A book intelligent allocation system and method based on an intelligent stereoscopic book library

Abstract

The present application belongs to the technical field of library information management, and specifically relates to a book intelligent allocation system and method based on an intelligent stereoscopic book library. The method collects reader borrowing and search behavior data, combines branch library geographic information and the Chinese Library Classification system, and analyzes regional reading demand differences. Based on preset removal rules, low circulation rate books are automatically screened to generate a removal list. At the same time, multi-source behavior data is weighted and calculated, and the book classification is matched with the intelligent stereoscopic book library inventory to automatically generate a precise shelving list. Finally, the allocation is executed through the interface of the automated warehousing and logistics system. The present application realizes the transition from experience-driven to data-driven, solves the problems of uneven allocation of book resources and slow allocation, and significantly improves the book circulation efficiency, space utilization rate and reader service satisfaction.

Description

Technical Field

[0001] This invention relates to the field of library information management technology, and in particular to a smart book allocation system and method based on an intelligent three-dimensional book repository. Background Technology

[0002] With the development of urban public library service systems, large libraries often have dozens or even hundreds of branches, forming a service network covering the entire city. However, traditional book allocation models mainly rely on manual experience or static borrowing data statistics, which has the following problems: (1) Strong lag in allocation: The decision-making cycle for book removal and removal is long, making it difficult to respond to changes in readers' needs; (2) Uneven resource allocation: Popular books are concentrated in a few central libraries, while remote branch libraries have few borrowable resources; (3) Lack of behavioral insights: The multi-dimensional behavioral data of readers such as searching, ordering, and borrowing are not fully explored; (4) Low space utilization efficiency: A large number of books with low circulation rates occupy limited bookshelf space for a long time; (5) Lack of intelligent hub: There is a lack of a unified data-driven platform combined with automated warehousing (such as intelligent three-dimensional library) to achieve efficient book circulation.

[0003] Some libraries have begun to introduce "intelligent automated book storage" systems as central storage nodes, which have functions such as high-density storage, automatic access, and RFID identification. However, most existing systems are only used for book storage and outbound operations and have not yet been deeply integrated into the book allocation decision-making process.

[0004] Therefore, there is an urgent need to build a smart book allocation system that integrates geographic information analysis, reader behavior modeling, the Chinese Library Classification system, and an intelligent three-dimensional book warehouse linkage mechanism, in order to achieve a transformation from "passive response" to "proactive prediction + precise delivery". Summary of the Invention

[0005] This invention proposes a smart book allocation system and method based on an intelligent three-dimensional library, aiming to solve the technical problems in the existing technology that traditional book allocation relies on manual experience and static data, resulting in allocation delays, uneven resource allocation, lack of reader behavior insights, and low utilization of bookshelf space.

[0006] In a first aspect, the present invention provides a smart book allocation system based on an intelligent three-dimensional library, comprising: The reader behavior analysis module is used to collect and analyze reader borrowing and search data from multiple branches; The collection status analysis module is used to analyze the circulation status and attributes of documents in each branch library; The allocation decision engine, connected to the reader behavior analysis module and the collection status analysis module respectively, is used to: generate a list of documents to be allocated and removed from shelves based on preset removal rules and the output of the collection status analysis module; and generate a list of documents to be allocated and put back on shelves based on a multi-source data weighted model and the output of the reader behavior analysis module, combined with the inventory information of the intelligent three-dimensional library.

[0007] The technical effect of the intelligent book allocation system based on intelligent three-dimensional library disclosed in this invention is that it significantly improves the matching degree between books and readers' needs by driving decision-making based on real and multi-dimensional reader behavior data.

[0008] Furthermore, the reader behavior analysis module is also used to generate a visual behavior map containing borrowing heat distribution based on the branch library's geographical information and reader behavior data.

[0009] Furthermore, the pre-defined delisting rules in the allocation decision engine include at least the following combinations of conditions: The number of copies of the document is greater than or equal to the first threshold, and the recent borrowing frequency is lower than the second threshold; The total number of times the document was borrowed was zero, and the number of years it had been held in the collection exceeded the third threshold; The literature is not a recommended reading list, and its recent borrowing frequency is higher than the fourth threshold; The literature is a recommended reading list, and its borrowing frequency within a specific period is lower than the fifth threshold.

[0010] Furthermore, the multi-source data weighting model in the allocation decision engine is: N = T × (w1·R1 + w2·R2 + ... + w n ·R n ); where N is the final recommended result of the documents to be allocated, T is the planned total allocation, and w1 to w n The weights for each data source are R1 to R2. n These are literature demand indicators obtained from analyses of different data sources.

