Intelligent logistics warehouse management method

Abstract

The invention discloses an intelligent logistics warehouse management method, and relates to the technical field of warehouse management, and the method comprises the steps: obtaining an original data set, carrying out the shared information quantitative analysis of the original data set, obtaining an information quantitative index, carrying out the preliminary filtering of the original data set according to the information quantitative index, and obtaining a call data set; a basic database is generated, basic data in the calling data set and the corresponding multi-level identification codes are recognized according to the basic database, other data in the calling data set are marked as identification data, and the multi-level identification codes and the identification data form a shared data set; data sharing is carried out according to the shared data set, and a warehouse management decision result is generated; by adding multi-dimensional data features, the technical effects of dynamically adapting to data changes in multiple dimensions and improving the accuracy and practicability of data evaluation are achieved; dynamic optimization and self-adaptive adjustment of the information quantification index are realized, and scientificity and efficiency of warehouse management decision making are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse management, and in particular to an intelligent logistics warehouse management method. Background Art

[0002] Warehouse management is an important link in the logistics transportation chain. In-depth research on logistics warehouse management methods is of great significance for reducing transportation costs and improving transportation efficiency. At present, the analysis of warehouse management mainly focuses on the optimization of a single-node warehouse. By analyzing the warehouse needs of a single enterprise, a central warehouse and a distribution center are established to meet the enterprise's needs for warehouse management. However, with the continuous improvement of the supply chain, there are more and more enterprises involved in the product production chain, and their businesses are intertwined. In the process of warehouse management, it is impossible to manage only one enterprise. It is necessary to share warehouse information to improve management efficiency and avoid the consequence that the warehouse cannot meet the requirements due to the too long update cycle of warehouse information. There is a technical problem in the prior art that the accuracy of data sharing analysis in logistics warehouse management is low, resulting in chaotic warehouse management and unable to carry out normal logistics storage.

[0003] The Chinese invention patent with the application number 202310035890.9 provides a logistics warehouse management method and system based on information sharing, which obtains multiple information source modules for logistics warehouse management; calls data according to the multiple information source modules to obtain multiple call data sets; inputs them into an information sharing platform; performs shared information quantification analysis on the multiple call data sets by a cloud server to obtain an information quantification index; obtains an information identification instruction according to the information quantification index; inputs it into a matrix index model for identification to obtain multiple filtered data sets; connects the shared modules corresponding to the multiple filtered data sets for data sharing to generate a warehouse management decision result.

[0004] However, in the above-mentioned invention patent technology, calculating the information quantification index by relying on the call type quantification index and the call traffic quantification index results in the inability to comprehensively evaluate the importance and complexity of data, and the inability to timely reflect data changes, which is a technical problem that needs to be solved urgently. Summary of the Invention

[0005] This application provides an intelligent logistics warehouse management method, which solves the problem in the prior art that the importance and complexity of data cannot be comprehensively and timely evaluated, and realizes the technical effects of dynamically adapting to data changes in multiple dimensions, improving the accuracy and practicality of data evaluation.

[0006] This application provides an intelligent logistics warehouse management method, including:

[0007] S100: Obtain the original data set, perform a quantification analysis of the shared information on the original data set to obtain an information quantification index, and preliminarily filter the original data set according to the information quantification index to obtain a call data set;

[0008] S200: Generate a basic database, identify the basic data and the corresponding multi-level identification codes in the call data set according to the basic database, mark the remaining data in the call data set as identification data, and form a shared data set with the multi-level identification codes and the identification data; perform data sharing according to the shared data set to generate a warehouse management decision result.

[0009] Further, in step S100, it further includes:

[0010] S110: Capture each data item in the data stream in real time to form multiple sub-data sets, and then fuse them to form an original data set, and use stream processing technology to perform a quantification analysis of the shared information on the original data set to form a processed original data set;

[0011] S120: Extract multi-dimensional data features from the processed original data set and analyze to obtain corresponding feature values, and perform a weighted sum of the feature values to obtain an information quantification index;

[0012] S130: Preset a quantification index threshold, and mark the data corresponding to the information quantification index less than the quantification index threshold as callable data;

[0013] Filter the data corresponding to the information quantification index not less than the quantification index threshold, and recalculate the information quantification index of the filtered data. If it is less than the quantification index threshold, mark it as callable data; if it is not less than the quantification index threshold, discard it;

[0014] S130: Obtain all callable data to form a call data set.

[0015] Further, the sub-data sets include: an original order information sub-data set, an original transport capacity information sub-data set, and an original transport monitoring sub-data set; the multi-dimensional data features include data timeliness, data integrity, data importance, and data transmission.

[0016] Further, in step S200, generating a basic database includes: obtaining a historical shared data set, classifying the data in the historical shared data set into basic data and identification data; setting a unique multi-level identification code for each basic data, and storing the basic data and the unique multi-level identification code in a database to form a basic database; wherein, the corresponding relationship between the basic data and the identification data is one-to-many.

