and data acquisition, using a barcode scanning device to scan the QR code on the vehicle body to obtain the frame number, and entering the obtained frame number information into the warehouse database through an interface or an automated program; database information search, searching for the frame number information in the warehouse database, and obtaining vehicle-related information matching the frame number, including: vehicle model: used to identify the vehicle category, source order number: indicating the source information of the vehicle (such as purchase, lease, etc.), customer: owner or user information of the associated vehicle; vehicle size acquisition, querying standardized vehicle size information (such as length, width, height) according to the vehicle model, ensuring the accuracy of the vehicle size information, so as to make correct parking space allocation; parking space allocation, in the currently vacant parking spaces in the parking lot, screening out candidate parking spaces that are greater than or equal to the vehicle size according to the parking space size, and using the parking space allocation model, considering other factors, allocating a suitable parking space for the vehicle. Automation and efficiency improvement eliminate the step of manually inputting the vehicle frame number, reduce manual errors, and improve data entry efficiency. The automatic acquisition and processing of data makes parking space allocation faster and more efficient. Accurate parking space allocation, based on the matching of the actual size of the vehicle, improves the effective utilization of parking spaces, avoids resource waste caused by size mismatch, optimizes the spatial arrangement of the parking lot, and improves the turnover rate of parking spaces. Customer experience is improved, providing fast and convenient parking services, reducing customer waiting time, and customized parking space allocation improves customer satisfaction, especially when considering customer preferences and vehicle types. Data management optimization, accurate management of vehicles and related information in the database, helps to improve the operation and management level of the entire parking lot, and improves the quality of subsequent analysis and decision-making through accurate data support. The parking and information records of each vehicle are traceable, which facilitates management and processing when problems arise, ensuring the transparency and security of vehicle and parking space management. By integrating QR code technology with the database system and combining the intelligent parking space allocation model, this technical solution makes full use of data resources, realizes the automation and intelligence of parking management, and significantly improves operational efficiency and service levels.

[0023] In one embodiment of the present invention, specifically, the parking space allocation model is: in, represents the priority of parking space i, represents the size fit of parking space i, Indicates the driving distance from the parking space to the garage entrance. represents the distance constant, Indicates the estimated parking time. represents the time constant, represents the functional matching factor, represents the seasonal impact factor. =1- , where R represents the remaining area of ​​the parking space after parking, and T represents the total area of ​​the parking space. Represents a distance constant, with a value of 1 and a unit of Stay consistent. Represents a time constant, with a value of 1 and a unit similar to Stay consistent. in, Indicates the current seasonal usage rate of the parking space. Indicates the average annual utilization rate of the parking space. Indicates the maximum parking space utilization rate, in, Function matching weight, the importance of this function to user needs. It can use a value from 0 to 1, where 1 means a perfect match and 0 means no match. It represents the user function usage demand index, which refers to the degree to which the actual parking space function meets the user's needs. It can be obtained through surveys or questionnaires, for example: electric parking space equipment, such as whether there are charging equipment; restriction boundaries, such as restrictions on disabled license plates; n represents the number of valid functional items, and the overall functional matching is obtained by comprehensively scoring each function. Assume that the following four parking space functions are evaluated, as well as the user's needs for each of these functions: Charging equipment (Electric Vehicle Charging) function matching weight M1 = 0.8 (users really need charging equipment), usage demand index U1 = 1.0 (the parking space has charging piles, which fully meets the needs), awning or covering (Shelter / Covering) function matching weight M2 = 0.5 (users need this function more), usage demand index U2 = 0.6 (the parking space is partially blocked), security monitoring (Security Surveillance) function matching weight M3 = 0.7 (users value the safety of parking spaces), usage demand index U3 = 0.9 (the parking space is within the monitoring coverage), accessibility for disabled facilities (Accessibility for Disabled) function matching weight M4 = 0.4 (users have general needs for this), usage demand index U4 = 0.0 (the parking space does not have such facilities), the contribution of each function to the matching degree is calculated item by item, the number of valid function items n=4, and the final function matching factor is calculated to be 0.4325, reflecting that the overall matching degree between the function of the parking space and the user needs is 43.25%.

[0024] The effects of the above technical solution are: optimizing parking space allocation, improving the rationality of parking space allocation by comprehensively considering multiple influencing factors, so as to achieve more precise services for different drivers and vehicle needs; dynamic adjustment capability, the method can be dynamically adjusted according to the actual season and parking space usage to maintain efficient and flexible parking space allocation; improving user satisfaction, by better matching user needs (such as charging facilities and safety, etc.), the method is expected to improve the overall user experience and satisfaction; resource management optimization, through scientific evaluation and allocation, it helps to better utilize existing parking resources and reduce resource waste caused by incorrect matching; by combining the model with considerations of multiple dimensions, it can effectively improve the intelligence level of parking space allocation and management, adapt to changing usage scenarios, and improve the efficiency of the overall parking system and user happiness.