[0011] Furthermore, the data sources include at least two of the following: local historical borrowing volume of branch libraries, online borrowing volume of branch libraries, online platform document search popularity, and central library borrowing ranking.

[0012] Furthermore, the system also includes a book classification mapping module, which is used to classify and index document and reader behavior data according to the Chinese Library Classification system.

[0013] Secondly, the present invention provides a method for intelligent book allocation based on an intelligent three-dimensional library, comprising the following steps: Reader behavior data analysis steps: Collect and analyze borrowing and search behavior data of readers in each branch library; Steps for analyzing the circulation status of collections: Analyze the borrowing history and attribute information of documents in each branch library; Allocation decision generation steps: Based on the preset removal rules and the analysis results of the circulation status of the collection, a removal list is generated; Based on the results of multi-source data weighting model and reader behavior data analysis, and combined with the intelligent three-dimensional book warehouse inventory, a list of books to be put on the shelves is generated; Intelligent allocation execution steps: The generated list of books to be put on or taken off the shelves is sent to the intelligent automated book warehouse to drive the automated allocation operation.

[0014] The technical advantages of the intelligent book allocation method based on an intelligent three-dimensional library disclosed in this invention are: constructing a closed loop of "data analysis - intelligent warehousing - logistics distribution," realizing a transformation from passive response to proactive prediction. The modular design facilitates the integration of new data sources such as social media sentiment and seasonal factors, enabling continuous model optimization.

[0015] Furthermore, the sub-steps for generating the shelving list include: assigning weights to different data sources; allocating the planned total allocation to each data source based on the weights; calculating the demand ratio and allocating the quantity for each data source according to the document classification; and determining the final shelving list by combining the allocation results of each data source with the inventory status.

[0016] Furthermore, in the reader behavior data analysis step, a geographical distance algorithm is used to assign readers' online behavior data to specific branch libraries for statistical analysis.

[0017] Furthermore, the method also includes a decision feedback optimization step: dynamically adjusting the threshold in the removal rule or the weight coefficient in the multi-source data weighting model based on the actual borrowing feedback of the target branch library on the newly added documents after the allocation is executed. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the overall system workflow of an embodiment of the present invention. Figure 2 This is a flowchart illustrating the logic for generating the delisting list in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the logic for generating the allocation and shelving list in an embodiment of the present invention; Figure 4 This is a schematic diagram of the system architecture in a specific application scenario of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Addressing the issue mentioned in the background that most existing systems are only used for book storage and retrieval operations, and have not been deeply integrated into the book allocation decision-making process, this invention proposes a smart book allocation system and method based on an intelligent automated book warehouse. This aims to solve the technical problems of traditional book allocation relying on manual experience and static data, resulting in allocation delays, uneven resource allocation, lack of reader behavior insights, and low utilization of bookshelf space. (Reference) Figures 1 to 4 As shown, it specifically includes: Branch configuration module: Manages basic information of branch libraries and obtains latitude and longitude coordinates through map API to establish a geographic database.

[0021] Administrators can input or import branch information (name, code, street, detailed address) through the backend interface. The system calls APIs such as Baidu Maps to parse the address into latitude and longitude coordinates (e.g., "longitude: 116.403, latitude: 39.915") and store it in the spatial database. When branch information changes, the system updates synchronously.

[0022] A geographic information database for the branch libraries was established, laying the foundation for subsequent location-based behavioral analysis and logistics route planning.

[0023] Book classification configuration module: Built-in classification tree of the Chinese Library Classification (CLC) up to the third level, used for book indexing and behavioral analysis classification.

[0024] The system comes pre-configured with the complete classification tree structure of the 5th edition of the Chinese Library Classification (CLC). For example, under "TP3 Computing Technology, Computer Technology," there are third-level categories such as "TP31 Computer Software" and "TP311 Programming, Software Engineering." Administrators can maintain the classification relationships through the interface and bind the ISBN metadata of books to the CLC classification number.

[0025] This module provides a unified standard for book content, making it possible to analyze borrowing preferences for books on different themes.

[0026] Reader behavior data collection module: Connects to the library's business system to collect borrowing orders, search logs, reservation records, etc. in real time.

[0027] Data is extracted daily from sources such as the Integrated Library System (ILS), online lending platform, and mobile app using ETL tools or API interfaces. Collected fields include: reader ID (anonymized), book ISBN, operation type (borrowing / search / reservation), timestamp, associated branch library code (or IP address), and search keywords.

