A flood-prevention emergency material allocation and reserve management method and system

Abstract

The application relates to the technical field of reserve management, in particular to a flood-prevention emergency material allocation reserve management method and system; the method comprises the following steps: collecting basic attribute data, inventory state data and historical calling data of flood-prevention emergency materials, establishing unique identification codes for various flood-prevention emergency materials, and constructing a flood-prevention emergency material information database; the storage environment of the flood-prevention emergency materials is divided into units, and a material storage structure model is constructed based on the unit division result; the storage, withdrawal and shifting operations of the flood-prevention emergency materials are automatically identified through a bar code identification mode, and the flood-prevention emergency material information database is synchronously updated at the same time when the identification is completed. The application realizes the whole-process closed-loop management of the flood-prevention emergency materials from storage, storage, allocation to tracking by constructing a unified coding system and a structured storage model, and by combining a bar code quality sensing self-adaptive identification, multi-source heterogeneous data preprocessing and a dynamic demand prediction mechanism.

Description

Technical Field

[0001] This invention relates to the field of reserve management technology, specifically to a method and system for the allocation and reserve management of flood control emergency supplies. Background Technology

[0002] In the current flood control and emergency management system, the allocation and storage of flood control emergency supplies have long faced problems such as delayed response, inaccurate inventory information, and low allocation efficiency. Traditional management models rely heavily on manual records and experience-based judgment, lacking dynamic perception and precise control over the entire lifecycle of materials. Especially in a multi-source heterogeneous data environment, key information such as basic attributes, inventory status, and historical data retrieval is difficult to integrate effectively, resulting in fragmented, low-standardization, and poor real-time performance in industrial data management. Furthermore, existing warehousing systems generally lack structured storage space models, leading to unclear material locations, errors in inbound and outbound operations, and significantly reduced identification accuracy under complex on-site conditions such as damaged barcodes and varying lighting conditions, further weakening the reliability and consistency of industrial data management. Simultaneously, material demand forecasting is mostly based on static thresholds, failing to integrate dynamic flood factors such as rainfall intensity variation rates and warning water level differences, making it difficult to support the evolution of industrial data management towards intelligence and predictability, thus hindering the scientific assessment of reserve adequacy and the improvement of forward-looking allocation capabilities. Summary of the Invention

[0003] The purpose of this invention is to address the problems existing in the background technology by proposing a method and system for the allocation, storage and management of flood control emergency materials.

[0004] The technical solution of this invention: a method for the allocation, storage, and management of flood control emergency supplies, comprising: S1. Collect basic attribute data, inventory status data and historical call data of flood control emergency supplies, establish unique identification codes for various flood control emergency supplies, and build a flood control emergency supplies information database. S2. Divide the storage environment of flood control emergency materials into units, and construct a material storage structure model based on the unit division results; S3. Automatically identify the entry, exit, and relocation of flood control emergency supplies through barcode recognition, and simultaneously update the flood control emergency supplies information database upon completion of the identification. S4. Based on the updated flood control emergency supplies information database, conduct a comprehensive assessment of the adequacy and availability of various flood control emergency supplies, and generate a supply availability assessment result and a supply allocation demand result. In the process of comprehensively assessing the adequacy of reserves, a material demand prediction model is constructed by using historical flood control emergency material call data, rainfall intensity change rate and warning water level difference to predict the consumption intensity of various flood control emergency materials within a preset time window, and to generate a predicted and corrected reserve adequacy index. S5. Based on the results of material availability assessment and material allocation demand, select flood control emergency materials that meet the allocation conditions, and combine them with the material storage structure model to analyze the allocation path and generate corresponding material transportation plans for flood control emergency material reserve and allocation management.

[0005] As a further improvement to this technical solution, in step S1, the basic attribute data, inventory status data, and historical call data of flood control emergency supplies are collected, a unique identification code is established for each type of flood control emergency supplies, and a flood control emergency supplies information database is constructed, including the following steps: S1.1 Collect basic attribute data of flood control emergency supplies. The basic attribute data shall include at least the category, specifications, production batch and expiration date of the supplies. S1.2 Collect inventory status data of flood control emergency supplies. The inventory status data shall include at least the current inventory quantity, storage warehouse, storage area and entry and exit time information. S1.3 Collect historical data on the use of flood control emergency supplies. The historical data should include at least the time and quantity of each use. S1.4 Preprocess the basic attribute data, inventory status data and historical call data of the collected flood control emergency materials; S1.5. Based on the preprocessed basic attribute data, inventory status data, and historical call data, generate a unique identification code for each type of flood control emergency supplies; S1.6 Construct a flood control emergency material information database based on unique identifier codes.

[0006] As a further improvement to this technical solution, in step S2, the storage environment for flood control emergency supplies is divided into units, and a storage structure model for the supplies is constructed based on the unit division results, including the following steps: S2.1. Collect information on the storage environment of flood control emergency supplies based on the flood control emergency supplies information database; S2.2 Based on storage environment information, the storage environment of flood control emergency materials is hierarchically divided, and the storage environment is split according to the hierarchical structure of warehouse, storage area and storage unit; S2.3. Generate a unique storage unit number for each storage unit based on the hierarchical partitioning result; S2.4. Based on the storage cell number, establish corresponding cell attribute information for each storage cell; S2.5. Establish the structural relationship between storage units based on unit attribute information; S2.6 Generate a material storage structure model for flood control emergency supplies based on the hierarchical structure of the storage environment, storage unit number, unit attribute information and structural relationships; S2.7 Map the unique identifier of each type of flood control emergency supplies to its corresponding storage unit.

[0007] As a further improvement to this technical solution, in step S3, the entry, exit, and relocation operations of flood control emergency supplies are automatically identified using barcode recognition, and the flood control emergency supplies information database is updated simultaneously upon completion of the identification. This includes the following steps: S3.1. The unique identification code of each type of flood control emergency supplies shall be processed into a barcode, and the barcode information of the flood control emergency supplies shall be collected before they are put into storage. S3.2 During the process of flood control emergency supplies entering the warehouse, the quality perception adaptive identification method of flood control emergency supplies barcodes is used to identify and verify the barcode information of flood control emergency supplies. After the identification and verification are completed, the information of material category, batch, quantity and entry time is obtained, and the identification results are updated to the flood control emergency supplies information database in real time. S3.3 During the process of material outbound, the quality perception adaptive identification method of flood control emergency material barcode is used to identify and verify the barcode information of flood control emergency material. After the identification and verification are completed, the unique identifier, quantity and outbound time of the outbound material are obtained and updated to the flood control emergency material information database simultaneously. S3.4 When moving flood control emergency supplies within the flood control emergency supplies storage unit, the quality perception adaptive identification method of the flood control emergency supplies barcode is used to identify and verify the information of the flood control emergency supplies barcode. After the identification and verification are completed, the starting unit, target unit and time information of the movement are recorded, and the storage unit mapping relationship of the supplies in the database is updated at the same time. S3.5 After the barcode information of flood control emergency supplies is identified, the information on warehousing, outbound and relocation operations will be synchronized with the flood control emergency supplies information database.

[0008] As a further improvement to this technical solution, in step S3.2, the barcode information of flood control emergency supplies is identified and verified using a quality-aware adaptive recognition method, including the following steps: S3.21. Obtain the original barcode image, perform quality parameter analysis on the original barcode image, and extract a set of quality parameters including stripe width deviation rate, stripe width uniformity index, damaged area ratio, damage dispersion index, effective black and white stripe contrast, boundary continuity index, and boundary breakage density. S3.22. Based on the set of quality parameters, an image preprocessing strategy that matches the degree of barcode damage is used to adaptively enhance the original barcode image to obtain the enhanced barcode image. S3.23. Based on the set of quality parameters and the enhanced barcode image, the barcode recognition results and corresponding confidence levels of flood control emergency supplies are obtained through a parameterized multi-strategy barcode recognition method. The barcode recognition results are then subjected to adaptive verification and error correction based on the quality parameters.

[0009] As a further improvement to this technical solution, in S3.23, a parameterized multi-strategy barcode recognition method is used, and the barcode recognition results are subjected to adaptive verification and error correction processing based on quality parameters, including the following steps: Using a set of quality parameters and an enhanced barcode image as input, a barcode recognition strategy selection function is constructed based on the stripe width deviation rate, the proportion of damaged area, the stripe width uniformity index, and the damage dispersion index. The function determines the recognizability of the enhanced barcode image and selects the recognition strategy type based on the determination result. The recognition strategy types include standard barcode decoding strategy, multi-frame fusion recognition strategy, and feature matching recognition strategy. Based on a set of quality parameters, a model for calculating recognition confidence is constructed to calculate the basic recognition confidence of the barcode recognition results. The barcode recognition results are verified in stages based on the recognition confidence level: when the recognition confidence level is higher than the first threshold 'a', the verification is passed directly; when the recognition confidence level is between the first threshold 'a' and the second threshold 'b', the barcode recognition results are checked for check bits and the consistency of the barcode encoding structure is checked; when the recognition confidence level is lower than the second threshold 'b', multiple recognition algorithms are triggered to recognize the barcodes in parallel, and the consistency of multiple barcode recognition results is analyzed.