[0017] Further, the basic data includes first basic data and second basic data. The first basic data refers to the basic data whose data update frequency is not greater than a preset first frequency threshold, and the second basic data is identification data that meets preset conditions; the multi-level identification code includes a head and a tail, the head corresponds to the first basic data, and the tail corresponds to the second basic data.

[0018] Further, the preset conditions are set as follows: a monitoring period is preset, all the identification data transmitted within the monitoring period is obtained, the transmission frequencies of different identification data are calculated, a second frequency threshold is preset, and the identification data corresponding to the transmission frequencies greater than the second frequency threshold is marked as the second basic data.

[0019] Further, the warehouse management decision result includes the current warehouse task decision result and the prediction result of the warehouse task for the next monitoring period. The current warehouse task decision result is obtained based on the shared data set;

[0020] The second basic data in multiple monitoring periods is obtained and marked as analysis data, the dynamic change factors in the warehouse logistics are identified, and based on the analysis data and the analysis result of the dynamic change factors, a prediction model is constructed, and then the prediction result of the warehouse task for the next monitoring period is generated.

[0021] Further, identifying the dynamic change factors in the warehouse logistics includes:

[0022] Using the association rule mining technology to perform data mining on the analysis data, and converting the analysis data into numerical data; generating frequent item sets in a layer-by-layer search manner, setting a minimum support and a minimum confidence, and constructing an association rule mining model;

[0023] Inputting the numerical data into the association rule mining model to obtain frequent item sets and association rules, and then identifying the dynamic change factors in the warehouse logistics and the relationships between the dynamic change factors;

[0024] Among them, the dynamic change factors include transportation vehicle scheduling, cargo scheduling, and transportation date.

[0025] Further, constructing a prediction model based on the analysis data and the analysis result of the dynamic change factors includes:

[0026] Preprocessing the analysis data to obtain the processed analysis data, dividing the processed analysis data into a training set and a validation set, using a time series model as the basic model, training the time series model with the training set to obtain an initial model, validating and optimizing the initial model based on the validation set to obtain a prediction model; using the prediction model to predict the dynamic change factors in the next monitoring period, and then obtaining the prediction result of the warehouse task for the next monitoring period.

[0027] Furthermore, the transportation vehicle scheduling includes the departure time, arrival time, driving route, load capacity, etc. of the vehicle, the cargo scheduling includes the warehousing time, outbound time, storage location, cargo type, cargo demand, etc. of the cargo, and the transportation date refers to the arrival date agreed upon in advance by the logistics.

[0028] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0029] By adding multi-dimensional data features, the technical effect of dynamically adapting to data changes in multiple dimensions and improving the accuracy and practicality of data evaluation is achieved; by quantifying the information index through multi-dimensional calculations, the dynamic optimization and adaptive adjustment of the information quantification index are realized, and the scientific nature and efficiency of warehouse management decisions are improved; by using stream processing technology, data is collected, processed, and analyzed in real time, further improving the timeliness of warehouse management decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the overall process of an intelligent logistics warehouse management method in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] To facilitate the understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, however, the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, these embodiments are provided to make the disclosure of the present invention more thorough and comprehensive.

[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs; 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; the term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0033] Embodiment 1: As Figure 1 shown, an intelligent logistics warehouse management method, the method includes:

[0034] S100: Obtain an original data set, perform shared information quantification analysis on the original data set to obtain an information quantification index, and preliminarily filter the original data set according to the information quantification index to obtain a call data set.

[0035] In some embodiments, to facilitate data sharing, an information sharing platform may be set up to share warehouse management information in real time. By communicating with the cloud server, reliable cloud storage and information data analysis of the shared information can be performed. During the process of logistics warehouse management, within the first time period, the utilization of transportation capacity resources and transportation process information corresponding to different warehouse orders can all be extracted from the shared information through the information sharing platform according to different types. Thus, the technical effect of providing basic analysis data for transportation capacity management and warehouse capacity management analysis in logistics warehousing is achieved.

[0036] Step S100 further includes: S110: Real-time capture each data item in the data stream to form multiple sub-datasets, and then fuse them to form an original dataset. Use stream processing technology to perform quantitative analysis of the shared information on the original dataset to form a processed original dataset.

