Intelligent warehouse management method and system based on multi-axis level

Abstract

The embodiment of the invention relates to the technical field of intelligent manufacturing, and discloses an intelligent warehouse management method based on a multi-axis level, and the method comprises the steps: obtaining the business operation data of each stored article; obtaining an inventory circulation time sequence according to the time information corresponding to each piece of business operation data; taking the preprocessed inventory circulation time sequence data and the business operation data as input of an LSTM model to obtain a cargo circulation trend in a set time range; calculating the three-dimensional space occupancy rate of each area of the warehouse in real time by counting the ratio of the volume of the space occupied by stored storage articles in the warehouse at the current moment to the volume of each available space of the warehouse; and determining whether to trigger a corresponding partition adjustment strategy according to the cargo circulation trend of each area and the space occupation condition of each area predicted by the LSTM, and updating the data of the corresponding area according to the partition adjustment strategy. Through the management method provided by the embodiment of the invention, the overall warehouse management efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to a multi-axis-based intelligent warehousing management method and system. Background Art

[0002] At present, traditional warehouse management mainly relies on manual operations and basic information systems, and has the following technical defects: Low space utilization: The cargo location management is extensive, relying on fixed partitions or experience-based placement, unable to dynamically optimize storage locations, and serious waste of warehouse space. Low degree of automation: Handling, sorting, and inventory rely on manual labor or single equipment (such as forklifts), which are inefficient and prone to errors, and are difficult to adapt to the high-frequency, multi-SKU modern logistics needs. Insufficient AGV / robot collaboration: Early automated equipment (such as AGV) can only perform simple path transportation, lacks intelligent scheduling, and cannot be deeply linked with inventory management and order systems. Therefore, designing an intelligent warehouse management method has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the Invention

[0003] In response to the above-mentioned defects, an embodiment of the present invention discloses an intelligent warehouse management method based on multi-axis level, which can improve the operational efficiency of warehouse management.

[0004] The first aspect of the embodiment of the present invention discloses a multi-axis-based intelligent warehouse management method, comprising:

[0005] Obtain the business operation data of each warehouse item, including warehouse entry information, warehouse transfer information, and warehouse exit information, and determine the time information corresponding to each business operation data through the timestamp module integrated in the warehouse management system;

[0006] Obtaining an inventory turnover time series based on the time information corresponding to each business operation data, wherein the inventory turnover time series includes an inbound quantity series, an outbound quantity series, and an inventory turnover rate series arranged in chronological order;

[0007] The pre-processed inventory turnover time series data and business operation data are used as inputs to the LSTM model to obtain the goods turnover trend within a set time range. The goods turnover trend includes the incoming and outgoing quantities and inventory turnover rate in the next time period.

[0008] By calculating the ratio of the volume of space occupied by stored items in the warehouse to the volume of available space in the warehouse at the current moment, the three-dimensional space occupancy rate of each area of ​​the warehouse is calculated in real time.

[0009] Whether to trigger the corresponding partition adjustment strategy is determined based on the cargo flow trend of each area and the space occupancy of each area predicted by LSTM, and the corresponding area data is updated according to the partition adjustment strategy.

[0010] As an optional implementation, in the first aspect of the embodiment of the present invention, determining whether to trigger a corresponding partition adjustment strategy based on the cargo flow trends of each region and the space occupancy of each region predicted by LSTM, and updating the corresponding regional data according to the partition adjustment strategy, includes:

[0011] If the cargo flow trend and space occupancy of the corresponding area match the first set condition, a first partition adjustment instruction is triggered to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume and spatial attributes of the corresponding area;

[0012] If the cargo flow trend and space occupancy of the corresponding area match the second set condition, a second partition adjustment instruction is triggered to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume and spatial attributes of the corresponding area;

[0013] If the cargo flow trend and space occupancy of the corresponding area match the third set condition, a third partition adjustment instruction is triggered to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume, and spatial attributes of the corresponding area;

[0014] If the cargo flow trend and space occupancy of the corresponding area match the fourth set condition, a fourth partition adjustment instruction is triggered to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume and spatial attributes of the corresponding area.

[0015] As an optional implementation, in the first aspect of the embodiment of the present invention, using the preprocessed inventory turnover time series data and business operation data as input to the LSTM model to obtain the goods turnover trend within a set time range includes:

[0016] The pre-processed data is encoded using spatiotemporal data fusion coding and event-driven coding. The spatiotemporal data fusion coding converts spatial data into fixed-length vectors through one-hot encoding or embedded encoding, and then concatenates them with time series data and business operation data according to time steps to form an input feature vector containing spatiotemporal correlations. The event-driven coding uses binary vectors to indicate whether a key warehouse event has occurred, and encodes the event's occurrence time and duration as numerical features.