[0010] As a further improvement to this technical solution, step S4, which generates material availability assessment results and material allocation demand results, includes the following steps: S4.1 Determine the baseline inventory and safe reserve threshold for each type of flood control emergency supplies based on basic attribute data; S4.2 Utilize the updated flood control emergency supplies information database to update the inventory quantity, current occupancy status and availability status of each storage unit in real time; S4.3. By using historical flood control emergency material dispatch data, rainfall intensity change rate and warning water level difference, construct a material demand prediction model to predict the consumption intensity of various flood control emergency materials within a preset time window, calculate the original reserve adequacy index, and perform risk correction on the original reserve adequacy index based on the predicted consumption intensity to generate the predicted and corrected reserve adequacy index. S4.4. Based on the predicted and corrected reserve adequacy index, a comprehensive score is generated for the availability status of each type of flood control emergency supplies. ; S4.5 Based on the predicted and corrected reserve adequacy index and the comprehensive score of availability, the allocation calculation module generates the allocation quantity and priority order of various flood control emergency materials, and generates the material allocation demand results.

[0011] As a further improvement to this technical solution, in step S4.3, the original reserve adequacy index is risk-corrected based on the predicted consumption intensity to generate a predicted and corrected reserve adequacy index, including the following steps: S4.31. Based on historical data on the allocation of flood control emergency supplies, extract the total allocation of various types of flood control emergency supplies in different flood control events. and the duration of the corresponding event By normalizing the total number of calls according to the duration of the corresponding events, the historical unit time consumption intensity of various flood control emergency materials was calculated. Furthermore, statistical aggregation of the consumption intensity per unit time was performed to construct a historical baseline for the consumption intensity per unit time of various flood control and emergency supplies. ; S4.32, Obtain the target area within the preset time window Rainfall intensity sequence within Based on rainfall intensity sequence Calculate the instantaneous rate of change of rainfall intensity The instantaneous rate of change The trend weights are mapped using a normalized function, and then used to establish a baseline for the intensity of historical unit-time consumption. A trend-weighted correction is performed, and a material demand forecasting model is generated to obtain the trend-corrected predicted consumption intensity. ; Predicting consumption intensity based on trend correction Calculate the original reserve adequacy index ; S4.33, Obtain the real-time water level of the target area. and corresponding warning water level threshold Calculate the difference between the real-time water level and the warning water level threshold. and the difference Mapped to warning water level safety margin parameter ; S4.34, Safety margin parameters based on warning water level The trend-corrected predicted consumption intensity is subjected to nonlinear risk amplification processing to obtain the risk-amplified predicted consumption intensity. ; S4.35, Predicted Consumption Intensity Based on Risk Amplification Within the preset time window Internal raw reserve adequacy index Perform risk corrections and generate a revised reserve adequacy index.

[0012] As a further improvement to this technical solution, in step S5, flood control emergency supplies that meet the allocation conditions are screened based on the material availability assessment results and material allocation demand results. Combined with the material storage structure model, the allocation path is analyzed to generate a corresponding material transportation plan, including the following steps: S5.1 Based on the results of the material availability assessment and the material allocation demand, various flood control emergency materials are screened, a list of materials that meet the allocation conditions is determined, and a set of materials that can be allocated is generated. S5.2. Combining the material storage structure model, use the graph theory shortest path algorithm to analyze the allocation path of each type of material to be allocated in the set of available materials. S5.3 Calculate the time required for material transportation based on the allocation route analysis results, and compare the time required for material transportation with the preset emergency response time to determine whether the flood control emergency materials can reach the target location on time. S5.4 Integrate the results of comparing the available materials, allocation routes, and transportation time with the preset emergency response time to form a complete material allocation plan. S5.5 Update the material transportation plan to the flood control emergency material information database simultaneously.

[0013] On the other hand, the present invention provides a flood control emergency material allocation and storage management system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the above-mentioned flood control emergency material allocation and storage management method.

[0014] Compared with existing technologies, the above-mentioned technical solution of the present invention has the following beneficial technical effects: by constructing a unified coding system and a structured storage model, and integrating barcode quality-aware adaptive recognition, multi-source heterogeneous data preprocessing and dynamic demand prediction mechanisms, the entire process of flood control emergency materials from warehousing, storage, allocation to tracking is realized; the standardization, real-time performance and intelligent decision-making capabilities of industrial data management are significantly improved, and the problem of low allocation efficiency caused by data silos, identification errors and response delays in traditional flood control material management is effectively solved, providing highly reliable, traceable and optimizable data support and scheduling basis for emergency response. Attached Figure Description

[0015] Figure 1 This is a flowchart of the overall method of the present invention. Detailed Implementation

[0016] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1: Please refer to Figure 1 As shown in the figure, this embodiment provides a method for the allocation, storage, and management of flood control emergency supplies, including the following steps: S1. Collect basic attribute data, inventory status data and historical call data of flood control emergency supplies, establish unique identification codes for various flood control emergency supplies, and build a flood control emergency supplies information database. In this embodiment, basic attribute data, inventory status data, and historical requisition data of flood control emergency supplies are collected. Unique identification codes are established for various types of flood control emergency supplies, and a flood control emergency supplies information database is constructed. This includes the following steps: S1.1 Collect basic attribute data of flood control emergency supplies. The basic attribute data shall include at least the category, specifications, production batch and expiration date of the supplies. The basic attribute data shall be obtained through manual entry, import from the business system or scanning and identification, and shall be used to characterize the static characteristics of the flood control emergency supplies. S1.2 Collect inventory status data of flood control emergency supplies. The inventory status data shall include at least the current inventory quantity, storage warehouse, storage area and entry / exit time information. The inventory status data shall be obtained in real time through the warehouse management system or entry / exit identification equipment to reflect the dynamic inventory status of flood control emergency supplies. S1.3 Collect historical data on the use of flood control emergency supplies. The historical data should include at least the time of each use, the quantity used, the reason for use, the destination of use, and the result of use. By summarizing and storing the historical data, the frequency of use and allocation patterns of flood control emergency supplies can be characterized. S1.4. Preprocess the basic attribute data, inventory status data and historical call data of the collected flood control emergency materials: unify the format, standardize the fields and perform data verification on the data from different sources, including unify the units, align the time format and correct abnormal data, to form a standardized flood control emergency materials dataset, which provides a consistent data foundation for subsequent unified coding and database construction; S1.5. Based on the preprocessed basic attribute data, inventory status data and historical call data, generate a unique identification code for each type or batch of flood control emergency materials. The unique identification code contains at least material category information and batch information, and serves as a unique index identifier for the materials to achieve consistent identification of the materials during the collection, storage, allocation and tracking process. S1.6. Construct a flood control emergency supplies information database based on the unique identifier code, and establish the relationship between the basic attribute data table, the inventory status data table and the historical call data table based on the unique identifier code, thereby constructing a unified flood control emergency supplies information database to provide data support for subsequent reserve assessment, allocation decision-making and process management. The flood control emergency supplies information database adopts a relational database architecture and standardized data model design. It mainly includes several logical sub-tables such as a basic material attribute table, an inventory status table, a historical call table, a storage unit table, a transportation plan table, and a transfer log table. Inter-table relationships are established through unique identifier codes and storage unit numbers. The basic material attribute table stores static attributes such as the unique identifier code, material category, specifications, production batch, and expiration date for each type or batch of flood control emergency supplies. The inventory status table records in real-time the current inventory quantity, warehouse, storage area, storage unit number, and most recent entry / exit time for each material code. The historical call table summarizes and stores the call time, quantity, reason, destination, and result of each call. The system includes: a time-series data retrieval table; a storage unit table describing storage structure information such as storage unit number, unit capacity, compatible material type, current occupancy status, and availability status; a transportation plan table storing transportation plan data such as material code, allocation quantity, priority, starting storage unit, target location, allocation route, and estimated transportation time for each transportation plan; and a transportation log table recording status updates during the material allocation process, including allocation execution time, actual transportation time, information on allocation personnel or equipment, and anomaly handling records. All these tables are linked by foreign keys using unique material identifiers and storage unit numbers, enabling unified access to material information, status tracking, and cross-table queries. This supports integrated flood control and emergency material management functions, including inventory management, reserve assessment, predictive modeling, and transportation decision-making.