[0037] In the prior art, when analyzing data, it is necessary to collect data for a period of time and wait until the data accumulates to a certain amount before performing batch analysis, resulting in a certain data delay and reducing real-time performance and value. In this application, stream processing technology is used to achieve instant data collection, processing, and analysis. The data is captured, processed, and analyzed at the moment it is generated without waiting for data accumulation, greatly shortening the data processing delay and improving the real-time performance and value of the data. In stream processing technology, data streams are usually transmitted and processed through message queues or stream processing platforms, providing an efficient data transmission mechanism, powerful data processing capabilities, and a flexible data processing model, enabling stream processing technology to handle various complex data processing scenarios. Stream processing technology mainly processes data generated in real time, such as order information, transportation capacity information, transportation monitoring data, etc. By real-time capturing each data item in the data stream and processing and analyzing it, the processing methods may include data filtering, data conversion, data aggregation, etc., which are specifically set according to actual needs. The result obtained through stream processing technology is a real-time updated information quantification index, and the index reflects the real-time changes and trends of the data, providing more timely and accurate information support for warehouse management decisions.

[0038] In some embodiments, the sub-datasets include: an original order information sub-dataset, an original transport capacity information sub-dataset, and an original transport monitoring sub-dataset; specifically, to obtain the original dataset, different information sources are first obtained during the process of logistics warehousing management, including: order information, transport capacity information, and transport monitoring information. The order information includes, but is not limited to: order type, order warehouse, order source, order payment status, etc.; the transport capacity information is the resource information for the logistics warehousing company to manage logistics transportation and storage, including, but not limited to: the number of personnel, the number of vehicles, the number of vehicles available for use in real time, etc.; the transport monitoring information is the cargo transfer information obtained by monitoring the logistics transportation process. By collecting the above information in real time or setting a certain time period for collection, the original dataset is the dataset generated during the logistics warehousing management and cargo transfer process collected in real time or within a fixed time period, including the original order information sub-dataset, the original transport capacity information sub-dataset, and the original transport monitoring sub-dataset.

[0039] Step S100 also includes: S120: Extract multi-dimensional data features from the processed original dataset and analyze to obtain corresponding feature values, and perform weighted summation on the feature values to obtain an information quantification index.

[0040] In some embodiments, the multi-dimensional data features include data timeliness, data integrity, data importance, and data transferability; the data timeliness reflects the update frequency and timeliness of the data; the data integrity measures the completeness and accuracy of the data; the evaluation of data importance includes the reputation and reliability of the data provider and the criticality of the data; the data transferability refers to the ease of data transmission. The larger the data volume and the larger the data traffic, the more difficult the transmission; when dealing with data, all features in the multi-dimensional data features should be comprehensively considered, rather than simply considering a single feature. When presetting weight values for different data features, they should be set according to the actual situation or historical data experience. Generally, the weight value of data importance should be set to the largest to ensure that important data can be shared preferentially.

[0041] In some embodiments, there are two ways to obtain the information quantification index. The first way is to regard the original dataset as a whole for calculation to obtain an information quantification index for subsequent processing, which is applicable to the case where the data volume is small and the data is easy to distinguish; the second way is to regard each sub-dataset in the original dataset as a whole, calculate the information quantification index separately for subsequent processing. Compared with the first way, it increases the calculation complexity, but in the subsequent processing process, especially when there are data greater than the quantification index threshold, it is easy to find the corresponding data for reprocessing; in practical applications, it needs to be dynamically adjusted according to the size of the overall data volume and the actual situation, and this application does not make specific limitations here.

[0042] Step S100 further includes: S130: preset a quantization index threshold, and mark the data corresponding to the information quantization index smaller than the quantization index threshold as callable data;

[0043] Filter the data corresponding to the information quantization index not less than the quantization index threshold, and recalculate the information quantization index of the filtered data. If it is less than the quantization index threshold, mark it as callable data; if it is not less than the quantization index threshold, discard it.

[0044] In some embodiments, the quantization index threshold is preset through historical experimental data, which is used to measure the comprehensive characteristics of the data, ensure the importance of the data and the data transferability. If it is not less than the quantization index threshold, it means that the data volume corresponding to the information quantization index is too large at this time and needs to be filtered. Specifically, multiple filtering indicators are preset. The multiple filtering indicators include data status indicators, data value indicators, data repetition indicators, and data timeliness indicators. Based on the data status indicators, data value indicators, data repetition indicators, and data timeliness indicators, generate the matrix index model, and perform identification filtering on the data to be filtered based on the matrix index model; the matrix index model is a functional model for identifying and filtering data, and is generated by multiple indicators. Among them, the multiple filtering indicators are multiple dimensions for evaluating data during the process of screening and eliminating data, including data status indicators, data value indicators, data repetition indicators, and data timeliness indicators. Among them, the data status indicator refers to whether the data is in the process of being executed. If not, sharing can be postponed first. If so, real-time sharing is required. Thus, the data is screened according to the sharing priority. The data value indicator is an indicator for analyzing the value of the data for warehouse management, including the contribution degree of the data to the warehouse. The use value of the data is evaluated according to the data value indicator. The data repetition indicator evaluates the sharing history of the data. If the data has been shared before, it will not be shared again. If it has not been shared, it needs to be uploaded and shared in time. The data timeliness indicator is an indicator for evaluating the timeliness of the data, including data validity. That is, it evaluates whether the data exceeds the usage time limit. Exemplarily, the warehouse management plan data of yesterday has lost its timeliness for today's warehouse management, and today's warehouse management needs to be rearranged according to today's order situation and transportation capacity resource situation. The matrix index model is a 2x2 matrix. Preferably, the data status indicator and the data value indicator are located in the first row of the matrix, and the data repetition indicator and the data timeliness indicator are located in the second row of the matrix. Furthermore, identification filtering is performed on the data according to the matrix index model. Eliminate the data in the data set that does not meet the indicators, recalculate the information quantization index of the filtered data. If it is less than the quantization index threshold, mark it as callable data; if it is not less than the quantization index threshold, discard it.