[0028] This module enables the centralized aggregation of heterogeneous reader behavior data from multiple sources, forming the foundation for data-driven decision-making.

[0029] The reader behavior analysis module includes functions such as branch-level search analysis, reader group preference analysis, and branch-level and city-wide behavior maps. It utilizes the Manhattan distance algorithm to classify order attribution and provides multi-dimensional data visualization on a GIS map. Specifically, it includes: Branch-level search analysis unit: Statistics on the search records of readers belonging to Branch A over the past month, matching search terms to the China Library Classification using natural language processing, revealing that readers of this branch had the highest search volume for "I Literature" category books.

[0030] Reader behavior distribution unit: By grouping readers by age (e.g., teenagers, adults, seniors) and occupation (students, employees), the analysis revealed that the youth group had a significant increase in borrowing of books in the "G Culture, Science, Education, Sports" category in Branch Library B.

[0031] Behavioral map unit (branch level): such as Figure 4 As shown, all branches are marked with dot icons on the electronic map. Clicking on branch C brings up an information box displaying: total borrowings in the past year were 5200 books, with an average of 2.1 books borrowed per person, and the most popular category was "F Economy". Order attribution uses the Manhattan distance algorithm: for an online order, the Manhattan distance between its delivery address and the coordinates of all surrounding branches is calculated, and the order is assigned to the nearest branch for statistical purposes.

[0032] Behavioral Map Unit (Citywide Level): Data is aggregated by administrative region (such as district or county) to generate a heat map, showing that the eastern part of the city has a strong demand for science and technology books, while the western part of the city has a clear preference for history books.

[0033] This module provides multi-dimensional insights into readers' needs from both micro and macro perspectives, transforming abstract data into intuitive geographic visualizations to help managers grasp overall trends.

[0034] Collection Borrowing Analysis Module: Calculates book lifecycle indicators (such as total number of borrowings, most recent borrowing time, years since acquisition, number of copies, and whether it is recommended in the bibliography) to assess its circulation value.

[0035] The system regularly imports the catalogs of each branch library and the borrowing and returning records for each book. For each book, the system calculates and updates the following metrics: total historical borrowing counts, most recent borrowing date, acquisition date (used to calculate acquisition years), number of copies in the current branch library, and a Boolean label "Whether it is a Selected Books" (from the acquisition system). For example, the system determines that the book "XX Programming" has 3 copies in branch library D, with a total of 15 borrowings, but the most recent borrowing was 2 years ago, and it has been in the collection for 8 years, therefore it is not a Selected Books book.

[0036] This module quantifies the circulation value and status of each book, providing an objective and quantitative basis for the decision to remove it from shelves.

[0037] The module for generating the list of books to be removed from shelves applies multi-condition combination rules to automatically identify and generate a list of books to be removed from shelves.

[0038] The system iterates through the collection list of branch library E, applying the following rules for logical judgment: Rule 1: Check the book "One Hundred Years of Solitude". The number of copies is 2. The system checks its borrowing records for the past 3 years. The number of times it was borrowed is 3 (<5). The rule is matched and it is marked as "recommended to remove from the shelves".

[0039] Rule 2: Check the book "Acta Physica Sinica 1985". The total number of borrowings is 0. The system calculates its storage period as 12 years (≥10 years). The rule is matched and it is marked as "recommended to be removed from the shelves".

[0040] Rule 3: Check the book "Old Edition Tax Guide". The total number of borrowings is 10, and the number of borrowings in the past 3 years is 4 (≥3). The system checks that its "Book Selection Helper" label is "No". If it hits the rule, it will be marked as "Recommended to remove from the shelves".

[0041] Rule 4: Check the book "Reader's Recommended Poetry Collection" (Selected Books). The system checks its borrowing count in the past 5 years and finds it is 0 (<1). This matches the rule and is marked as "Recommended to be Removed from Shelf". For all marked books, the system automatically generates a structured "Branch Library E-Book Transfer and Removal List.xlsx" containing the ISBN, book title, current location, and reason for removal.

[0042] This module automates and batches the identification of books with low circulation rates, freeing up valuable bookshelf space. The rules are clear and explicit, reducing human subjectivity.

[0043] The allocation and shelving list generation module automatically generates a recommended list of books to be shelved, based on a multi-source data weighting, classification calculation, and inventory matching process.

[0044] The process of generating a shelving list for branch library F is as follows: S71: The administrator sets the total number of books to be allocated in this transfer plan to T=100.