[0018] S2. Divide the storage environment of flood control emergency materials into units, and construct a material storage structure model based on the unit division results; In this embodiment, the storage environment for flood control emergency supplies is divided into units, and a storage structure model is constructed based on the unit division results, including the following steps: S2.1. Collect information on the storage environment of flood control emergency supplies based on the flood control emergency supplies information database. The storage environment information includes at least the number of reserve warehouses, warehouse space layout, storage area division, shelf or stacking structure and aisle distribution information, which are used to characterize the actual storage environment of flood control emergency supplies. S2.2 Based on storage environment information, the storage environment for flood control emergency supplies is hierarchically divided. The storage environment is split according to the hierarchical structure of warehouse, storage area and storage unit. Each storage unit is used to accommodate one type or batch of flood control emergency supplies, thus forming a storage environment structure with a clear hierarchical relationship. S2.3. Based on the hierarchical partitioning results, generate a unique storage unit number for each storage unit. The storage unit number is used to uniquely identify the location and hierarchical attributes of the storage unit and serves as the basic identifier for associating flood control emergency supplies with storage locations. S2.4. Based on the storage unit number, establish corresponding unit attribute information for each storage unit. The unit attribute information includes at least the unit capacity, the type of material to be adapted, the current occupancy status, and the availability status, which are used to describe the carrying capacity and usage constraints of the storage unit. S2.5. Establish the structural relationship between storage units based on unit attribute information, including the hierarchical membership and relative spatial relationship of storage units, thereby forming a structured relationship model that can reflect the spatial organization of flood control emergency materials storage. S2.6. Generate a material storage structure model for flood control emergency supplies based on the hierarchical structure of the storage environment, storage unit numbers, unit attribute information, and structural relationships. (When constructing the flood control emergency supplies storage structure model, firstly, the hierarchical structure information of the storage environment is integrated, including the hierarchical relationships of warehouses, storage areas, and storage units, and the storage area and warehouse to which each storage unit belongs are clearly defined, thus forming a complete hierarchical framework to provide the basic skeleton for the model; then, the unique number of each storage unit is associated with its hierarchical position to ensure that the unit can be uniquely identified in the model and its physical location and hierarchical attributes can be traced; next, the attribute information of each storage unit (such as capacity, compatible material types, current occupancy status, and availability) is...) By associating status with numbering, a model is introduced that represents not only spatial layout but also storage capacity and usage constraints. Based on this, structural relationships between storage units are established according to their spatial layout and hierarchical relationships, including hierarchical relationships, neighboring unit relationships, and channel connectivity, to support material allocation route planning and spatial optimization. Finally, the integrated hierarchical structure, unit numbering, unit attributes, and structural relationships are uniformly encapsulated to generate a queryable and updatable material storage structure model. This model can be quickly retrieved by warehouse, storage area, material type, or unit number, facilitating inventory management and allocation decisions. This material storage structure model is used for subsequent flood control emergency material inventory management, allocation route analysis, and transportation plan generation. The input layer of the material storage structure model receives basic attribute data of flood control emergency materials, storage unit division information, and historical inventory and allocation data. The middle layer constructs a material storage network through graph structure or matrix representation methods, and performs correlation calculations on the physical location relationship, available capacity, and retrieval path between storage units. At the same time, it uses node weights and edge weights to represent storage efficiency and retrieval cost. The output layer generates the storage location distribution, available inventory, and allocation priority of each type of material in each storage unit, providing direct decision-making basis for material allocation and inventory optimization. S2.7 Map the unique identification code of each type or batch of flood control emergency supplies to the storage unit where they are actually stored. By recording the material code, storage unit number, entry time and quantity, unified identification and traceable management of materials during storage, entry and exit and allocation can be achieved, providing data support for subsequent automated allocation operations.