[0045] The technical solutions in the embodiments of the present application at least have the following technical effects or advantages:

[0046] By adding multi-dimensional data features, the present application achieves the technical effect of dynamically adapting to data changes in multiple dimensions, improving the accuracy and practicality of data evaluation; by calculating the information quantization index in multiple dimensions, the dynamic optimization and adaptive adjustment of the information quantization index are realized, improving the scientificity and efficiency of warehouse management decision-making; by using stream processing technology, data is collected, processed, and analyzed in real time, further improving the timeliness of warehouse management decision-making.

[0047] Embodiment 2: In Embodiment 1, by using stream processing technology and adding multi-dimensional data features, the technical effect of dynamically adapting to data changes in multiple dimensions and improving the accuracy and practicality of data evaluation is achieved. However, a large amount of data information still needs to be transmitted inevitably during data transmission, affecting the data transmission efficiency. This embodiment makes further improvements on the basis of the above embodiment.

[0048] S200: Generate a basic database, identify the basic data and corresponding multi-level identification codes in the call dataset according to the basic database, mark the remaining data in the call dataset as identification data, and form a shared dataset with the multi-level identification codes and identification data; perform data sharing according to the shared dataset to generate a warehouse management decision result.

[0049] In step S200, generating a basic database includes: obtaining a historical shared dataset, classifying the data in the historical shared dataset into basic data and identification data; setting a unique multi-level identification code for each piece of basic data, and storing the basic data and the unique multi-level identification code in the database to form a basic database, and the corresponding relationship between the basic data and the identification data is one-to-many.

[0050] In some embodiments, the basic data refers to data that can be commonly used in the long term, i.e., data that can be used in a single sub-dataset or multiple sub-datasets; the identification data refers to data that cannot be commonly used and needs to be updated each time; the basic data and identification data corresponding to different sub-datasets are also different. For example, in the original sub-dataset of order information, the basic data includes: order type description (such as "ordinary order", "urgent order"), order source category (such as "e-commerce platform", "offline store"), etc.; the identification data includes: order number, specific order time, order payment status, etc. In the original sub-dataset of transportation capacity information, the basic data includes: vehicle type description (such as "truck", "refrigerated truck"), personnel position description (such as "driver", "stevedore"), etc.; the identification data includes: real-time available vehicle number, current status of personnel (such as "on duty", "resting"), etc. In the original sub-dataset of transportation monitoring, the basic data includes: transportation status description (such as "in transportation", "arrived"), etc.; the identification data includes: real-time location of goods, records of abnormal events during transportation, etc.

[0051] In some embodiments, the same basic database is stored in both the managers who receive data and those who share data. A unique multi-level identification code is set for each basic data in the basic database. When sharing data, according to the obtained basic data, the corresponding multi-level identification code is found through the basic database, and the multi-level identification code and the identification data are shared. The receiving party searches for the corresponding basic data according to the multi-level identification code, thereby reducing the amount of data shared. After receiving the data, the receiving party first parses the multi-level identification code, searches for the corresponding basic data in the basic database according to the multi-level identification code, combines the basic data with the identification data, and restores it to a complete shared dataset.

[0052] In this embodiment, by only transmitting the identification code and the identification data, the amount of data transmitted is greatly reduced, and the data transmission efficiency is improved; the receiving party can flexibly search for and combine data according to the identification code, adapting to the data requirements in different scenarios; the update and maintenance mechanism of the basic database can easily handle the emergence and changes of new data items, enhancing the scalability and at the same time improving the adaptability to complex environments.

[0053] By classifying the data in the historical shared dataset into basic data and identification data, setting a unique multi-level identification code for each basic data, and only transmitting the multi-level identification code and the identification data, the receiving party searches for the corresponding basic data in the basic database according to the multi-level identification code.

[0054] This technical means greatly reduces the amount of data that needs to be transmitted because only the identification code and a small amount of identification data need to be transmitted, rather than the complete basic data.