[0017] The pre-processed warehouse spatiotemporal data and business data are encoded and then input into the constructed LSTM model. The LSTM model learns the spatiotemporal correlation and long-term dependency in the data through the gating mechanism and network structure to predict the cargo flow trend. The LSTM model is constructed as follows:

[0018] An LSTM model is constructed, which includes an input layer, a hidden layer and an output layer, wherein the input layer receives encoded data; the hidden layer adopts an LSTM hidden layer structure with hierarchical functional division and a bidirectional LSTM structure, wherein the hidden layer includes an upper hidden layer and a lower hidden layer, the upper hidden layer is used to mine long-term dependencies and complex patterns, and the lower hidden layer is used to learn cargo flow patterns and short-term dependencies. During the training process, the number of neurons in each hidden layer is determined by hyperparameter tuning, and residual connections are added between the hidden layers; the output layer sets the number of neurons according to the prediction target and outputs the cargo flow trend prediction result.

[0019] As an optional implementation, in the first aspect of the embodiment of the present invention, the real-time calculation of the three-dimensional space occupancy rate of each area of ​​the warehouse by calculating the ratio of the space volume occupied by stored items in the warehouse at the current moment to the volume of each available space in the warehouse includes:

[0020] Counting the volume of space currently occupied by stored items in the warehouse, where the warehouse space is divided into multiple three-dimensional grid cells, each of which represents a minimum storage unit. A three-dimensional array or hash table is established to store status information for each three-dimensional grid cell, including whether it is occupied, the type of goods stored, and the storage time.

[0021] The overall three-dimensional space occupancy rate of the warehouse is calculated according to the overall space occupancy formula, which is: Among them, V i represents the volume of the i-th occupied grid cell, V tot is the total available space volume of the warehouse, T o is the overall occupancy rate, m is the number of occupied grid cells;

[0022] The regional three-dimensional space occupancy rate is calculated according to the regional space occupancy formula, and the regional space occupancy formula is: Among them, T k is the area occupancy rate, V j represents the volume of the jth occupied grid cell, V reg It represents the total available space volume of the corresponding area, and n is the number of occupied grids in the area.

[0023] As an optional implementation, in the first aspect of the embodiment of the present invention, determining whether to trigger a corresponding partition adjustment strategy based on the cargo flow trends of each area and the space occupancy of each area predicted by LSTM includes:

[0024] When the space occupancy of the corresponding area remains unchanged, if the cargo turnover rate drops by more than the first set threshold for multiple consecutive statistical periods, the partition expansion operation is triggered; if the cargo turnover rate rises by more than the second set threshold for multiple consecutive statistical periods, the partition compression operation is triggered; the first set threshold and the second set threshold are determined based on the fluctuations in the historical turnover rate.

[0025] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the warehouse management method further includes:

[0026] Divide the storage area into a first storage area, a second storage area, and a third storage area, and arrange high-turnover items, medium-turnover items, and low-turnover items in the first storage area, the second storage area, and the third storage area according to different item arrangement ratios;

[0027] Determine the sorting time for the corresponding area based on the working parameters of the sorting robot, the type of items in each storage area, and the item configuration ratio, and determine the congestion level of the corresponding path based on the sorting time;

[0028] If the degree of congestion exceeds a set value, the item configuration ratio or the area space occupancy rate of the corresponding storage area is adjusted.

[0029] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the warehouse management method further includes:

[0030] Processing the received front-end promotion data to determine association information between the items, the front-end promotion data including promotion combination information; generating a corresponding first shift operation if the position between the corresponding items exceeds a first set value;

[0031] The co-occurrence frequency between the products is calculated based on the historical order data. If the co-occurrence frequency exceeds 80%, it is determined that there is a strong correlation between the two products. If the position between the two items in the storage area exceeds a second set value, a corresponding second shift operation is generated.

[0032] A second aspect of an embodiment of the present invention discloses a multi-axis-based intelligent warehouse management system, comprising:

[0033] Acquisition module: used to obtain the business operation data of each warehouse item, including the information of warehouse entry, warehouse transfer and warehouse exit, and determine the time information corresponding to each business operation data through the timestamp module integrated in the warehouse management system;

[0034] Calculation module: used to obtain the inventory turnover time series based on the time information corresponding to each business operation data. The inventory turnover time series includes the inventory inflow quantity series, the inventory outflow quantity series, and the inventory turnover rate series arranged in chronological order.

[0035] Model input module: This module uses pre-processed inventory turnover time series data and business operation data as input to the LSTM model to obtain the goods turnover trend within a set time range. The goods turnover trend includes the incoming and outgoing quantities and inventory turnover rate in the next time period.

[0036] Statistics module: used to calculate the three-dimensional space occupancy rate of each area of ​​the warehouse in real time by counting the ratio of the space volume occupied by the stored items in the warehouse at the current moment to the volume of each available space in the warehouse;

[0037] Update module: used to determine whether to trigger the corresponding partition adjustment strategy based on the cargo flow trend of each area and the space occupancy of each area predicted by LSTM, and update the corresponding area data according to the partition adjustment strategy.