[0025] One embodiment of the present invention is a digital storage system based on task behavior, the system comprising: The prediction allocation module is used to analyze historical data through machine learning algorithms to predict the arrival and departure patterns of vehicles in the future and allocate parking spaces to vehicles in advance based on the predicted information; The warehousing allocation module is used to obtain the frame number by scanning the QR code on the vehicle body through a barcode scanning device when the vehicle enters the warehouse, and to match the frame number information with the predicted vehicle arrival mode information to see if they are consistent. If they are consistent, a parking space is directly allocated. If they are inconsistent, the vehicle information in the storage database is matched based on the frame number to obtain vehicle-related information, and a parking space is allocated to the vehicle based on the vehicle-related information.

[0026] The working principle and effect of the above technical solution are as follows: Data collection and processing: Collect historical vehicle arrival, stay and departure data, record each vehicle's frame number, entry time, departure time and other relevant information; Prediction model construction, use machine learning algorithms (time series analysis) to analyze historical data, predict future vehicle arrival and departure patterns, the model considers time factors, seasonal changes, weather conditions, special events and other influencing factors; Advance parking space allocation, according to the predicted vehicle arrival pattern, plan the general arrival time period of the vehicle in advance, and allocate potential parking spaces in advance according to the vehicle's historical behavior pattern; Entry verification and parking space allocation, when the vehicle arrives, Scan the QR code on the vehicle body with a barcode scanning device, extract the frame number, match the frame number with the arrival mode information in the prediction model, if the frame number is consistent with the arrival mode information in the prediction model, guide the vehicle into the warehouse directly according to the pre-allocated parking space, if the frame number is inconsistent with the arrival mode information in the prediction model, retrieve the vehicle information (such as vehicle model, frequency of use, historical stay duration, etc.) from the warehouse database based on the frame number, and re-implement parking space allocation based on this information; Dynamic adjustment and optimization: real-time monitoring of vehicle entry and exit, adjust the prediction model to cope with changes, and continuously calibrate and optimize the machine learning model to improve prediction accuracy. Improve efficiency. Predicting and allocating parking spaces in advance reduces waiting time when vehicles arrive, improves the operational efficiency of parking lots, reduces human intervention, and speeds up parking space allocation through automated processes. Enhance user experience. Vehicles do not need to search for parking spaces for a long time, which improves driver satisfaction. Systematic management makes the parking process smoother. Improve parking space utilization. Through accurate prediction and allocation optimization, the turnover rate and utilization rate of parking spaces are improved, and the distribution of parking spaces can be reasonably planned to reduce idle parking spaces, especially during peak vehicle traffic. The system can adapt to changes and dynamically adjust the allocation logic to deal with temporary events or unexpected situations. The solution can be expanded to multiple parking lots for collaborative management to improve overall efficiency. Data-driven decision-making helps managers understand the usage patterns and trends of parking lots, thereby optimizing strategies and resource allocation, and providing a clear basis for future development and layout by collecting and analyzing data. It has good results in improving management efficiency, improving user experience, and optimizing resource utilization, reflecting the value of machine learning and automation technology in practical applications.

[0027] In one embodiment of the present invention, the prediction allocation module includes: A module for acquiring vehicle historical driving data is used to collect vehicle historical driving data from the monitoring channel of the warehouse management system, wherein the vehicle historical driving data includes the vehicle arrival timestamp, the vehicle departure timestamp and the vehicle parking time; A preprocessing module, used to remove duplicate data, erroneous data and outliers in the historical driving data of the vehicle, convert the arrival timestamp and departure timestamp of the vehicle into a unified date and time format, and extract useful features from the converted data, wherein the useful features include the distribution of arrival and departure times, vehicle type and parking space usage; A data set partitioning module is used to partition the historical driving data of vehicles from which historical features have been extracted into a training set and a test set; A training module is used to train a time series prediction model using training set data and adjust parameters of the time series prediction model to optimize prediction performance; The testing module is used to verify the accuracy and generalization ability of the model through the test set data; The prediction-based parking allocation module is used to use the trained machine learning model to predict the vehicle arrival and departure patterns in the future, generate prediction results, and calculate the parking space demand in each future time period based on the prediction results of vehicle arrival and departure.