[0055] The technical solutions in the embodiments of the present application at least have the following technical effects or advantages:

[0056] By only transmitting the identification code and the identification data, the present application greatly reduces the amount of data transmitted, achieving the effect of improving the efficiency of data transmission and sharing; flexibly searching for and combining data according to the identification code, adapting to the data requirements in different scenarios, easily coping with the emergence and changes of new data items, enhancing the scalability, and achieving the effect of improving the adaptability to complex environments.

[0057] Embodiment 3: This embodiment makes further improvements on the basis of the above embodiment.

[0058] The basic data includes first basic data and second basic data. The first basic data refers to the basic data whose data update frequency is not greater than a preset first frequency threshold, and the second basic data is the identification data that meets the preset conditions.

[0059] The preset conditions are set as follows: a monitoring period is preset, all the identification data transmitted within the monitoring period is obtained, the transmission frequency of different identification data is calculated, a second frequency threshold is preset, and the identification data corresponding to the transmission frequency greater than the second frequency threshold is marked as the second basic data.

[0060] The multi-level identification code includes a head and a tail. The head corresponds to the first basic data, and the tail corresponds to the second basic data.

[0061] In some embodiments, the data update frequency refers to the number of times a certain data item changes within a unit time. For the first basic data, its update frequency is low and remains stable; the transmission frequency refers to the number of times a certain identification data is transmitted within the monitoring period, which is used to measure the activity degree of the identification data in data sharing. For each identification data, the total number of times it is transmitted within the monitoring period is counted, and the transmission frequency is obtained by dividing the number of transmission times by the length of the monitoring period (such as the number of days or weeks).

[0062] In some embodiments, both the first frequency threshold and the second frequency threshold are preset according to the actual situation and historical data. For example, the setting of the first frequency threshold: According to the business characteristics of warehousing and logistics management, analyze the historical update records of various types of basic data. For data items that are long-term stable or rarely change, set a lower first frequency threshold (such as no more than 1 update per month); for data items that need to be updated occasionally, set a slightly higher first frequency threshold (such as no more than 1 update per quarter). The setting of the second frequency threshold: Preset a monitoring period, such as one month or one quarter, collect the transmission records of all identification data within the monitoring period, analyze the transmission records, calculate the transmission frequency of each identification data, and set the second frequency threshold according to business requirements and data analysis results. For example, for a customer ID, if a certain customer appears much more frequently than other customers within the monitoring period and has an important impact on the business, its transmission frequency can be set as the second frequency threshold. Specific adjustments need to be made according to the actual situation, and this application does not make specific restrictions here.

[0063] In some embodiments, when calling a data set for identification and judgment, first determine whether there is first basic data. If it exists, continue to search to see if there is second basic data. If both exist, generate a multi-level identification code and share the multi-level identification code and the identification data; the head of the multi-level identification code is used to uniquely identify the first basic data, and the tail is used to identify the second basic data or the category / attribute of specific identification data, so as to achieve efficient transmission and accurate search during data sharing.

[0064] For example, for order information: Use the order type description, order source category, etc. as the first basic data, and promote frequently occurring specific order numbers or customer IDs (such as exceeding the threshold) to the second basic data; for transportation capacity information: Use the vehicle type description, personnel position description as the first basic data, and promote commonly used vehicle numbers or personnel IDs (such as exceeding the threshold) to the second basic data; for transportation monitoring: Use the transportation status description as the first basic data, and promote the identifier of a specific transportation route (such as exceeding the threshold) to the second basic data.

[0065] Suppose in logistics warehousing management, there are the following data items: Order information: Order type description (such as "ordinary order"), order source category (such as "e-commerce platform"), order number, customer ID, specific order time. Transportation capacity information: Vehicle type description (such as "truck"), personnel position description (such as "driver"), vehicle number, personnel ID, current status of personnel.

[0066] The initial classification is as follows:

[0067] First basic data: Order type description, order source category, vehicle type description, personnel position description.

[0068] Identified data: order number, customer ID, specific order time, vehicle number, personnel ID, current status of personnel.

[0069] After statistical analysis of subsequent identified data, it is found that orders of a certain customer ID (such as "Customer A") appear frequently, exceeding the first threshold; it is found that a certain vehicle number (such as "Vehicle 123") appears frequently in the transport capacity information, exceeding the first threshold.

[0070] Promote "Customer A" to the second basic data (customer category); promote "Vehicle 123" to the second basic data (vehicle category).

[0071] The head of the multi-level identification code for the order type description "ordinary order" is "01"; the tail of the multi-level identification code for the customer category "Customer A" is "A01".

[0072] The head of the multi-level identification code for the vehicle type description "truck" is "V01"; the tail of the multi-level identification code for the vehicle category "Vehicle 123" is "V123".

[0073] Sender transmission: multi-level identification code "01A01" (ordinary order - Customer A), specific order time, other relevant identified data.