[0045] S72 / S73: The system uses fixed weights: local borrowing volume of branch libraries (w1=2), online borrowing + database borrowing volume (w2=1), search popularity (w3=1), and online borrowing database ranking (w4=1). The allocated quantity for each data source is calculated as follows: N1=100*2 / (2+1+1+1)=40 books; N2=N3=N4=100*1 / 5=20 books.

[0046] S74: For the 40-book quota for "local borrowing from branch libraries," the system calculates the Chinese Library Classification (CLC) of all local borrowing records from branch library F over the past six months. Assume the statistics are: Category I (Literature) 30%, Category TP (Automation Technology) 25%, Category K (History & Geography) 20%, and others 25%. Therefore, the allocation for Category I under this data source is: 40 books * 30% = 12 books. Similarly, calculate 10 books for Category TP, 8 books for Category K, and so on.

[0047] Repeat step S74 for the three data sources: "Online Borrowing + Library Borrowing Quantity" (20 books), "Search Popularity" (20 books), and "Online Borrowing Library Ranking" (20 books) to obtain the suggested quantity for each category based on the different data sources.

[0048] S75: Summarize the suggested number of "Category I" books from the four data sources. For example: (12+5+6+7)=30 books. This means the final suggestion is for branch library F to stock 30 Category I books. The same logic applies to obtaining the numbers for other categories.

[0049] Inventory matching and deduplication: The system queries the intelligent automated library for available inventory and finds 50 titles in Category I with sufficient stock. After excluding copies already available in branch library F, the system selects 30 titles from these 50 titles based on a comprehensive score (e.g., the city's borrowing index over the past six months * 1.0 + search popularity * 0.8) from highest to lowest.

[0050] Out-of-stock replacement: If the required "K-category history and geography" books are in short supply, the system will activate a topic similarity algorithm (such as cosine similarity calculation based on the keywords of the book's metadata) to select the most similar books from the category with sufficient stock for replacement.

[0051] Finally, a "Branch F Book Allocation and Shelving List" is generated, which includes book details, allocation quantity, source warehouse, and estimated delivery time.

[0052] This module integrates multiple demand signals, intelligently quantifies the demand intensity of various types of books, and combines inventory to achieve accurate recommendations, ensuring that the books put on the shelves are the ones most likely to be borrowed at the moment.

[0053] Intelligent automated book storage docking module: Enables command interaction with automated storage and retrieval systems (WMS / WCS).

[0054] The shelving and de-shelf lists are converted into standard outbound instruction messages via the intelligent automated library interface module and sent to the automated library's Warehouse Control System (WCS). The WCS dispatches stacker cranes to retrieve the corresponding books from the high-density shelves and transport them to the packing station. After packing, the logistics information is synchronized to the allocation task management module and assigned to the logistics carrier. After the books are delivered to the branch library and signed for, the administrator scans the code to confirm receipt, and the system automatically updates the book's location status to "on the shelf at Branch Library F".

[0055] This module enables seamless integration of information flow and logistics, allowing data-driven decisions to be executed efficiently and accurately by automated equipment, forming a complete closed loop.

[0056] Allocation Task Management Module: Tracks allocation tasks throughout the entire process, manages logistics and feedback.

[0057] Based on the same inventive concept, embodiments of the present invention provide a method for intelligent book allocation based on an intelligent three-dimensional library, comprising the following steps: Reader behavior data analysis steps: Collect and analyze borrowing and search behavior data of readers in each branch library.

[0058] The steps for analyzing the circulation status of the collection are as follows: Analyze the borrowing history and attribute information of the documents in each branch library; the system runs continuously, collecting the previous day's behavior and borrowing data daily, and updating the reader behavior analysis model and collection analysis indicators.

[0059] Allocation decision generation steps: Based on the preset removal rules and the analysis results of the circulation status of the collection, a removal list is generated; Based on the multi-source data weighted model and reader behavior data analysis results, combined with the intelligent automated book warehouse inventory, a shelf-reading list is generated; a relocation decision task is triggered monthly (or quarterly). The system executes the de-shelf list generation process in parallel. Figure 2 ) and the process of generating the shelf list ( Figure 3 The system merges the lists, triggers the intelligent automated book warehouse outbound operation, and tracks the entire logistics process.

[0060] Intelligent allocation execution steps: The generated shelving and de-shelving list instructions are sent to the intelligent automated book storage system to drive the automated allocation operation. One month after the allocation is completed, the system analyzes the actual borrowing rate of the newly added books in branch library F and compares it with the predicted value. If the actual borrowing rate of a certain type of book (such as "O Mathematical Sciences") remains low, the system can automatically suggest lowering its influence factor in the weighting formula, or reducing the proportion of this type of book in the next list generation, thus achieving self-iterative optimization of the model.