[0028] The working principle and effect of the above technical solution are as follows: collect historical vehicle driving data from the warehouse management system, including arrival timestamp, departure timestamp and parking duration, etc.; data preprocessing, remove duplicate data, erroneous data and outliers, improve data quality, and convert timestamp data to a unified date and time format to ensure data consistency; feature extraction, extract useful features from the cleaned data, such as: the distribution of arrival and departure times (peak time period), vehicle type (such as small cars, large cars), parking space usage (such as daily usage rate, average stay time); data set division, divide the processed data set into training set and test set to ensure the independence of model training and evaluation, model training and optimization, use the training set data to train the time series prediction model, and optimize the model's prediction performance through parameter tuning and cross-validation; model verification, use the test set data to verify the accuracy and generalization ability of the model, and evaluate its prediction effect in different time periods; Perform forecasting and demand calculations, apply trained and verified machine learning models to predict vehicle arrival and departure patterns in future time periods, calculate parking space demand in future time periods based on predicted vehicle flow information, and help parking lot managers plan resources in advance. Improved prediction accuracy: Through high-quality data preprocessing and model optimization, the accuracy of parking demand prediction is improved, and prediction errors caused by data noise and outliers are reduced; optimized parking space management: accurate parking demand prediction is provided, so that parking lot managers can allocate and manage parking space resources more efficiently, plan parking space configuration and allocation in advance, reduce parking lot congestion, and improve parking space utilization; reduced operating costs: through accurate demand prediction, the parking lot’s manpower and equipment scheduling is optimized, operating costs are reduced, and economic losses caused by parking space shortages or improper allocation are reduced; improved user experience: accurate parking space allocation reduces vehicle waiting time, improves user satisfaction, and helps drivers meet parking needs while achieving a smooth parking experience; practicality and scalability: the system can be expanded to multi-location parking lots, improving the efficiency of the overall parking network, with wide adaptability, and can integrate more external data sources (such as weather, urban events) to improve the comprehensiveness of predictions; data-driven decision support: detailed historical features and predictive analysis are provided to support data-based decision making, better understand vehicle flow patterns and parking behaviors, and optimize long-term parking lot development plans. Through the application of machine learning, the intelligent technology potential is demonstrated in improving parking resource management efficiency and user satisfaction.

[0029] In one embodiment of the present invention, the warehousing allocation module includes: The module for obtaining the vehicle frame number is used to obtain the vehicle frame number by scanning the vehicle body QR code through a barcode scanning device; A module for searching vehicle related information, used for automatically entering the frame number into a storage database, wherein the storage database searches for the frame number information to obtain vehicle related information matching the frame number information, wherein the vehicle related information includes: frame number, vehicle model, source order number and customer; The parking space allocation module is used to obtain the vehicle model in the vehicle-related information, obtain the vehicle size based on the vehicle model, and allocate a parking space to the vehicle in an idle parking space whose size is greater than or equal to the vehicle size by using the parking space allocation model.

[0030] The working principle and effect of the above technical solution are as follows: QR code scanning and data acquisition, using a barcode scanning device to scan the QR code on the vehicle body to obtain the frame number, and entering the obtained frame number information into the warehouse database through an interface or an automated program; database information search, searching for the frame number information in the warehouse database, and obtaining vehicle-related information matching the frame number, including: vehicle model: used to identify the vehicle category, source order number: indicating the source information of the vehicle (such as purchase, lease, etc.), customer: owner or user information of the associated vehicle; vehicle size acquisition, querying standardized vehicle size information (such as length, width, height) according to the vehicle model, ensuring the accuracy of the vehicle size information, so as to make correct parking space allocation; parking space allocation, in the currently vacant parking spaces in the parking lot, screening out candidate parking spaces that are greater than or equal to the vehicle size according to the parking space size, and using the parking space allocation model, considering other factors, allocating a suitable parking space for the vehicle. Automation and efficiency improvement eliminate the step of manually inputting the vehicle frame number, reduce manual errors, and improve data entry efficiency. The automatic acquisition and processing of data makes parking space allocation faster and more efficient. Accurate parking space allocation, based on the matching of the actual size of the vehicle, improves the effective utilization of parking spaces, avoids resource waste caused by size mismatch, optimizes the spatial arrangement of the parking lot, and improves the turnover rate of parking spaces. Customer experience is improved, providing fast and convenient parking services, reducing customer waiting time, and customized parking space allocation improves customer satisfaction, especially when considering customer preferences and vehicle types. Data management optimization, accurate management of vehicles and related information in the database, helps to improve the operation and management level of the entire parking lot, and improves the quality of subsequent analysis and decision-making through accurate data support. The parking and information records of each vehicle are traceable, which facilitates management and processing when problems arise, ensuring the transparency and security of vehicle and parking space management. By integrating QR code technology with the database system and combining the intelligent parking space allocation model, this technical solution makes full use of data resources, realizes the automation and intelligence of parking management, and significantly improves operational efficiency and service levels.