[0019] S3. Automatically identify the entry, exit, and relocation of flood control emergency supplies through barcode recognition, and simultaneously update the flood control emergency supplies information database upon completion of the identification. In this embodiment, barcode recognition is used to automatically identify the entry, exit, and relocation operations of flood control emergency supplies, and the flood control emergency supplies information database is updated simultaneously upon completion of the identification. This includes the following steps: S3.1. Barcode the unique identification code of each type or batch of flood control emergency supplies, and collect the barcode information of flood control emergency supplies through scanning equipment before warehousing to ensure that each item of supplies has a traceable unique identifier in the system, providing basic data for automated warehousing operations; S3.2 During the process of flood control emergency supplies entering the warehouse, the quality perception adaptive identification method of flood control emergency supplies barcodes is used to identify and verify the barcode information of flood control emergency supplies. After the identification and verification are completed, the information of material category, batch, quantity and entry time is obtained, and the identification results are updated to the flood control emergency supplies information database in real time to ensure that the inventory status data reflects the entry status of materials in a timely and accurate manner. In this embodiment, the quality-aware adaptive recognition method for barcodes used in flood control emergency supplies addresses the specific problems in flood control emergency scenarios where barcodes are physically damaged or their image quality severely degraded due to long-term storage, frequent transportation, and harsh environments (such as water immersion, soiling, and uneven lighting). Traditional fixed recognition algorithms suffer from high failure rates and error rates, leading to delays in inventory data updates and inaccurate allocation decisions. Its core solution is to achieve highly reliable and automated barcode information collection under complex damage conditions to ensure the real-time accuracy and synchronization of material traceability and inventory status. This method dynamically couples barcode quality assessment with the recognition process. By extracting quality parameters such as stripe width deviation and damaged area, it adaptively selects or fuses multiple recognition strategies and performs graded verification and error correction based on confidence levels. This significantly improves the first-time recognition success rate and the automation level of data entry under harsh conditions, ensuring the reliability of inventory data and the timeliness of allocation responses from the source. The method of quality-aware adaptive recognition of flood control emergency supplies barcodes for identification and verification includes the following steps: S3.21. Obtain the original barcode image (during the warehousing, retrieval, or relocation of flood control emergency supplies, photograph or collect the unique identification barcode on each item using barcode scanning equipment (such as a fixed scanner, handheld barcode scanner, or industrial camera) to obtain the original barcode image), perform quality parameter analysis on the original barcode image, and extract parameters including stripe width deviation rate. Stripe width uniformity index Percentage of damaged area Damage dispersion index Effective black and white stripe contrast Boundary continuity index and boundary fracture density A set of quality parameters is used to characterize the identifiability of barcodes for flood control emergency supplies; Among them, the stripe width deviation rate is extracted. Stripe width uniformity index Percentage of damaged area Damage dispersion index Effective black and white stripe contrast Boundary continuity index and boundary fracture density The specific set of quality parameters is as follows: First, the input barcode image is grayscaled and denoised. The effective area of ​​the barcode is extracted through projection analysis and morphological constraints. The effective area of ​​the barcode is then initially binarized based on the standard Otsu threshold to obtain an initial binary barcode image. Within this initial binary barcode image, a stripe set is obtained. Within the valid area of ​​the barcode, calculate the actual width of each stripe. and the corresponding standard stripe width (Determined by barcode encoding rules or templates) Compare and calculate the stripe width deviation rate. (This indicator is used to characterize the degree of systematic width distortion of barcodes caused by printing, scanning, or perspective distortion.) ; In the formula, The total number of stripes included in the calculation within the valid area of ​​the barcode. Assign a stripe index number; Meanwhile, to characterize the local width stability between stripes, the stripe width uniformity index is calculated. : ; In the formula, This is the standard deviation function, used to calculate the dispersion of the actual width of all stripes. This is a mean function used to calculate the average of the actual widths of all stripes; Since stripe width distortion directly affects grayscale distribution and foreground / background segmentation accuracy, a width deviation rate is introduced. The global threshold is modulated, and an adaptive threshold coupled with stripe distortion is constructed. (Using adaptive threshold) (The valid area of ​​the barcode is subjected to refined binarization processing to obtain the final binary barcode image for subsequent damage area detection and boundary analysis). ; In the formula, To extract barcode images from raw images using the Otsu method The globally optimal threshold is calculated in [the following context]. The width deviation modulation coefficient is used to control the influence of the stripe width deviation rate on the threshold adjustment amplitude. Its value ranges from 0.1 to 0.5 and is determined by expert experience. Based on the final binary barcode image, connected component analysis is used to identify broken, missing, or contaminated regions in the stripes, defining them as a set of damaged regions; the total area of ​​the damaged regions is then calculated. Area of ​​the barcode's effective area Calculate the percentage of damaged area (Used to characterize the overall severity of physical damage or imaging degradation of barcodes): ; Based on the proportion of damaged area Furthermore, the spatial distribution characteristics of the damage are considered; the spatial distribution of the damaged area within the valid barcode area is statistically analyzed, and the damage dispersion index is calculated. Specifically: In the final binary barcode image, all damaged regions (i.e., missing, broken, or contaminated regions inconsistent with the ideal barcode structure) are identified through connected component analysis, and each damaged region is treated as an independent damaged patch; if there are a total of There are 10 damaged plaques, each with an area of ​​100 square meters. Its normalized coordinates within the valid area of ​​the barcode are Then the damage dispersion index The coefficient of variation of the spatial distribution among damaged patches is calculated as follows: First, the average Euclidean distance between the centroids of the damaged patches is calculated. and standard deviation ,but ; The larger the value, the more dispersed and random the damage distribution, and the more significant the impact on barcode recognizability. In the original grayscale image, the mean and standard deviation of the white and black bar regions are calculated separately, and the contrast of the black and white bars is calculated. : ; In the formula, The average gray value of the white bar area. The average gray value of the black bar area. The standard deviation of the gray values ​​in the white bar area. The standard deviation of the gray values ​​in the black bar area; Given that damage can lead to localized grayscale aliasing and decreased contrast, a contrast attenuation coefficient based on the proportion of damaged area is introduced. (Used to characterize the exponential effect of barcode damage on grayscale contrast): ; In the formula, The damage attenuation adjustment coefficient is an exponential function used to control the influence of the proportion of damaged area on the contrast attenuation intensity. Its value ranges from 2 to 10 and is determined experimentally. Using contrast attenuation coefficient Contrast of black and white stripes Corrections were made to obtain an effective black-and-white stripe contrast for imperfect perception. This is used to more accurately reflect the actual identifiability of barcodes under damaged conditions; Edge detection was performed on the final binary barcode image using the Sobel operator to extract the stripe boundary structure; connectivity analysis was then performed on the detected boundaries to calculate the length of continuous boundaries. Total length of reference boundary The boundary continuity index is obtained. : ; Simultaneously, the boundary fracture density was statistically analyzed. : ; S3.22. Based on the quality parameter set, an image preprocessing strategy matching the degree of barcode damage is used (the image preprocessing strategy includes CLAHE local contrast enhancement for mild damage, combined enhancement with illumination correction, direction-guided filtering, and adaptive binarization for moderate damage, and structural reconstruction of the damaged area by calling a pre-trained barcode repair generative adversarial network model) to adaptively enhance the original barcode image (implemented through the adaptive enhancement processing module) to obtain enhanced barcode images (enhanced barcode images include mildly enhanced barcode images, moderately enhanced barcode images, and structurally repaired barcode images) to improve the stability and accuracy of barcode recognition. Specifically, based on the set of barcode quality parameters obtained in step S3.21, the stripe width deviation rate is calculated. Percentage of damaged area Based on this, the distortion level of the barcode is determined. The original barcode image is then input into the adaptive enhancement processing module. The module dynamically calls the corresponding image preprocessing strategy according to the determined distortion level to perform enhancement processing on the original barcode image, generating an enhanced barcode image. When the stripe width deviation rate And the proportion of damaged area At that time, the distortion level was determined to be minor damage. The adaptive enhancement processing module adopted a fast enhancement strategy, using the original barcode image as input, and performed local contrast enhancement on the original barcode image through the Limited Contrast Adaptive Histogram Equalization (CLAHE) algorithm (wherein, the contrast limitation parameter is adaptively set according to the proportion of damaged area, and the block size is fixed). Output a slightly enhanced barcode image; When the stripe width deviation rate or percentage of damaged area At that time, the distortion level was determined to be moderate damage. The adaptive enhancement processing module performed joint enhancement of illumination correction and structural repair: illumination correction was performed on the original barcode image based on Retinex theory (first, the barcode image was decomposed into illumination and reflection components, the illumination of each pixel was estimated using the multi-scale Retinex method, and the influence of uneven illumination was removed by logarithmic operation and mean filtering of local areas, while retaining stripe edge and detail information. Finally, the corrected reflection component was fused with the original image to generate an illumination-corrected barcode image), and the stripe width deviation rate was combined with the correction. The Gaussian smoothing scale parameter is adaptively set; principal component analysis is performed on the barcode edge points to obtain the main direction of the stripes, and direction-guided filtering is applied along the main direction. At the same time, image statistical features (quantitative features that describe the brightness distribution and stripe structure information of the image in a numerical way by calculating the mean, variance, entropy, gradient and other indicators of image gray level or local texture) are fused with barcode quality parameters to construct an adaptive binarization threshold. The illuminated barcode image is then subjected to binary enhancement processing to output a moderately enhanced barcode image. When the stripe width deviation rate And the proportion of damaged area When the distortion level is determined to be severely damaged, the adaptive enhancement processing module uses a pre-trained barcode repair generative adversarial network model. Taking the original barcode image and the corresponding damaged area mask as input, it reconstructs and enhances the barcode structure, generates a barcode image after structural repair, and recalculates and updates the quality parameter set for the repaired barcode image. The pre-trained barcode repair generative adversarial network (GAN) model is used to reconstruct the structure of severely damaged barcode images. Its input is the original barcode image and the corresponding damaged area mask, and its output is the structurally repaired barcode image. The network consists of a generator and a discriminator. The generator employs an encoder-decoder structure: the encoder includes several convolutional layers (Conv2D) and batch normalization layers (BatchNorm) to extract local and global features of the barcode image; the decoder includes several deconvolutional layers (ConvTranspose2D) and activation layers (ReLU / LeakyReLU), combining residual connections and skip connections to recover stripe details; the discriminator is a multi-layer convolutional discriminant network. A network is used to determine the similarity between the generated image and the real barcode image. The network is trained using an adversarial training method. The generator loss function includes reconstruction loss (L1 or L2 loss), structural similarity loss (SSIM), and discriminator-guided adversarial loss. The discriminator loss is the standard binary classification cross-entropy. The pre-training method is to use a large-scale barcode image dataset for iterative training. The generator and discriminator weights are initialized using Xavier initialization. The training optimizer is the Adam optimizer. The learning rate can be set to 0.0002, and the batch size can be set to 16~64. Training continues until the generator output image converges and the discriminator can hardly distinguish between the real and generated images, thus obtaining a pre-trained model that can be directly used for the repair of severely damaged barcodes. S3.23. Based on the set of quality parameters and the enhanced barcode image (for severely damaged cases, the quality parameters are recalculated after repair), the barcode recognition results and corresponding confidence levels of flood control emergency supplies are obtained through a parameterized multi-strategy barcode recognition method, and the barcode recognition results are subjected to adaptive verification and error correction based on quality parameters. Among them, the parameterized multi-strategy barcode recognition method refers to: using a set of barcode quality parameters as a unified driving variable, parameterizing key steps in the barcode recognition process such as image enhancement, binarization, edge detection, stripe reconstruction and decoding, and pre-configuring multiple complementary recognition strategies (including standard barcode decoding strategy, multi-frame fusion recognition strategy and feature matching recognition strategy). During actual recognition, the corresponding strategy and its parameter combination are dynamically selected or weighted and fused according to quality parameters such as stripe width deviation rate, damaged area ratio, contrast index and damage dispersion. This enables the recognition process to adaptively match barcodes with different damage types and degrees, thereby improving the success rate and stability of barcode recognition in complex damage scenarios without relying on a single fixed algorithm. In this embodiment, the parameterized multi-strategy barcode recognition method addresses the specific problems of harsh environments in flood control emergency sites, where material barcodes are prone to severe image quality degradation due to water immersion, wear, dirt, or uneven lighting. This leads to high failure rates and misread rates of traditional fixed recognition algorithms, resulting in inaccurate inventory data and delayed allocation responses. Its core solution is how to achieve highly robust and accurate automatic barcode recognition under complex physical damage conditions, ensuring the reliability of the data source for subsequent real-time inventory status updates and allocation decisions. It quantifies the degree of physical damage to barcodes as a quality parameter, driving the dynamic selection and verification / correction process of recognition strategies, rather than relying on a single algorithm or fixed threshold. Existing technologies typically use general-purpose decoders, which fail directly or require manual intervention when damaged. This invention, through a closed loop of quality perception, strategy adaptation, and confidence-level verification, achieves a leap from "whether it can be recognized" to "how to accurately recognize," significantly improving the first-time recognition success rate and data entry automation level under harsh emergency conditions, fundamentally supporting the real-time performance and reliability of the entire closed-loop management process. Furthermore, a parameterized multi-strategy barcode recognition method is used, and the barcode recognition results are subjected to adaptive verification and error correction based on quality parameters, including the following steps: Using a set of quality parameters and the enhanced barcode image as input, a barcode recognition strategy selection function is constructed based on the stripe width deviation rate, damaged area ratio, stripe width uniformity index, and damage dispersion index. This function generates an overall barcode recognizability score by normalizing and weighting the enhanced barcode image and its quality parameter set. The recognizability of the enhanced barcode image is then determined by comparing the overall recognizability score with high, medium, and low recognizability thresholds to form a quantifiable barcode recognizability judgment result: when the overall recognizability score is greater than or equal to the high recognizability threshold... When the overall barcode recognizability score is less than the high recognizability threshold, select the standard barcode decoding strategy; when the overall barcode recognizability score is less than the high recognizability threshold... And greater than or equal to the identifiability threshold When the overall barcode recognizability score is less than the medium recognizability threshold, a multi-frame fusion recognition strategy is selected; When selecting a feature matching and recognition strategy, in this embodiment, , (determined through expert experience), and the identification strategy type is selected based on the judgment result. The identification strategy types include standard barcode decoding strategy, multi-frame fusion identification strategy, and feature matching identification strategy. When selecting a standard barcode decoding strategy, the stripe width deviation rate is used as a reference. and percentage of damaged area The decoding timeout, smoothing filter parameters, and number of decoding attempts are adaptively set. The standard decoding algorithm is executed on the enhanced barcode image according to the barcode encoding rules to obtain the stripe sequence and parse the barcode information to generate a preliminary barcode recognition result. When selecting a multi-frame fusion recognition strategy, multiple frames of barcode images of the same flood control emergency supplies are continuously acquired, and the number of barcode image frames participating in the fusion is determined based on the proportion of damaged area (the proportion of damaged area is calculated frame by frame for each of the continuously acquired barcode images). and each frame With preset high, medium and low thresholds Comparison; for Frames that are of high quality are considered high-quality frames and are included in the fusion by default; for The frames are weighted using a linear or exponential decay function. Only frames with weights greater than the minimum weight threshold are selected for fusion; for Low-quality frames can be excluded or used as alternative frames only when the total number of frames is insufficient; ultimately, the maximum number of frames to be merged is limited according to the strategy. (For example, 5), select the top-weighted frames from the high-weighted frames, sorted by weight. The frames are used as a set of fused frames, and the number of barcode image frames participating in the fusion is generated. The recognition quality of each frame is evaluated, and a weighting function is constructed based on the mean square error between barcode images. After weighted fusion of the multiple barcode images, barcode decoding is performed to obtain the fused barcode recognition result. (First, preliminary barcode recognition is performed on each frame, and the recognition quality score of each frame is calculated based on quality parameters (weighted calculation). Then, the frame with the highest recognition quality score is used as the reference frame, and the mean square error (MSE) between other frames and the reference frame in the striped grayscale or binarized image is calculated. A weighting function is constructed based on the MSE difference.) , This represents the target image to be fused (or the initial fused image). For the first Frame barcode image, For the frame index participating in the fusion, To prevent division by zero (using a small constant for smoothing or normalizing weight calculations), frames with small differences from the reference frame receive higher weights. This weighting function is then used to weight and superimpose or pixel-level fuse all participating barcode images to generate a fused image. Finally, standard barcode decoding is performed on the fused image to parse the stripe sequence and output the fused barcode recognition result. Simultaneously, the fusion weights and recognition parameters are recorded, providing a data foundation for confidence calculation and subsequent error correction. When selecting a feature matching and recognition strategy, the local feature type (including at least corner features, edge features, and local binary patterns) is adaptively selected based on the stripe width deviation rate. Local feature points are extracted from the enhanced barcode image, and the feature matching results are constrained by a random consistency algorithm to obtain matching results that satisfy geometric consistency. Based on the matching results, the barcode encoding information is judged, and the barcode recognition result is generated. Based on a set of quality parameters, a recognition confidence calculation model is constructed to calculate a basic recognition confidence score for barcode recognition results, characterizing the reliability of the current barcode recognition outcome. This model employs a three-layer structured architecture: the input layer receives two core elements: first, a set of quality parameters extracted from the original barcode image, including stripe width deviation rate, damaged area ratio, stripe width uniformity index, damaged dispersion index, effective black and white stripe contrast, boundary continuity index, and boundary break density; second, the currently adaptively selected barcode recognition strategy type (standard barcode decoding strategy, multi-frame fusion recognition strategy, or feature matching recognition strategy). The middle layer is the core of the calculation. It first normalizes and weights each quality parameter to generate a comprehensive image quality score; simultaneously, it assigns a corresponding basic strategy confidence score to different recognition strategy types; then, through a weighted fusion function or a nonlinear mapping function (such as the Sigmoid function), it combines the image quality score with the basic strategy confidence score to calculate a preliminary confidence value. The output layer outputs the final recognition confidence score, a continuous value between 0 and 1, which is directly used in the subsequent hierarchical verification process: high confidence scores pass directly, medium confidence scores trigger check bits and structural verification, and low confidence scores trigger parallel recognition by multiple algorithms and consistency analysis. This architecture ensures that the confidence score assessment is closely related to the objective quality of the image and the inherent reliability of the selected recognition algorithm. The barcode recognition results are verified in stages based on the recognition confidence level: when the recognition confidence level is higher than the first threshold 'a' (a is 0.92 in this embodiment), the verification is passed directly; when the recognition confidence level is between the first threshold 'a' and the second threshold 'b' (b is 0.7 in this embodiment), the barcode recognition results undergo check bit verification and barcode encoding structure consistency verification (firstly, check bits are extracted from the barcode recognition results according to the barcode standard (such as parity check bits or CRC check codes in UPC / EAN barcodes), and the recognized data is mathematically verified according to the corresponding algorithm to determine whether the check bits match the data content, thereby verifying the integrity of the recognition; at the same time, the barcode encoding structure consistency verification is performed, including checking whether the barcode start character, stop character, separator, and the encoding length and stripe spacing of each stripe conform to the standard encoding rules to ensure that the barcode recognition results are legal in encoding format; if the verification passes, the barcode recognition results are considered reliable, otherwise error correction or multi-strategy recognition processes are triggered to improve the accuracy and robustness of the barcode recognition results); when the recognition confidence level is lower than the second threshold 'b', the verification is triggered... Parallel recognition using multiple recognition algorithms is employed, and consistency analysis is performed on multiple barcode recognition results. This involves simultaneously executing standard barcode decoding, multi-frame fusion recognition, and feature matching recognition strategies on the same barcode image or multiple enhanced images to obtain multiple candidate barcode recognition results. Subsequently, consistency analysis is performed on these candidate barcode recognition results, including comparing the similarity of the encoding sequences, check digit matching, and barcode structure legality. Results that are completely consistent or highly similar are aggregated and weighted. For multiple barcode recognition results (i.e., the initial barcode recognition result obtained through the standard barcode decoding strategy, the fused barcode recognition result obtained through the multi-frame fusion recognition strategy, and the barcode recognition result obtained through the feature matching recognition strategy), a weight is first assigned to each result (firstly, normalization is performed based on the algorithm's own output confidence level; simultaneously, key parameters that significantly affect recognition stability (at least including stripe width deviation rate, damaged area ratio, and stripe width uniformity index) are selected from the barcode quality parameter set, and single-parameter reliability mapping functions are constructed to map the quality parameters to... The quality reliability score within the interval is weighted and fused according to preset weights to obtain the quality perception score; the success rate of the recognition algorithm in historical recognition tasks is further introduced to reflect the long-term reliability of the algorithm in similar scenarios; finally, the above three factors are fused by weighted product or weighted summation to form the comprehensive weight coefficient of the recognition algorithm; then, for barcodes with the same value among multiple barcode recognition results, their corresponding weights are accumulated to form the total weighted score of the barcode recognition result; finally, the barcode recognition result with the highest total weighted score is selected as the final barcode recognition result; the barcode recognition result with the highest weighted score is selected as the final barcode decoding result, and the contribution weight and confidence of each candidate result are recorded to provide a basis for subsequent error correction or manual review. If the sub-verification fails, the barcode recognition result is corrected based on the set of quality parameters, including replacing or reconstructing easily confused code bits; multiple candidate barcode recognition results are weighted and voted to determine the final barcode recognition result, and the corresponding final recognition confidence is output; when the divergence of multiple candidate barcode recognition results exceeds a preset threshold, the barcode recognition result is marked to enter the manual review process. S3.3 During the process of material outbound or allocation, the quality perception adaptive identification method of flood control emergency material barcode is used to identify and verify the information of flood control emergency material barcode. After the identification and verification are completed, the unique identifier, quantity and outbound time of the outbound material are obtained, and the relevant information is updated to the flood control emergency material information database in real time to record the changes in material inventory. S3.4 When moving or adjusting the position of flood control emergency materials within the flood control emergency material storage unit, the quality perception adaptive identification method of the flood control emergency material barcode is used to identify and verify the barcode information of the flood control emergency material. After the identification and verification are completed, the starting unit, target unit and time information of the movement are recorded, and the storage unit mapping relationship of the materials in the database is updated to ensure the real-time accuracy of the material position. S3.5 After the barcode information of flood control emergency supplies is identified, the information on warehousing, outbound and relocation operations will be synchronized with the flood control emergency supplies information database: update the inventory status, current occupancy status and storage unit information of flood control emergency supplies, and provide a reliable data foundation for subsequent reserve assessment, allocation analysis and transportation plan generation.