[0038] The third aspect of an embodiment of the present invention discloses an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the multi-axis-based intelligent warehouse management method disclosed in the first aspect of the embodiment of the present invention.

[0039] A fourth aspect of an embodiment of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the multi-axis-based intelligent warehouse management method disclosed in the first aspect of an embodiment of the present invention.

[0040] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0041] The multi-axis-based intelligent warehousing method in this embodiment of the present invention determines whether to trigger a partition adjustment strategy based on LSTM-predicted cargo flow trends and the space occupancy of each area, and updates the corresponding regional data, enabling dynamic optimization of warehouse resources. If a mismatch between cargo flow trends and space occupancy is detected in a certain area, the partitions can be adjusted promptly to optimize the warehouse's storage space and cargo distribution, improving overall operational efficiency and adapting to different business needs and changing cargo flow patterns. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 1 is a flow chart of a multi-axis-based intelligent warehouse management method disclosed in an embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of the partition adjustment and update process disclosed in an embodiment of the present invention;

[0045] Figure 3 Schematic diagram of a multi-axis intelligent warehouse management system provided by an embodiment of the present invention;

[0046] Figure 4 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0048] It should be noted that the terms "first," "second," "third," "fourth," etc. in the description and claims of the present invention are used to distinguish different objects rather than to describe a specific order. The terms "including" and "having," as well as any variations thereof, in the embodiments of the present invention, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or apparatus.

[0049] Example 1

[0050] See also Figure 1 , Figure 1It is a flow chart of the intelligent warehouse management method based on multi-axis level disclosed in the embodiment of the present invention. Among them, the execution subject of the method described in the embodiment of the present invention is an execution subject composed of software and / or hardware, and the execution subject can receive relevant information by wired or / and wireless means, and can send certain instructions. Of course, it can also have certain processing functions and storage functions. The execution subject can control multiple devices, such as remote physical servers or cloud servers and related software, or it can be a local host or server and related software that performs related operations on a device placed somewhere. In some scenarios, multiple storage devices can also be controlled, and the storage devices can be placed in the same place or different places as the devices. For example Figure 1 As shown, the multi-axis-based intelligent warehouse management method includes the following steps:

[0051] S101: Acquire business operation data of each stored item, including warehouse entry information, warehouse transfer information, and warehouse exit information, and determine the time information corresponding to each business operation data through a timestamp module integrated in the warehouse management system;

[0052] S102: Obtaining an inventory turnover time series based on the time information corresponding to each business operation data, wherein the inventory turnover time series includes an inbound quantity series, an outbound quantity series, and an inventory turnover rate series arranged in chronological order;

[0053] S103: Using the pre-processed inventory turnover time series data and business operation data as input to the LSTM model to obtain a goods turnover trend within a set time range, the goods turnover trend including the incoming inventory, outgoing inventory, and inventory turnover rate within the next time period;

[0054] S104: Calculate the three-dimensional space occupancy rate of each area of ​​the warehouse in real time by calculating the ratio of the space volume occupied by the stored items in the warehouse at the current moment to the volume of each available space in the warehouse;

[0055] S105: Determine whether to trigger a corresponding partition adjustment strategy based on the cargo flow trends of each area and the space occupancy of each area predicted by LSTM, and update the corresponding area data according to the partition adjustment strategy.

[0056] The multi-axis level in this embodiment refers to the time axis, space axis, and business axis. The time axis covers real-time dynamic tracking of inventory throughout its entire lifecycle. The space axis refers to the warehouse's three-dimensional storage dimensions, enabling precise positioning within three dimensions. The business axis integrates multiple aspects of warehouse operations to form a unified management closed loop. This integration of data across multiple dimensions enables efficient unified management.

[0057] The solution of the embodiment of the present invention can accurately record the time when various business operations of stored items occur, and provide accurate time dimension data support for subsequent analysis and decision-making. Through the timestamp module integrated into the warehouse management system, the accuracy and consistency of time information can be ensured, and the errors and inconsistencies that may occur when manually recording time are avoided, which helps to achieve accurate tracing and monitoring of warehousing business operations. Sorting business operation data into inventory flow time series in chronological order can clearly show the trend of inventory changes over time. Including the incoming quantity sequence, the outgoing quantity sequence and the inventory turnover rate sequence, it can help managers intuitively understand the flow of goods in different time periods, provide a comprehensive and intuitive data basis for inventory management, and facilitate the discovery of patterns and problems in inventory management.