[0074] Receiver parsing: Search for the first basic data "ordinary order" according to "01", and search for or confirm the second basic data "Customer A" according to "A01".

[0075] Combination restoration: Combine "ordinary order", "Customer A" with the specific order time and other identified data to form complete order information.

[0076] The specific implementation process is as follows: Collect the historical shared data set, classify the data, initially divide it into the first basic data and identified data, determine the first basic data items according to the data update frequency, set a unique multi-level identification code head for it, establish a basic database, and store the first basic data and its corresponding multi-level identification code head. Set a monitoring period, such as one month. During the monitoring period, record the transmission situation of all identified data, calculate the transmission frequency of each identified data, and the method is the same as that described in the above design scheme. Set the second frequency threshold, mark the identified data items whose transmission frequency exceeds this threshold as the second basic data, set the multi-level identification code tail for the second basic data, and combine it with the corresponding head to form a complete multi-level identification code.

[0077] When sharing data, the sender converts the basic data into multi-level identification codes according to the basic database, and only transmits the multi-level identification codes and the remaining identification data. After receiving the data, the receiver first parses the multi-level identification codes, looks up the corresponding first basic data in the basic database according to the header. If the multi-level identification code contains a tail, the second basic data is looked up or confirmed according to the tail. The first basic data, the second basic data (if any) and the identification data are combined to restore the complete shared data set.

[0078] The basic database is updated dynamically. Regularly (such as every quarter) or as needed, the basic database is reviewed and adjusted. According to the transmission frequency data in the new monitoring period, the second basic data items and the multi-level identification code system are updated to ensure the efficiency and accuracy of data sharing.

[0079] The technical solutions in the embodiments of the present application at least have the following technical effects or advantages:

[0080] In the present application, by dividing the basic data into the first basic data and the second basic data, the data volume of the identification data is further reduced, and the effects of further improving the accuracy and efficiency of data sharing are achieved.

[0081] Embodiment 4: This embodiment makes further improvements on the basis of the above embodiment.

[0082] In step S200, the warehousing management decision result includes the current warehousing task decision result and the warehousing task prediction result for the next monitoring period. The current warehousing task decision result is obtained based on the shared data set; obtaining the warehousing task prediction result for the next monitoring period includes: obtaining the second basic data in multiple monitoring periods and marking it as analysis data, identifying the dynamic change factors in the warehousing logistics, and based on the analysis data and the analysis result of the dynamic change factors, constructing a prediction model, and then generating the warehousing task prediction result for the next monitoring period.

[0083] In some embodiments, obtaining the second basic data in multiple monitoring periods and marking it as analysis data, the length of the obtained data needs to be set according to the actual situation, which can be 5 monitoring periods, 10 monitoring periods, etc. The present application does not make specific limitations here. The second basic data is selected from a large amount of identification data and represents those data items that appear frequently and have important impacts on the business. Therefore, using the second basic data as the analysis data can greatly reduce the data volume and improve the efficiency of data processing and analysis.

[0084] Since the second basic data contains those factors that have important impacts on the warehousing logistics tasks and change frequently, using these data as the analysis data of the prediction model can more accurately capture the dynamic changes in the warehousing logistics, thereby improving the accuracy of the prediction model.

[0085] The selection of the second basic data is based on the actual data transmission frequency and business requirements, so it has strong adaptability. As the business develops and the data changes, the second basic data can be updated by adjusting the frequency threshold to ensure that the prediction model always makes predictions based on the most relevant and important data.

[0086] As the business develops and the data changes, the second basic data can be updated by adjusting the frequency threshold to ensure that the prediction model always makes predictions based on the most relevant and important data, thereby continuously optimizing the warehousing management decision-making process.

[0087] In some embodiments, dynamic change factors in warehousing logistics are identified, including: using association rule mining technology to perform data mining on the analysis data, converting the analysis data into numerical data; generating frequent item sets in a layer-by-layer search manner, setting a minimum support and a minimum confidence, and constructing an association rule mining model; inputting the numerical data into the association rule mining model to obtain frequent item sets and association rules, and then identifying the dynamic change factors in warehousing logistics and the relationships between the dynamic change factors; specifically, the dynamic change factors include but are not limited to: transportation vehicle scheduling, cargo scheduling, and transportation date.

[0088] The transportation vehicle scheduling data includes the departure time, arrival time, driving route, and cargo capacity of the vehicle, etc., the cargo scheduling data includes the warehousing time, outbound time, storage location, cargo type, and cargo demand of the cargo, etc., and the transportation date refers to the pre-agreed arrival date of the logistics.

[0089] Specifically, the minimum support represents the minimum frequency of an item set appearing in the dataset. The lower the support, the more frequent item sets are generated, but they may contain noisy data; the minimum confidence represents the credibility of an association rule. The higher the confidence, the more reliable the rule.