[0020] S4. Based on the updated flood control emergency supplies information database, conduct a comprehensive assessment of the adequacy and availability of various flood control emergency supplies, and generate a supply availability assessment result and a supply allocation demand result. In the process of comprehensively assessing the adequacy of reserves, a material demand prediction model is constructed by using historical flood control emergency material call data, rainfall intensity change rate and warning water level difference to predict the consumption intensity of various flood control emergency materials within a preset time window, and to generate a predicted and corrected reserve adequacy index. In this embodiment, a material demand forecasting model is constructed to predict the consumption intensity of various flood control emergency materials within a preset time window, and a revised reserve adequacy index is generated. This addresses the specific problem in flood control emergency management where material demand is highly dynamic and prone to sudden changes, and traditional static inventory models are slow to respond, leading to a disconnect between reserve quantities and actual emergencies, resulting in "coexistence of material stockpiling and shortages." Its core solution is how to transform real-time flood dynamics (rainfall intensity change rate, water level difference above warning level) into a forward-looking prediction of material consumption intensity, thereby upgrading inventory management from passive response to proactive early warning. This invention couples dynamic environmental factors (rainfall change rate, water level difference) into demand forecasting in real time through a mathematical model and performs risk correction on reserve indicators, establishing a three-layer forecasting architecture of historical baseline, trend correction, and risk amplification. This allows the reserve adequacy index to dynamically reflect the urgency of the flood control situation, achieving a leap from static inventory management to dynamic risk-aware allocation, significantly improving the scientific and forward-looking nature of emergency material allocation. The process of generating resource availability assessment results and resource allocation requirements includes the following steps: S4.1. Based on basic attribute data, determine the inventory baseline and safety reserve threshold for each type of flood control emergency supplies. Specifically, based on basic attribute data, calculate the minimum guarantee quantity of this type of supplies under normal and emergency conditions to obtain the inventory baseline quantity (first, classify various flood control emergency supplies according to material category, specifications, and applicable flood control operation type to determine the standard usage scenario corresponding to each type of material; then, calculate the demand quantity of this type of material in a unit flood control operation scenario based on the standard usage quantity per operation, the range of operations that can be covered, and the specified minimum guarantee duration; on this basis, according to the preset number of flood control operation scenarios or the scale of typical flood control tasks, accumulate the demand quantity per unit scenario to obtain the total demand quantity of this type of flood control emergency supplies in the complete flood control cycle; finally, use the total demand quantity as the inventory baseline quantity of this type of flood control emergency supplies for subsequent inventory status assessment and material allocation analysis); then, based on the material supply cycle, transportation timeliness, and importance of flood control tasks, set a safety reserve threshold to trigger replenishment or allocation strategies when the inventory is lower than the threshold. S4.2 Utilize the updated flood control emergency supplies information database to update the inventory quantity, current occupancy status and availability status of each storage unit in real time; for supplies that are close to or have exceeded their expiration date, mark their availability status as "restricted" through the database to distinguish between supplies that can be directly allocated and restricted supplies, and ensure the authenticity and reliability of the assessment results. S4.3. By using historical flood control emergency material dispatch data, rainfall intensity change rate and warning water level difference, construct a material demand prediction model to predict the consumption intensity of various flood control emergency materials within a preset time window, calculate the original reserve adequacy index, and perform risk correction on the original reserve adequacy index based on the predicted consumption intensity to generate the predicted and corrected reserve adequacy index. The process of adjusting the original reserve adequacy index based on the predicted consumption intensity to generate a revised reserve adequacy index includes the following steps: S4.31. Based on historical data on the allocation of flood control emergency supplies, extract the total allocation of various types of flood control emergency supplies in different flood control events, targeting different types of flood control events. and the duration of the corresponding event By normalizing the total amount of calls according to the duration of the corresponding events, the historical unit-time consumption intensity of various flood control emergency materials under different flood control events was calculated. Furthermore, statistical aggregation of the unit-time consumption intensity of multiple historical flood control events was conducted to construct a baseline for the historical unit-time consumption intensity of various flood control emergency materials. (In the formula, For the first The number of historical flood control events in which flood control emergency supplies were included in the statistics. Numbering historical flood control events (This serves as an index for flood control emergency supplies categories, providing a reference standard for subsequent forecasts); S4.32, Obtain the target area within the preset time window Rainfall intensity sequence within (Call on meteorological monitoring stations or radar rainfall data interfaces to collect rainfall intensity sequences of the target area at preset time intervals) (forming discrete time series), based on rainfall intensity series Calculate the instantaneous rate of change of rainfall intensity (In the formula, For time, For time The first derivative), the instantaneous rate of change Mapped to trend weights using a normalization function. (In the formula, This is the minimum adjustment coefficient (when rainfall changes are gradual). This is the maximum adjustment coefficient (when rainfall changes rapidly). (As a parameter for smoothing the rate of change to prevent small disturbances from being amplified), through trend weighting. Baseline of historical unit time consumption intensity A trend-weighted correction is performed, and a material demand forecasting model is generated to obtain a trend-corrected forecast of consumption intensity that takes into account the changing trend of rainfall intensity. ; The overall material demand forecasting model adopts a layered architecture design with baseline forecasting, dynamic correction, and risk enhancement. Its input layer receives multi-source time-series features such as historical flood control emergency material requisition data, basic attributes of material types, event duration, real-time rainfall intensity and its rate of change, and the difference between water level and warning water level, and performs time alignment and normalization on each input feature. The intermediate layer first models the historical requisition intensity through the baseline demand forecasting sublayer and outputs the baseline demand for various flood control emergency materials. Subsequently, a dynamic response correction sublayer was introduced to map the rate of change of rainfall intensity into an adjustment coefficient. Dynamically adjust the baseline demand to generate the predicted transitional demand. Furthermore, in the risk enhancement sublayer, the water level risk perception factor is calculated by combining the difference between the water level and the warning level. Risk sensitivity enhancement is applied to the forecasted transitional demand to form a revised forecast demand. The output layer uniformly outputs the final predicted demand for various flood control and emergency supplies within a preset time window. Predicting consumption intensity based on trend correction Calculate the original reserve adequacy index (In the formula, (Inventory quantity); S4.33, Obtain the real-time water level of the target area. and corresponding warning water level threshold Calculate the difference between the real-time water level and the warning water level threshold. and the difference Mapped to warning water level safety margin parameter (In the formula, The water level risk growth coefficient (unit: 1 / meter) is used to quantitatively characterize the degree to which the current water level is close to the warning threshold and the urgency of the flood control situation. S4.34, Safety margin parameters based on warning water level The trend-corrected predicted consumption intensity is subjected to nonlinear risk amplification processing; when the approach to the warning water level is higher than a preset threshold, the predicted consumption intensity of the corresponding flood control emergency materials is significantly increased through a nonlinear amplification function, resulting in the risk-amplified predicted consumption intensity. (In the formula, This is a risk amplification adjustment coefficient used to control the overall magnitude of the amplification of water consumption intensity due to the warning water level. It is based on analysis of historical flood control events. (This is a risk sensitivity coefficient, used to control the rate at which changes in the safety margin of the warning water level affect the amplification factor; it is determined through expert experience.) S4.35, Predicted Consumption Intensity Based on Risk Amplification Within the preset time window Internal raw reserve adequacy index Perform risk correction and generate a revised reserve adequacy index. This is used to reflect the actual reserve pressure under critical flood control conditions; S4.4. Based on the predicted and corrected reserve adequacy index, a comprehensive score is generated for the availability status of each type of flood control emergency supplies. (In the formula, This is the availability factor for materials, which can take values ​​from 0 to 1. It is used to quantify whether materials are restricted (such as restricted, nearing expiration, or damaged). To determine the weight of the reserve sufficiency indicator, As a weight for the availability of supplies, ); S4.5 Based on the predicted and corrected reserve adequacy index and the comprehensive score of availability, the allocation calculation module generates the allocation quantity and priority order of various flood control emergency materials, and generates the material allocation demand results. Specifically, this involves using the revised reserve adequacy index based on forecasts. Compared with inventory baseline For input, calculate the reserve shortage for each type of flood control emergency supplies. ; in, This refers to the current available inventory of this type of flood control emergency supplies; when there is a reserve shortage... At that time, it was determined that there was a need to allocate such materials; Comprehensive score based on availability Adjust the availability constraints for flood control emergency supplies that have allocation needs; when the supply availability status factor... When the quantity of materials allocated is lower than the preset available threshold F (in this embodiment, F is 0.6), the allocation quantity is reduced or marked as restricted allocation to prevent unavailable or high-risk materials from entering the allocation decision. Based on reserve gap Predicted revised reserve adequacy index and overall availability score Construct an emergency index for material allocation : ; In the formula, The weight of the inventory gap item, To reserve the weights of risk items, The weight of the material availability risk item, and It is used to comprehensively depict the impact of material shortages, reserve risk levels, and availability status. According to the urgency index of allocation The various flood control emergency supplies with allocation needs are sorted to generate a priority sequence for allocation; the higher the urgency index value, the higher the allocation priority of the flood control emergency supplies. Based on allocation priority sequence and reserve gap Determine the recommended allocation range for each type of flood control emergency supplies; when the available inventory is insufficient to fully meet the gap, allocate supplies according to priority to form a tiered allocation demand. The allocation demand, allocation priority, and availability constraints of various flood control emergency materials are uniformly packaged to generate material allocation demand results, which serve as direct inputs for the subsequent steps S5, including material screening, allocation path analysis, and transportation plan generation.