[0058] The solution in this embodiment of the present invention uses an LSTM model to analyze preprocessed inventory turnover time series data and business operation data, accurately predicting cargo turnover trends within a set timeframe, including incoming and outgoing inventory quantities and inventory turnover rates for the next timeframe. This helps companies proactively allocate resources, arrange personnel, and plan inventory, improving the foresight and planning of warehouse management, reducing inventory backlogs and stockouts, and lowering inventory costs.

[0059] The solution of the embodiment of the present invention calculates the three-dimensional space occupancy rate of each area of ​​the warehouse in real time, which enables managers to understand the usage of warehouse space in a timely manner. By counting the ratio of the space volume occupied by stored items to the available space volume, the weak links in space utilization and potential optimization space can be discovered, providing data support for rationally planning warehouse layout and improving space utilization, thereby reducing storage costs and improving warehouse operating efficiency. According to the LSTM predicted cargo flow trend and the space occupancy of each area, it is determined whether to trigger the partition adjustment strategy and update the corresponding area data, which can realize the dynamic optimization configuration of warehouse resources. When it is found that the cargo flow trend of certain areas does not match the space occupancy, the partition can be adjusted in time to make the storage space and cargo distribution of the warehouse more reasonable, improve the overall operational efficiency of the warehouse, and adapt to different business needs and changes in cargo flow patterns.

[0060] More preferably, determining whether to trigger a corresponding partition adjustment strategy based on the cargo flow trends of each area and the space occupancy of each area predicted by LSTM, and updating the corresponding area data according to the partition adjustment strategy, includes:

[0061] S1051: If the cargo flow trend and space occupancy of the corresponding area match the first set condition, triggering the generation of a first partition adjustment instruction to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume, and spatial attributes of the corresponding area;

[0062] S1052: If the cargo flow trend and space occupancy of the corresponding area match the second set condition, triggering the generation of a second partition adjustment instruction to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume, and spatial attributes of the corresponding area;

[0063] S1053: If the cargo flow trend and space occupancy of the corresponding area match the third set condition, triggering the generation of a third partition adjustment instruction to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume, and spatial attributes of the corresponding area;

[0064] S1054: If the cargo flow trend and space occupancy of the corresponding area match the fourth set condition, a fourth partition adjustment instruction is triggered to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume and spatial attributes of the corresponding area.

[0065] Specifically, the solution of the embodiment of the present invention dynamically adjusts the location of the storage area according to the flow trend of goods (such as the frequency of entering / exiting the warehouse), for example, high-frequency circulation items are placed closer to the entry and exit ports to reduce the length of the transportation path and improve work efficiency; the space volume is dynamically allocated to areas with different circulation rates, such as reserving more flexible space for high-turnover areas to avoid frequent warehouse transfer operations and reduce labor costs.

[0066] Specifically, the first setting condition is that the space occupancy rate is ≥80%, and the occupancy rate of the adjacent area is ≤60%, and the LSTM prediction indicator: the outbound volume in the next three days will increase by ≥30% month-on-month, or the turnover rate forecast value will increase by ≥20%, then the expansion operation will be triggered to allocate space from the adjacent low-turnover area, or enable a temporary flexible storage area; its main applicable scenario is when the current area is space-constrained and the flow of goods will accelerate in the future.

[0067] The second condition is that the space occupancy rate is ≤ 40% for more than five consecutive days, the LSTM indicator predicts the turnover rate for the next seven days is ≤ 50% of the industry average, or the incoming inventory forecast continues to decline. In this case, the storage space in the area is consolidated to free up space in the common buffer pool, or the goods are moved to high-rise shelves or edge storage areas.

[0068] The third set condition is that the space occupancy rate is ≥90% or ≤20%, and deviates from the forecast trend: at high occupancy, the forecast outbound volume drops sharply, or at low occupancy, the forecast inbound volume surges. In this case, cross-regional transfers (such as temporarily transferring goods to other warehouses) are initiated. The forecast data is manually reviewed and safety stock parameters are adjusted.

[0069] Fourth setting condition: Set a threshold for changes in cargo turnover rate based on the warehouse's historical operating data and business objectives. For example, if the cargo turnover rate in a certain area drops by more than 30% or rises by more than 50% within two consecutive statistical periods (such as one week as a statistical period), it is determined that the change in cargo turnover rate in that area exceeds the threshold, triggering the partition adjustment strategy. For example, if the turnover rate changes by more than 200% in the next cycle, the attribute conversion strategy is triggered, such as temporarily converting the general storage area into a sorting area to quickly respond to business changes. The changed attributes can be used as the basis for subsequent storage space adjustments.

[0070] More preferably, the method of using the pre-processed inventory turnover time series data and business operation data as inputs of the LSTM model to obtain the goods turnover trend within a set time range includes:

[0071] The pre-processed data is encoded using spatiotemporal data fusion coding and event-driven coding. The spatiotemporal data fusion coding converts spatial data into fixed-length vectors through one-hot encoding or embedded encoding, and then concatenates them with time series data and business operation data according to time steps to form an input feature vector containing spatiotemporal correlations. The event-driven coding uses binary vectors to indicate whether a key warehouse event has occurred, and encodes the event's occurrence time and duration as numerical features.