[0090] In some embodiments, an association rule mining algorithm needs to be selected to construct an association rule mining model. Here, the Apriori algorithm that generates frequent item sets in a layer-by-layer search manner is selected. The minimum support is initially set to 0.2, and the minimum confidence is set to 0.7. Based on the initial parameters and the association rule mining algorithm, an association rule mining model is constructed, the analysis data is input, the corresponding frequent item sets and association rules are obtained, the model results are evaluated, valuable association rules are screened out, and then the dynamic change factors in warehousing logistics and the relationships between the dynamic change factors are identified.

[0091] In some embodiments, a prediction model is constructed based on the analysis data and the analysis results of the dynamic change factors, including: preprocessing the analysis data to obtain the processed analysis data, dividing the processed analysis data into a training set and a validation set, using a time series model as the basic model, training the time series model with the training set to obtain an initial model, and validating and optimizing the initial model based on the validation set to obtain a prediction model; using the prediction model to predict the dynamic change factors in the next monitoring period, and then obtaining the prediction result of the warehousing task in the next monitoring period. According to the prediction result, adjust the transportation vehicle scheduling and cargo scheduling plans to cope with the possible impact of date restrictions.

[0092] For example, a logistics company needs to predict the transportation vehicle scheduling volume in the next monitoring period (one month) to reasonably arrange vehicles and personnel. Collect the second basic data of the past 12 monitoring periods (one year), identify the dynamic change factors, and take the transportation vehicle scheduling data as an example to obtain the daily vehicle scheduling volume. Preprocess the data to remove outliers and missing values and perform normalization processing. Use the prediction model for analysis to obtain whether there is obvious autocorrelation and seasonality. According to the analysis results, the transportation vehicle scheduling volume in the next monitoring period will show a trend of increasing first and then decreasing, and it is necessary to increase the vehicle and personnel arrangements during the peak period. According to the prediction result, formulate a transportation vehicle scheduling plan to ensure that there are enough vehicles and personnel during the peak period. Monitor the difference between the actual scheduling volume and the predicted volume in real time, and adjust the plan in time to cope with emergencies. Regularly evaluate the model prediction effect and iteratively optimize the model according to actual needs.

[0093] In some embodiments, a prediction model is constructed. Specifically, the second basic data within multiple past monitoring cycles (such as the past year, with a total of 12 monitoring cycles) is collected, and this data should cover key information such as transportation vehicle scheduling, cargo scheduling, and transportation dates; outliers (such as unreasonable departure times, arrival times, cargo demand quantities, etc.) and missing values are removed. For missing values, methods such as interpolation, forward filling, and backward filling can be used for processing, and non-numerical data (such as cargo types) is converted into numerical data. For example, the cargo type can be converted into a binary vector using one-hot encoding, and the numerical data is normalized so that the data is at the same magnitude, facilitating model training. According to the results of association rule mining, dynamic change factors such as transportation vehicle scheduling, cargo scheduling, and transportation dates are identified. The preprocessed analysis data is divided into a training set and a validation set. The training set is used to train the model, and the validation set is used to evaluate the model performance. According to the characteristics of the data and the prediction requirements, a suitable time series model is selected, such as the ARIMA model, SARIMA model, Holt-Winters model, etc. The selected time series model is trained using the training set. During the training process, the parameters of the model need to be adjusted. The trained model is validated using the validation set to evaluate the prediction accuracy and generalization ability of the model.

[0094] In some embodiments, the trained time series model is used to predict the dynamic change factors within the next monitoring cycle, such as the transportation vehicle scheduling volume, cargo scheduling volume, etc. According to the prediction results, reasonable transportation vehicle scheduling and cargo scheduling plans are formulated. For example, if the prediction results show that the transportation vehicle scheduling volume within the next monitoring cycle will show a trend of increasing first and then decreasing, then vehicle and personnel arrangements can be increased during the peak period to cope with possible transportation pressure. The difference between the actual scheduling volume and the predicted volume is monitored in real time, and the plan is adjusted in a timely manner to cope with emergencies. For example, if the actual scheduling volume suddenly increases, it may be necessary to temporarily allocate vehicles and personnel to meet the transportation demand. The prediction effect of the model is evaluated regularly, and the model is iteratively optimized according to actual needs. As new data is continuously added, the model can be retrained to improve the prediction accuracy and generalization ability.

[0095] The technical solutions in the embodiments of the present application at least have the following technical effects or advantages:

[0096] By obtaining the second basic data within multiple monitoring cycles, combining association rule mining and a prediction model, the present application can more accurately capture the dynamic changes in warehousing logistics, thereby improving the prediction accuracy of warehousing tasks within the next monitoring cycle; by using the second basic data as the data source for prediction analysis, efficient data processing and accurate prediction are achieved, while the efficiency and accuracy of warehousing management decisions are improved, and the adaptability and scalability of the prediction model are enhanced.