[0021] S5. Based on the results of material availability assessment and material allocation demand, select flood control emergency materials that meet the allocation conditions, and combine them with the material storage structure model to analyze the allocation path and generate corresponding material transportation plans for flood control emergency material storage and allocation management. In this embodiment, flood control emergency supplies that meet the allocation conditions are screened based on the results of the material availability assessment and the material allocation demand. Combined with the material storage structure model, the allocation path is analyzed to generate a corresponding material transportation plan, including the following steps: S5.1. Based on the material availability assessment results and material allocation demand results, various flood control emergency materials are screened to determine the material list that meets the allocation conditions and generate an allocation set of materials. Specifically, based on the obtained material availability assessment results and material allocation demand results, the flood control emergency materials in the inventory are first compared and screened by material category to identify the material types with allocation needs. Then, based on the material availability assessment results, the inventory batches corresponding to the material types are checked item by item to determine whether their availability status meets the preset allocation conditions, including that the inventory quantity is not lower than the minimum retention threshold, the material status is available or can be quickly restored to an available state, and it does not affect the local basic flood control guarantee needs. Under the premise of meeting the above conditions, the allocation quantity of the batch of materials is calculated after deducting the local retention quantity, and the material items with an allocation quantity of zero are removed. Finally, the material categories, corresponding inventory batches and their allocation quantities that have passed the screening are summarized to form a material list that meets the allocation conditions, and an allocation set of materials is generated accordingly. S5.2. Combining the material storage structure model, the graph theory shortest path algorithm is used to analyze the allocation path for each type of material to be allocated in the set of available materials. Specifically, based on the material storage structure model, the material storage units and their spatial layout, hierarchical relationships, and channel connectivity information are first abstracted into a graph structure, where the storage units are the nodes of the graph, and the traversable paths between units and the transportation distance or time are the weights of the edges. Then, for each type of material to be allocated in the set of available materials, the starting storage unit node and the target allocation warehouse or distribution point node are determined. On this basis, the graph theory shortest path algorithm (Dijkstra's algorithm) is used to calculate the optimal allocation path from the starting node to the target node, and the shortest transportation route required for material allocation is obtained. Finally, the optimal allocation path for each type of material is associated with the corresponding storage unit information and the available allocation quantity to form a complete allocation path analysis result, providing an accurate data foundation for generating allocation plans. S5.3. Calculate the required transportation time for materials based on the allocation path analysis results, and compare the required transportation time with the preset emergency response time to determine whether flood control emergency materials can arrive at the target location on time, and provide early warning and scheduling optimization for the risk of exceeding the time limit; specifically: First, calculate the total transportation time from the starting storage unit to the target allocation warehouse or distribution point by accumulating the edge weight information (including path length, speed limits, and possible channel congestion or traffic constraints) in each allocation path to obtain the transportation time of each material allocation path; then compare the calculated transportation time with the preset flood control emergency response time (different response times are set for different types of materials, for example, the response time for critical materials such as life-saving equipment and water pumps can be set to 1-2 hours) to determine whether each type of material to be allocated can arrive at the target location within the emergency response time; finally, record the transportation time and comparison results in the material allocation analysis table to provide a basis for allocation decisions, and mark the time-out path to trigger the backup allocation plan or expedited transportation strategy; S5.4 Integrate the results of comparing the available materials, allocation routes, and transportation time with the preset emergency response time to form a complete material allocation plan; the allocation plan shall include at least the unique identification code of each type of material to be allocated, the quantity and specifications of the material, the information of the starting warehouse and the target location, the corresponding storage unit number, the allocation route and the shortest transportation route, to ensure that the allocation process is traceable and executable; S5.5. Update the material transportation plan to the flood control emergency material information database simultaneously (write the material transportation plan into the flood control emergency material information database through the database interface or management system, and update the inventory status, current occupancy status and allocation route information in real time).