[0072] The pre-processed warehouse spatiotemporal data and business data are encoded and then input into the constructed LSTM model. The LSTM model learns the spatiotemporal correlation and long-term dependency in the data through the gating mechanism and network structure to predict the cargo flow trend. The LSTM model is constructed as follows:

[0073] An LSTM model is constructed, which includes an input layer, a hidden layer and an output layer, wherein the input layer receives encoded data; the hidden layer adopts an LSTM hidden layer structure with hierarchical functional division and a bidirectional LSTM structure, wherein the hidden layer includes an upper hidden layer and a lower hidden layer, the upper hidden layer is used to mine long-term dependencies and complex patterns, and the lower hidden layer is used to learn cargo flow patterns and short-term dependencies. During the training process, the number of neurons in each hidden layer is determined by hyperparameter tuning, and residual connections are added between the hidden layers; the output layer sets the number of neurons according to the prediction target and outputs the cargo flow trend prediction result.

[0074] By converting spatial data (such as storage area ID) into vectors and splicing them with time series data, the embodiment of the present invention enables the model to simultaneously learn the temporal patterns (such as seasonal fluctuations) and spatial features (such as differences in flow between different regions) of cargo flows, thereby improving the integrity of the prediction dimension. One-hot encoding or embedded encoding converts high-dimensional sparse spatial data into low-dimensional dense vectors, reducing the dimensionality disaster while retaining semantic information. For example, the embedded vectors of adjacent storage areas are closer in space, which helps the model capture spatial correlation. Splicing data by time step ensures the alignment of different types of features in the time dimension. For example, the inventory volume at a certain moment is associated with the corresponding storage area ID feature to strengthen the spatiotemporal causal relationship modeling.

[0075] In the embodiment of the present invention, key events (such as equipment failures and promotional activities) are represented by binary vectors. The model can identify the impact of unconventional factors on the flow of goods, such as predicting a surge in outbound shipments after a promotional event, thereby improving anti-interference capabilities. Event time and duration are encoded as numerical features, enabling the model to learn the temporal impact patterns of events, such as predicting the length of the inventory recovery period after the end of a promotional activity and optimizing replenishment strategies. Event encoding provides a basis for explainable predictions, such as quantifying the degree of impact of different types of events on the flow of goods by analyzing event feature weights.

[0076] The LSTM model of the embodiment of the present invention realizes multi-scale dependency learning by setting up a double hidden layer. The upper hidden layer: predicts trend changes by capturing long-term dependencies (such as quarterly inventory cycles), such as identifying slow-moving products in advance to reduce inventory backlogs. The lower hidden layer: focuses on short-term dependencies (such as daily / weekly fluctuations), accurately predicts recent inbound and outbound peaks, and optimizes personnel scheduling and equipment scheduling. The bidirectional LSTM learns historical data and future context simultaneously, such as optimizing current inventory allocation through order information for the next week, and improving the forward-looking nature of predictions. The residual connection alleviates the gradient vanishing problem, accelerates model convergence and improves the training efficiency of deep networks, enabling the model to learn more complex flow patterns.

[0077] The embodiments of the present invention dynamically determine the number of neurons in each layer through hyperparameter tuning, enabling the model to adapt to the data characteristics of warehouses of different sizes. For example, small warehouses can use a more streamlined network structure. The layered design makes the model learning process more transparent. For example, by analyzing the activation values ​​of the upper and lower hidden layers, the factors affecting long-term trends and short-term fluctuations can be explained respectively. The model output can be directly connected to the WMS system to generate executable scheduling instructions (such as replenishment plans and storage location allocation), shortening the decision-making chain.

[0078] More preferably, the three-dimensional space occupancy rate of each area of ​​the warehouse is calculated in real time by counting the ratio of the space volume occupied by the stored items in the warehouse at the current moment to the volume of each available space in the warehouse, including:

[0079] Counting the volume of space currently occupied by stored items in the warehouse, where the warehouse space is divided into multiple three-dimensional grid cells, each of which represents a minimum storage unit. A three-dimensional array or hash table is established to store status information for each three-dimensional grid cell, including whether it is occupied, the type of goods stored, and the storage time.

[0080] The overall three-dimensional space occupancy rate of the warehouse is calculated according to the overall space occupancy formula, which is: Among them, V i represents the volume of the i-th occupied grid cell, V tot is the total available space volume of the warehouse, T o is the overall occupancy rate, m is the number of occupied grid cells;

[0081] The regional three-dimensional space occupancy rate is calculated according to the regional space occupancy formula, and the regional space occupancy formula is: Among them, T k is the area occupancy rate, V j represents the volume of the jth occupied grid cell, V reg It represents the total available space volume of the corresponding area, and n is the number of occupied grids in the area.