[0031] In one embodiment of the present invention, the parking space allocation module includes: The storage allocation parking space model module, specifically, the parking space allocation model is: in, represents the priority of parking space i, represents the size fit of parking space i, Indicates the driving distance from the parking space to the garage entrance. represents the distance constant, Indicates the estimated parking time. represents the time constant, represents the functional matching factor, represents the seasonal impact factor.

[0032] =1- , where R represents the remaining area of ​​the parking space after parking, and T represents the total area of ​​the parking space. Represents a distance constant, with a value of 1 and a unit of Stay consistent. Represents a time constant, with a value of 1 and a unit similar to Stay consistent. in, Indicates the current seasonal usage rate of the parking space. Indicates the average annual utilization rate of the parking space. Indicates the maximum parking space utilization rate, in, Function matching weight, the importance of this function to user needs. It can use a value from 0 to 1, where 1 means a perfect match and 0 means no match. It represents the user function usage demand index, which refers to the degree to which the actual parking space function meets the user's needs. It can be obtained through surveys or questionnaires, for example: electric parking space equipment, such as whether there are charging equipment; restriction boundaries, such as restrictions on disabled license plates; n represents the number of valid functional items, and the overall functional matching is obtained by comprehensively scoring each function. Assume that the following four parking space functions are evaluated, as well as the user's needs for each of these functions: Charging equipment (Electric Vehicle Charging) function matching weight M1 = 0.8 (users really need charging equipment), usage demand index U1 = 1.0 (the parking space has charging piles, which fully meets the needs), awning or covering (Shelter / Covering) function matching weight M2 = 0.5 (users need this function more), usage demand index U2 = 0.6 (the parking space is partially blocked), security monitoring (Security Surveillance) function matching weight M3 = 0.7 (users value the safety of parking spaces), usage demand index U3 = 0.9 (the parking space is within the monitoring coverage), accessibility for disabled facilities (Accessibility for Disabled) function matching weight M4 = 0.4 (users have general needs for this), usage demand index U4 = 0.0 (the parking space does not have such facilities), the contribution of each function to the matching degree is calculated item by item, the number of valid function items n=4, and the final function matching factor is calculated to be 0.4325, reflecting that the overall matching degree between the function of the parking space and the user needs is 43.25%.

[0033] The effects of the above technical solution are: optimizing parking space allocation, improving the rationality of parking space allocation by comprehensively considering multiple influencing factors, so as to achieve more precise services for different drivers and vehicle needs; dynamic adjustment capability, the method can be dynamically adjusted according to the actual season and parking space usage to maintain efficient and flexible parking space allocation; improving user satisfaction, by better matching user needs (such as charging facilities and safety, etc.), the method is expected to improve the overall user experience and satisfaction; resource management optimization, through scientific evaluation and allocation, it helps to better utilize existing parking resources and reduce resource waste caused by incorrect matching; by combining the model with considerations of multiple dimensions, it can effectively improve the intelligence level of parking space allocation and management, adapt to changing usage scenarios, and improve the efficiency of the overall parking system and user happiness.

[0034] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A digital warehousing method based on task behavior, characterized in that: The method comprises: Analyze historical data through machine learning algorithms to predict the arrival and departure patterns of vehicles in the future, and allocate parking spaces to vehicles in advance based on the predicted information; When a vehicle enters the warehouse, the barcode scanning device is used to scan the QR code on the vehicle body to obtain the frame number, and the frame number information is matched with the predicted vehicle arrival mode information to see if they are consistent. If they are consistent, a parking space is directly allocated. If they are inconsistent, the vehicle information in the storage database is matched based on the frame number to obtain vehicle-related information, and a parking space is allocated to the vehicle based on the vehicle-related information.

2. According to the task behavior-based digital warehousing method of claim 1, it is characterized in that: Analyze historical data through machine learning algorithms to predict the arrival and departure patterns of vehicles in the future, and allocate parking spaces to vehicles in advance based on the predicted information, including: Collecting historical vehicle travel data from the monitoring channel of the warehouse management system, wherein the historical vehicle travel data includes the arrival timestamp of the vehicle, the departure timestamp of the vehicle, and the parking time of the vehicle; Removing duplicate data, erroneous data and outliers from historical vehicle travel data, converting arrival timestamps and departure timestamps of vehicles into a unified date and time format, and extracting useful features from the converted data, wherein the useful features include the distribution of arrival and departure times, vehicle types and parking space usage; Divide the historical driving data of vehicles from which historical features have been extracted into a training set and a test set; Using training set data to train a time series prediction model, and adjusting parameters of the time series prediction model to optimize prediction performance; Verify the accuracy and generalization ability of the model through test set data; The trained machine learning model is used to predict the vehicle arrival and departure patterns in the future, generate prediction results, and calculate the parking space demand in each future time period based on the predicted results of vehicle arrival and departure.