[0022] Example 2: This example provides a flood control emergency material allocation and storage management system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the flood control emergency material allocation and storage management method described in Example 1.

[0023] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A method for the allocation, storage, and management of flood control emergency supplies, characterized in that, include: S1. Collect basic attribute data, inventory status data and historical call data of flood control emergency supplies, establish unique identification codes for various flood control emergency supplies, and build a flood control emergency supplies information database. S2. Divide the storage environment of flood control emergency materials into units, and construct a material storage structure model based on the unit division results; S3. Automatically identify the entry, exit, and relocation of flood control emergency supplies through barcode recognition, and simultaneously update the flood control emergency supplies information database upon completion of the identification. S4. Based on the updated flood control emergency supplies information database, conduct a comprehensive assessment of the adequacy and availability of various flood control emergency supplies, and generate a supply availability assessment result and a supply allocation demand result. In the process of comprehensively assessing the adequacy of reserves, a material demand prediction model is constructed by using historical flood control emergency material call data, rainfall intensity change rate and warning water level difference to predict the consumption intensity of various flood control emergency materials within a preset time window, and to generate a predicted and corrected reserve adequacy index. S5. Based on the results of material availability assessment and material allocation demand, select flood control emergency materials that meet the allocation conditions, and combine them with the material storage structure model to analyze the allocation path and generate corresponding material transportation plans for flood control emergency material reserve and allocation management.

2. The method for allocating and storing flood control emergency supplies according to claim 1, characterized in that, In step S1, basic attribute data, inventory status data, and historical requisition data of flood control emergency supplies are collected. A unique identifier code is established for each type of flood control emergency supplies, and a flood control emergency supplies information database is constructed. This includes the following steps: S1.1 Collect basic attribute data of flood control emergency supplies. The basic attribute data shall include at least the category, specifications, production batch and expiration date of the supplies. S1.2 Collect inventory status data of flood control emergency supplies. The inventory status data shall include at least the current inventory quantity, storage warehouse, storage area and entry and exit time information. S1.3 Collect historical data on the use of flood control emergency supplies. The historical data should include at least the time and quantity of each use. S1.4 Preprocess the basic attribute data, inventory status data and historical call data of the collected flood control emergency materials; S1.

5. Based on the preprocessed basic attribute data, inventory status data, and historical call data, generate a unique identification code for each type of flood control emergency supplies; S1.6 Construct a flood control emergency material information database based on unique identifier codes.

3. The method for allocating and storing flood control emergency supplies according to claim 1, characterized in that, In step S2, the storage environment for flood control emergency supplies is divided into units, and a storage structure model is constructed based on the unit division results, including the following steps: S2.