[0082] This embodiment of the present invention divides warehouse space into minimum storage units (three-dimensional grids), achieving millimeter-level spatial precision management. Compared to traditional two-dimensional or extensive spatial calculations, this method can improve the accuracy of space utilization assessment. By storing grid status (occupied / free, cargo type, storage time) in a three-dimensional array or hash table, it supports real-time query of spatial status at any location, providing a data foundation for dynamic storage allocation. Grid units can be resized according to actual storage needs (e.g., large grids for large-item areas and small grids for small-item areas), flexibly adapting to a variety of different product types.

[0083] Combined with the storage time dimension, "long-term occupied areas" (such as areas where unsaleable goods accumulate) and "high-frequency circulation areas" are identified to guide warehouse adjustment strategies (such as moving high-frequency goods to areas near the entrance and exit). Future storage needs are predicted based on real-time occupancy rates. For example, when the overall occupancy rate exceeds 85% for three consecutive days, the expansion plan or adjustment of the procurement plan is automatically triggered to reduce temporary rental costs. Combined with cargo type information, intelligent zoning is achieved (such as storing heavy goods in the bottom grid) to improve space stability and picking efficiency. When planning the optimal path for automated equipment such as AGV, consider the three-dimensional space occupancy to avoid invalid paths (such as crossing high-occupancy areas)

[0084] The 3D grid status can be synchronized with the WMS system in real time, automatically updating the corresponding grid status when goods enter and exit the warehouse, ensuring the timeliness of spatial data. Occupancy data can be connected to the TMS (Transportation Management System) to optimize vehicle scheduling based on warehouse space status (for example, prioritizing outbound transportation when fully loaded), achieving full supply chain collaboration. It also supports the display of 3D space occupancy heat maps, visually displaying the usage of each warehouse area and assisting managers in making space decisions.

[0085] More preferably, determining whether to trigger a corresponding partition adjustment strategy based on the cargo flow trends of each area and the space occupancy of each area predicted by LSTM includes:

[0086] When the space occupancy of the corresponding area remains unchanged, if the cargo turnover rate drops by more than the first set threshold for multiple consecutive statistical periods, the partition expansion operation is triggered; if the cargo turnover rate rises by more than the second set threshold for multiple consecutive statistical periods, the partition compression operation is triggered; the first set threshold and the second set threshold are determined based on the fluctuations in the historical turnover rate.

[0087] When the turnover rate of a certain area continues to rise and exceeds a threshold (such as a year-on-year increase of 20%), its space volume is automatically compressed, and the released space is allocated to more urgently needed high-flow areas, thereby improving global throughput efficiency. When the turnover rate of a certain area continues to fall and exceeds a threshold, an expansion operation is triggered to avoid operational bottlenecks caused by insufficient space. For example, more storage space is reserved for slow-moving products to reduce frequent warehouse transfers. By dynamically adjusting the partition volume, elastic scaling of storage space is achieved. For example, more space is allocated to hot-selling areas during promotional seasons, and normal configuration is restored during non-promotional periods.

[0088] In this embodiment of the present invention, a "multiple consecutive statistical periods" judgment mechanism filters out short-term fluctuations and captures true trend changes, for example, identifying seasonal slowdowns rather than temporary order fluctuations. This adjusted data is fed back into the LSTM model to further optimize turnover rate prediction accuracy, forming a closed loop of intelligent decision-making called "prediction-adjustment-verification." This embodiment of the present invention proactively adjusts zoning by predicting turnover rate changes, reducing frequent cargo handling due to space shortages. For example, if turnover in a particular area decreases, capacity can be expanded in advance to avoid subsequent emergency warehouse transfers.

[0089] More preferably, the warehouse management method further includes:

[0090] Divide the storage area into a first storage area, a second storage area, and a third storage area, and arrange high-turnover items, medium-turnover items, and low-turnover items in the first storage area, the second storage area, and the third storage area according to different item arrangement ratios;

[0091] Determine the sorting time for the corresponding area based on the working parameters of the sorting robot, the type of items in each storage area, and the item configuration ratio, and determine the congestion level of the corresponding path based on the sorting time;

[0092] If the degree of congestion exceeds a set value, the item configuration ratio or the area space occupancy rate of the corresponding storage area is adjusted.

[0093] Existing solutions either completely segment the areas, with each area storing only one type of item, or integrate multiple functions into transport carts, both of which increase costs. In this embodiment, dynamic shelf area adjustment allows for appropriate item storage, enabling more efficient warehouse management. Because a three-dimensional grid is used to manage each storage area, subsequent refined management becomes possible.