3. According to the digital warehousing method based on task behavior as claimed in claim 1, it is characterized in that: When a vehicle enters the warehouse, the barcode scanning device scans the QR code on the vehicle body to obtain the frame number, matches the data entry in the storage database based on the frame number to obtain vehicle-related information, and allocates a parking space for the vehicle based on the vehicle-related information, including: Scan the QR code on the vehicle body with a barcode scanner to obtain the vehicle frame number; Automatically input the frame number into a storage database, and the storage database searches for the frame number information to obtain vehicle related information matching the frame number information, the vehicle related information including: frame number, vehicle model, source order number and customer; The vehicle model in the vehicle-related information is obtained, the vehicle size is obtained based on the vehicle model, and a parking space is allocated to the vehicle in an idle parking space whose size is greater than or equal to the vehicle size by using a parking space allocation model.

4. According to the task behavior-based digital warehousing method of claim 1, it is characterized in that: Specifically, the parking space allocation model is: in, represents the priority of parking space i, represents the size fit of parking space i, Indicates the driving distance from the parking space to the garage entrance. represents the distance constant, Indicates the estimated parking time. represents the time constant, represents the functional matching factor, represents the seasonal impact factor.

5. A digital warehousing system based on task behavior, characterized in that: The system comprises: The prediction allocation module is used to analyze historical data through machine learning algorithms to predict the arrival and departure patterns of vehicles in the future and allocate parking spaces to vehicles in advance based on the predicted information; The warehousing allocation module is used to obtain the frame number by scanning the QR code on the vehicle body through a barcode scanning device when the vehicle enters the warehouse, and to match the frame number information with the predicted vehicle arrival mode information to see if they are consistent. If they are consistent, a parking space is directly allocated. If they are inconsistent, the vehicle information in the storage database is matched based on the frame number to obtain vehicle-related information, and a parking space is allocated to the vehicle based on the vehicle-related information.

6. A digital warehousing system based on task behavior according to claim 5, characterized in that: The prediction allocation module includes: A module for acquiring vehicle historical driving data is used to collect vehicle historical driving data from the monitoring channel of the warehouse management system, wherein the vehicle historical driving data includes the vehicle arrival timestamp, the vehicle departure timestamp and the vehicle parking time; A preprocessing module, used to remove duplicate data, erroneous data and outliers in the historical driving data of the vehicle, convert the arrival timestamp and departure timestamp of the vehicle into a unified date and time format, and extract useful features from the converted data, wherein the useful features include the distribution of arrival and departure times, vehicle type and parking space usage; A data set partitioning module is used to partition the historical driving data of vehicles from which historical features have been extracted into a training set and a test set; A training module is used to train a time series prediction model using training set data and adjust parameters of the time series prediction model to optimize prediction performance; The testing module is used to verify the accuracy and generalization ability of the model through the test set data; The prediction-based parking allocation module is used to use the trained machine learning model to predict the vehicle arrival and departure patterns in the future, generate prediction results, and calculate the parking space demand in each future time period based on the prediction results of vehicle arrival and departure.

7. A task-behavior-based digital warehousing system according to claim 5, characterized in that: The warehousing allocation module includes: The module for obtaining the vehicle frame number is used to obtain the vehicle frame number by scanning the vehicle body QR code through a barcode scanning device; A module for searching vehicle related information, used for automatically entering the frame number into a storage database, wherein the storage database searches for the frame number information to obtain vehicle related information matching the frame number information, wherein the vehicle related information includes: frame number, vehicle model, source order number and customer; The parking space allocation module is used to obtain the vehicle model in the vehicle-related information, obtain the vehicle size based on the vehicle model, and allocate a parking space to the vehicle in an idle parking space whose size is greater than or equal to the vehicle size by using the parking space allocation model.

8. A task-behavior-based digital warehousing system according to claim 7, characterized in that: The parking space allocation module includes: The storage allocation parking space model module, specifically, the parking space allocation model is: in, represents the priority of parking space i, represents the size fit of parking space i, Indicates the driving distance from the parking space to the garage entrance. represents the distance constant, Indicates the estimated parking time. represents the time constant, represents the functional matching factor, represents the seasonal impact factor.

Images