1. Collect information on the storage environment of flood control emergency supplies based on the flood control emergency supplies information database; S2.2 Based on storage environment information, the storage environment of flood control emergency materials is hierarchically divided, and the storage environment is split according to the hierarchical structure of warehouse, storage area and storage unit; S2.

3. Generate a unique storage unit number for each storage unit based on the hierarchical partitioning result; S2.

4. Based on the storage cell number, establish corresponding cell attribute information for each storage cell; S2.

5. Establish the structural relationship between storage units based on unit attribute information; S2.6 Generate a material storage structure model for flood control emergency supplies based on the hierarchical structure of the storage environment, storage unit number, unit attribute information and structural relationships; S2.7 Map the unique identifier of each type of flood control emergency supplies to its corresponding storage unit.

4. The method for allocating and storing flood control emergency supplies according to claim 1, characterized in that, In step S3, barcode recognition is used to automatically identify the entry, exit, and relocation operations of flood control emergency supplies, and the flood control emergency supplies information database is updated simultaneously upon completion of the identification. This includes the following steps: S3.

1. The unique identification code of each type of flood control emergency supplies shall be processed into a barcode, and the barcode information of the flood control emergency supplies shall be collected before they are put into storage. S3.2 During the process of flood control emergency supplies entering the warehouse, the quality perception adaptive identification method of flood control emergency supplies barcodes is used to identify and verify the barcode information of flood control emergency supplies. After the identification and verification are completed, the information of material category, batch, quantity and entry time is obtained, and the identification results are updated to the flood control emergency supplies information database in real time. S3.3 During the process of material outbound, the quality perception adaptive identification method of flood control emergency material barcode is used to identify and verify the barcode information of flood control emergency material. After the identification and verification are completed, the unique identifier, quantity and outbound time of the outbound material are obtained and updated to the flood control emergency material information database simultaneously. S3.4 When moving flood control emergency supplies within the flood control emergency supplies storage unit, the quality perception adaptive identification method of the flood control emergency supplies barcode is used to identify and verify the information of the flood control emergency supplies barcode. After the identification and verification are completed, the starting unit, target unit and time information of the movement are recorded, and the storage unit mapping relationship of the supplies in the database is updated at the same time. S3.5 After the barcode information of flood control emergency supplies is identified, the information on warehousing, outbound and relocation operations will be synchronized with the flood control emergency supplies information database.

5. The method for allocating and storing flood control emergency supplies according to claim 4, characterized in that, In step S3.2, the barcode information of flood control emergency supplies is identified and verified using a quality-aware adaptive recognition method, including the following steps: S3.

21. Obtain the original barcode image, perform quality parameter analysis on the original barcode image, and extract a set of quality parameters including stripe width deviation rate, stripe width uniformity index, damaged area ratio, damage dispersion index, effective black and white stripe contrast, boundary continuity index, and boundary breakage density. S3.

22. Based on the set of quality parameters, an image preprocessing strategy that matches the degree of barcode damage is used to adaptively enhance the original barcode image to obtain the enhanced barcode image. S3.

23. Based on the set of quality parameters and the enhanced barcode image, the barcode recognition results and corresponding confidence levels of flood control emergency supplies are obtained through a parameterized multi-strategy barcode recognition method. The barcode recognition results are then subjected to adaptive verification and error correction based on the quality parameters.

6. The method for allocating and storing flood control emergency supplies according to claim 5, characterized in that, In step S3.23, a parameterized multi-strategy barcode recognition method is used, and the barcode recognition results are subjected to adaptive verification and error correction based on quality parameters, including the following steps: Using a set of quality parameters and an enhanced barcode image as input, a barcode recognition strategy selection function is constructed based on the stripe width deviation rate, the proportion of damaged area, the stripe width uniformity index, and the damage dispersion index. The function determines the recognizability of the enhanced barcode image and selects the recognition strategy type based on the determination result. The recognition strategy types include standard barcode decoding strategy, multi-frame fusion recognition strategy, and feature matching recognition strategy. Based on a set of quality parameters, a model for calculating recognition confidence is constructed to calculate the basic recognition confidence of the barcode recognition results. The barcode recognition results are verified in stages based on the recognition confidence level: when the recognition confidence level is higher than the first threshold 'a', the verification is passed directly; when the recognition confidence level is between the first threshold 'a' and the second threshold 'b', the barcode recognition results are checked for check bits and the consistency of the barcode encoding structure is checked; when the recognition confidence level is lower than the second threshold 'b', multiple recognition algorithms are triggered to recognize the barcodes in parallel, and the consistency of multiple barcode recognition results is analyzed.

7. The method for allocating and storing flood control emergency supplies according to claim 1, characterized in that, In step S4, the formation of material availability assessment results and material allocation demand results includes the following steps: S4.1 Determine the baseline inventory and safe reserve threshold for each type of flood control emergency supplies based on basic attribute data; S4.2 Utilize the updated flood control emergency supplies information database to update the inventory quantity, current occupancy status and availability status of each storage unit in real time; S4.

3. By using historical flood control emergency material dispatch data, rainfall intensity change rate and warning water level difference, construct a material demand prediction model to predict the consumption intensity of various flood control emergency materials within a preset time window, calculate the original reserve adequacy index, and perform risk correction on the original reserve adequacy index based on the predicted consumption intensity to generate the predicted and corrected reserve adequacy index. S4.

4. Based on the predicted and corrected reserve adequacy index, a comprehensive score is generated for the availability status of each type of flood control emergency supplies. ; S4.5 Based on the predicted and corrected reserve adequacy index and the comprehensive score of availability, the allocation calculation module generates the allocation quantity and priority order of various flood control emergency materials, and generates the material allocation demand results.

8. The method for allocating and storing flood control emergency supplies according to claim 7, characterized in that, In step S4.3, the original reserve adequacy index is risk-corrected based on the predicted consumption intensity to generate a predicted-corrected reserve adequacy index, including the following steps: S4.

31. Based on historical data on the allocation of flood control emergency supplies, extract the total allocation of various types of flood control emergency supplies in different flood control events. and the duration of the corresponding event By normalizing the total number of calls according to the duration of the corresponding events, the historical unit time consumption intensity of various flood control emergency materials was calculated. Furthermore, statistical aggregation of the consumption intensity per unit time was performed to construct a historical baseline for the consumption intensity per unit time of various flood control and emergency supplies. ; S4.32, Obtain the target area within the preset time window Rainfall intensity sequence within Based on rainfall intensity sequence Calculate the instantaneous rate of change of rainfall intensity The instantaneous rate of change The trend weights are mapped using a normalized function, and then used to establish a baseline for the intensity of historical unit-time consumption. A trend-weighted correction is performed, and a material demand forecasting model is generated to obtain the trend-corrected predicted consumption intensity. ; Predicting consumption intensity based on trend correction Calculate the original reserve adequacy index ; S4.33, Obtain the real-time water level of the target area. and corresponding warning water level threshold Calculate the difference between the real-time water level and the warning water level threshold. and the difference Mapped to warning water level safety margin parameter ; S4.34, Safety margin parameters based on warning water level The trend-corrected predicted consumption intensity is subjected to nonlinear risk amplification processing to obtain the risk-amplified predicted consumption intensity. ; S4.35, Predicted Consumption Intensity Based on Risk Amplification Within the preset time window Internal raw reserve adequacy index Perform risk corrections and generate a revised reserve adequacy index.

9. The method for allocating and storing flood control emergency supplies according to claim 1, characterized in that, In step S5, flood control emergency supplies that meet the allocation conditions are screened based on the material availability assessment results and material allocation demand results. Combined with the material storage structure model, the allocation path is analyzed to generate a corresponding material transportation plan, including the following steps: S5.1 Based on the results of the material availability assessment and the material allocation demand, various flood control emergency materials are screened, a list of materials that meet the allocation conditions is determined, and a set of materials that can be allocated is generated. S5.

2. Combining the material storage structure model, use the graph theory shortest path algorithm to analyze the allocation path of each type of material to be allocated in the set of available materials. S5.3 Calculate the time required for material transportation based on the allocation route analysis results, and compare the time required for material transportation with the preset emergency response time to determine whether the flood control emergency materials can reach the target location on time. S5.4 Integrate the results of comparing the available materials, allocation routes, and transportation time with the preset emergency response time to form a complete material allocation plan. S5.5 Update the material transportation plan to the flood control emergency material information database simultaneously.

10. A flood control emergency material allocation and storage management system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes a computer program to implement the method for allocating and storing flood control emergency supplies as described in any one of claims 1-9.

Images