[0094] During specific implementation, sometimes one of the reasons for shelf congestion is excessive density. Therefore, excessive density can cause road congestion. Assisting in determining the optimal route based on different densities and the status of picking up goods is also a specific method. When the space occupancy rate is in a certain range and the number of intervals reaches a certain value, a problem will arise. If the space occupancy rate is too high, congestion will occur; if it is too low, faster intervals cannot be achieved. The number of high-frequency pick-ups should also be large. How to deal with it well and balance the number of pick-ups, transportation quantity, and transportation routes? The space occupancy rate predicted by LSTM can be used to achieve early warning of the route, which can help improve warehouse operation efficiency.

[0095] High-turnover, low-turnover, and medium-turnover product combinations can be placed on the same shelf. When making specific arrangements, they can be combined according to the set ratio; a first storage area, a second storage area, and a third storage area are generated for the area; specifically, the ratio of the first storage area is 7:2:1; the second storage area is: 2:7:1; the third storage area is: 1:2:7; through the above method, on the one hand, congestion can be prevented, and on the other hand, the efficiency of entering and leaving the warehouse can be improved; the solution of the embodiment of the present invention can set different management goals in different time periods, and can set high mobility (high operating speed of the transport trolley, with the highest number of outbound shipments as the goal) or high storage (with the goal of improving the overall warehouse storage rate).

[0096] The solution of the present invention achieves a precise balance between space and efficiency. High-turnover items are prioritized (e.g., the first storage area is allocated 70% for high-turnover items), shortening the sorting path in the core area and improving outbound delivery speed. Low-frequency items are stored in a decentralized manner (e.g., the third storage area is allocated 70% for low-turnover items), avoiding occupying space along the main path while maintaining storage capacity utilization. The degree of path congestion can be inferred by inversely linking it to operational efficiency (e.g., a 20% increase in sorting time triggers a congestion threshold).

[0097] The embodiments of the present invention can also provide real-time congestion warning and congestion prediction: based on the operating parameters of the sorting robot (such as driving speed and task completion time), the degree of congestion is quantified, and bottlenecks are discovered earlier than manually; if the congestion in a certain area exceeds a threshold (such as sorting time > 120% of the set value), the system automatically reduces the proportion of high-turnover items in the area or expands the capacity of adjacent areas.

[0098] If the first storage area is congested during a promotional event, the system adjusts the ratio of high-turnover items from 7:2:1 to 6:3:1, diverting them to the second storage area, alleviating congestion by 15%. Sorting robots are prioritized based on the type of items in each area (high, medium, or low turnover), ensuring the fastest response times in high-turnover areas. By adjusting the allocation ratio, robots are prevented from being concentrated in a single area (for example, the task density of robots in low-turnover areas is reduced by 30%). During promotional events, the ratio of high-turnover items in the first storage area is increased to 8:1:1, and robot speed is increased by 20%, with outbound volume being the priority. A balanced ratio is restored daily, prioritizing storage capacity utilization.

[0099] More preferably, the warehouse management method further includes:

[0100] Processing the received front-end promotion data to determine association information between the items, the front-end promotion data including promotion combination information; generating a corresponding first shift operation if the position between the corresponding items exceeds a first set value;

[0101] The co-occurrence frequency between the products is calculated based on the historical order data. If the co-occurrence frequency exceeds 80%, it is determined that there is a strong correlation between the two products. If the position between the two items in the storage area exceeds a second set value, a corresponding second shift operation is generated.

[0102] Specifically, during a major sales event, there are sales strategies that require certain items to be updated. This can be achieved by splitting business-side data to update the corresponding storage locations. For example, products that were previously stored in different areas can be relocated to the same area, significantly improving subsequent transportation efficiency.

[0103] By processing front-end promotional data (such as "buy A get B free" and "AB set"), the embodiment of the present invention allows the system to automatically identify related products in the promotional combination. For example, when an "A+B" promotion is detected, the system can quickly locate the storage locations of the two products.

[0104] When the distance between related product locations exceeds a threshold (e.g., the first set value is 5 meters), the first shift operation is triggered, adjusting the product to an adjacent location (such as the same shelf or adjacent aisle), reducing the picking path length. Compared to manual adjustments, automated shifting operations can complete inventory optimization before promotional activities go online, supporting rapid response to market changes.

[0105] Calculate product co-occurrence frequencies based on historical order data (e.g., pairs of products that appear simultaneously in more than 80% of orders) to identify strong correlations in users' actual purchasing behavior. For low-frequency but strongly correlated products (e.g., specific accessories), optimize storage locations through co-occurrence analysis to avoid inefficient picking due to dispersed storage.

[0106] When the distance between the storage locations of strongly related products exceeds a threshold (such as the second set value is set to 10 meters), the second shift operation is triggered to store them in the same area (such as the related product area) to improve order picking efficiency.