[0097] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, various modifications and variations can be made to the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent logistics warehousing management method, characterized in that: include: S100: Acquire an original data set, perform a shared information quantitative analysis on the original data set to obtain an information quantification index, and perform a preliminary filtering on the original data set according to the information quantification index to obtain a calling data set; S200: Generate a basic database, identify the basic data and the corresponding multi-level identification code in the calling data set according to the basic database, mark the remaining data in the calling data set as identification data, and form a shared data set with the multi-level identification code and the identification data; share data according to the shared data set to generate a warehouse management decision result.

2. The intelligent logistics warehousing management method according to claim 1, characterized in that: The step S100 also includes: S110: capturing each data item in the data stream in real time to form multiple sub-data sets, and then fusing them to form an original data set, and using stream processing technology to perform shared information quantitative analysis on the original data set to form a processed original data set; S120: extracting multi-dimensional data features from the processed original data set and analyzing to obtain corresponding eigenvalues, and performing weighted summation on the eigenvalues ​​to obtain an information quantization index; S130: presetting a quantization index threshold, and marking data corresponding to information quantization indexes less than the quantization index threshold as callable data; Filter the data corresponding to the information quantization index not less than the quantization index threshold, and recalculate the information quantization index of the filtered data. If it is less than the quantization index threshold, mark it as callable data; if it is not less than the quantization index threshold, discard it; S130: Acquire all callable data to form a call data set.

3. An intelligent logistics warehousing management method as claimed in claim 2, characterized in that: The sub-datasets include: an original sub-dataset of order information, an original sub-dataset of capacity information, and an original sub-dataset of transportation monitoring; the multi-dimensional data features include data real-time, data integrity, data importance, and data transmission.

4. The intelligent logistics warehousing management method according to claim 1, characterized in that: In step S200, a basic database is generated, including: obtaining a historical shared data set, classifying the data in the historical shared data set into basic data and identification data; setting a unique multi-level identification code for each basic data, storing the basic data and the unique multi-level identification code in a database to form a basic database; wherein the correspondence between the basic data and the identification data is one-to-many.

5. The intelligent logistics warehousing management method according to claim 1, characterized in that: The basic data includes first basic data and second basic data, the first basic data refers to basic data whose data update frequency is not greater than a preset first frequency threshold, and the second basic data is identification data that meets preset conditions; the multi-level identification code includes a header and a tail, the header corresponds to the first basic data, and the tail corresponds to the second basic data.

6. An intelligent logistics warehousing management method as claimed in claim 5, characterized in that: The preset conditions are set as: presetting a monitoring period, obtaining all identification data transmitted within the monitoring period, calculating the transmission frequency of different identification data, presetting a second frequency threshold, and marking the identification data corresponding to the transmission frequency greater than the second frequency threshold as second basic data.

7. The intelligent logistics warehousing management method according to claim 1, characterized in that: The warehouse management decision result includes the current warehouse task decision result and the warehouse task prediction result of the next monitoring cycle, and the current warehouse task decision result is obtained based on the shared data set; The second basic data within multiple monitoring cycles is obtained and marked as analysis data, the dynamic change factors in warehousing logistics are identified, and a prediction model is constructed based on the analysis data and the analysis results of the dynamic change factors, thereby generating a warehousing task prediction result for the next monitoring cycle.

8. An intelligent logistics warehousing management method as claimed in claim 7, characterized in that: Identify dynamic factors in warehouse logistics, including: Use association rule mining technology to mine the analysis data and convert the analysis data into numerical data; use layer-by-layer search to generate frequent item sets, set the minimum support and minimum confidence, and build an association rule mining model; Input numerical data into the association rule mining model to obtain frequent item sets and association rules, and then identify the dynamic change factors in warehousing logistics and the relationship between them; The dynamically changing factors include transport vehicle scheduling, cargo scheduling and transport date.

9. The intelligent logistics warehousing management method according to claim 7, characterized in that: Based on the analysis data and the analysis results of the dynamic change factors, a prediction model is constructed, including: The analytical data is preprocessed to obtain the processed analytical data, and the processed analytical data is divided into a training set and a validation set. The time series model is used as the basic model, and the training set is used to train the time series model to obtain the initial model. The initial model is validated and optimized based on the validation set to obtain the prediction model; the prediction model is used to predict the dynamic change factors in the next monitoring cycle, and then the warehousing task prediction results of the next monitoring cycle are obtained.

10. The intelligent logistics warehousing management method according to claim 8, characterized in that: The transport vehicle scheduling includes the vehicle's departure time, arrival time, driving route and cargo capacity, etc. The cargo scheduling includes the cargo entry time, exit time, storage location, cargo type and cargo demand, etc. The transportation date refers to the arrival date agreed in advance by the logistics department.

Images