[0061] Beneficial effects: Precision service: Decision-making is driven by real, multi-dimensional reader behavior data, which greatly improves the matching degree between books and readers' needs.

[0062] Intelligent decision-making: Construct a closed loop of "data analysis - intelligent warehousing - logistics distribution" to achieve a transformation from passive response to proactive prediction.

[0063] Space efficiency: Regularly remove inefficient books to free up physical space for high-demand books and improve the utilization rate of branch library space.

[0064] Resource equalization: Through geographical and behavioral analysis, we promote the flow of high-quality resources to peripheral branches and facilitate the equalization of services.

[0065] Green operations: Extending the book lifecycle, reducing unnecessary purchases and copies, and aligning with the concept of sustainable development.

[0066] System scalability: The modular design facilitates the integration of new data sources such as social media sentiment and seasonal factors, enabling continuous model optimization.

[0067] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A smart book allocation system based on an intelligent three-dimensional library, characterized in that, include: The reader behavior analysis module is used to collect and analyze reader borrowing and search data from multiple branches; The collection status analysis module is used to analyze the circulation status and attributes of documents in each branch library; The allocation decision engine, connected to the reader behavior analysis module and the collection status analysis module respectively, is used to: generate a list of documents to be allocated and removed from shelves based on preset removal rules and the output of the collection status analysis module; and generate a list of documents to be allocated and put back on shelves based on a multi-source data weighted model and the output of the reader behavior analysis module, combined with the inventory information of the intelligent three-dimensional library.

2. The system according to claim 1, characterized in that, The reader behavior analysis module is also used to generate a visual behavior map containing borrowing heat distribution based on the branch library's geographical information and reader behavior data.

3. The system according to claim 1, characterized in that, The pre-set delisting rules in the allocation decision engine include at least the following combinations of conditions: The number of copies of the document is greater than or equal to the first threshold, and the recent borrowing frequency is lower than the second threshold; The total number of times the document was borrowed was zero, and the number of years it had been held in the collection exceeded the third threshold; The literature is not a recommended reading list, and its recent borrowing frequency is higher than the fourth threshold; The literature is a recommended reading list, and its borrowing frequency within a specific period is lower than the fifth threshold.

4. The system according to claim 1, characterized in that, The multi-source data weighting model in the allocation decision engine is: N = T × (w1·R1 + w2·R2 + ... + w n ·R n ); where N is the final recommended result of the documents to be allocated, T is the planned total allocation, and w1 to w n Here are the weighting coefficients for each data source, R1 to R... n These are literature demand indicators obtained from analyses of different data sources.

5. The system according to claim 4, characterized in that, The data sources include at least two of the following: local historical borrowing volume of branch libraries, online borrowing volume of branch libraries, online platform document search popularity, and central stacks borrowing rankings.

6. The system according to claim 1, characterized in that, The system also includes a book classification mapping module, which is used to classify and index document and reader behavior data according to the Chinese Library Classification system.

7. A method for intelligent book allocation based on an intelligent three-dimensional library, characterized in that, Includes the following steps: Reader behavior data analysis steps: Collect and analyze borrowing and search behavior data of readers in each branch library; Steps for analyzing the circulation status of collections: Analyze the borrowing history and attribute information of documents in each branch library; Allocation decision generation steps: Based on the preset removal rules and the analysis results of the circulation status of the collection, a removal list is generated; Based on the results of multi-source data weighting model and reader behavior data analysis, and combined with the intelligent three-dimensional book warehouse inventory, a list of books to be put on the shelves is generated; Intelligent allocation execution steps: The generated list of books to be put on or taken off the shelves is sent to the intelligent automated book warehouse to drive the automated allocation operation.

8. The method according to claim 7, characterized in that, The sub-steps for generating the shelving list include: assigning weights to different data sources; allocating the planned total allocation to each data source based on the weights; calculating the demand ratio and allocating the quantity for each data source according to the document classification; and determining the final shelving list by combining the allocation results of each data source with the inventory status.

9. The method according to claim 7, characterized in that, In the reader behavior data analysis step, a geographical distance algorithm is used to assign readers' online behavior data to specific branch libraries for statistical analysis.

10. The method according to claim 7, characterized in that, The method also includes a decision feedback optimization step: dynamically adjusting the threshold in the removal rule or the weight coefficient in the multi-source data weighting model based on the actual borrowing feedback of the target branch library on the newly added documents after the allocation is executed.

Images