[0107] The embodiment of the present invention uses co-occurrence analysis to find slow-moving products among related products (such as an accessory with a high co-occurrence rate but low sales), to provide early warning of inventory backlog risks and guide purchasing decisions. For strongly related products, the safety stock level is dynamically adjusted based on promotional data and co-occurrence frequency. For example, gifts in promotional combinations can reduce safety stock and reduce capital occupation. By clustering related products, the storage space fragmentation is reduced.

[0108] Specifically, historical order data shows that liquid foundation X and loose powder Y co-occur in 85% of orders (exceeding the 80% threshold), making them strongly related products. Currently, liquid foundation X is in Area B on the first floor (a high-turnover area), while loose powder Y is in Area D on the third floor (a low-turnover area), a distance of 18 meters (exceeding the second set value of 10 meters). This triggers the second relocation operation: loose powder Y is relocated to Area C on the first floor (on the same level as liquid foundation X, on an adjacent shelf). This adjustment shortens the distance to 3 meters.

[0109] The multi-axis-based intelligent warehousing method in this embodiment of the present invention determines whether to trigger a partition adjustment strategy based on LSTM-predicted cargo flow trends and the space occupancy of each area, and updates the corresponding regional data, enabling dynamic optimization of warehouse resources. If a mismatch between cargo flow trends and space occupancy is detected in a certain area, the partitions can be adjusted promptly to optimize the warehouse's storage space and cargo distribution, improving overall operational efficiency and adapting to different business needs and changing cargo flow patterns.

[0110] Example 2

[0111] See also Figure 3 , Figure 3 This is a schematic diagram of the structure of the multi-axis intelligent warehouse management system disclosed in the embodiment of the present invention. Figure 3 As shown, the multi-axis-based intelligent warehouse management system may include:

[0112] Acquisition module 21: used to acquire business operation data of each stored item, including storage information, transfer information and outbound information, and determine the time information corresponding to each business operation data through the timestamp module integrated in the warehouse management system;

[0113] Calculation module 22: used to obtain an inventory turnover time series based on the time information corresponding to each business operation data, wherein the inventory turnover time series includes an inbound quantity series, an outbound quantity series, and an inventory turnover rate series arranged in chronological order;

[0114] Model input module 23: used to use the pre-processed inventory flow time series data and business operation data as input to the LSTM model to obtain the goods flow trend within a set time range, the goods flow trend including the incoming and outgoing quantities and inventory turnover rate in the next time period;

[0115] Statistics module 24: used to calculate the three-dimensional space occupancy rate of each area of ​​the warehouse in real time by counting the ratio of the space volume occupied by the stored items in the warehouse at the current moment to the volume of each available space in the warehouse;

[0116] Update module 25: used to determine whether to trigger the corresponding partition adjustment strategy based on the cargo flow trend of each area and the space occupancy of each area predicted by LSTM, and update the corresponding area data according to the partition adjustment strategy.

[0117] The multi-axis-based intelligent warehousing method in this embodiment of the present invention determines whether to trigger a partition adjustment strategy based on LSTM-predicted cargo flow trends and the space occupancy of each area, and updates the corresponding regional data, enabling dynamic optimization of warehouse resources. If a mismatch between cargo flow trends and space occupancy is detected in a certain area, the partitions can be adjusted promptly to optimize the warehouse's storage space and cargo distribution, improving overall operational efficiency and adapting to different business needs and changing cargo flow patterns.

[0118] Example 3

[0119] See also Figure 4 , Figure 4This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention. The electronic device can be a computer, a server, etc. Of course, in certain circumstances, it can also be a smart device such as a mobile phone, a tablet computer, a monitoring terminal, and an image acquisition device with processing functions. Figure 4 As shown, the electronic device may include:

[0120] A memory 510 storing executable program code;

[0121] a processor 520 coupled to the memory 510;

[0122] The processor 520 calls the executable program code stored in the memory 510 to execute part or all of the steps in the multi-axis-based intelligent warehouse management method in the first embodiment.

[0123] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute part or all of the steps in the multi-axis-based intelligent warehouse management method in embodiment one.

[0124] An embodiment of the present invention further discloses a computer program product, wherein when the computer program product is run on a computer, the computer is enabled to execute part or all of the steps in the multi-axis-based intelligent warehouse management method in embodiment one.

[0125] An embodiment of the present invention also discloses an application publishing platform, wherein the application publishing platform is used to publish a computer program product, wherein when the computer program product runs on a computer, the computer executes some or all of the steps in the multi-axis-based intelligent warehouse management method in Example 1.

[0126] In various embodiments of the present invention, it should be understood that the size of the serial numbers of the processes does not necessarily mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0127] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one place or distributed across multiple network units. Some or all of these units may be selected based on actual needs to achieve the objectives of this embodiment.

[0128] In addition, the functional units in the embodiments of the present invention may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The integrated unit may be implemented in the form of hardware or software functional units.

[0129] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible memory. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a memory and includes several requests for causing a computer device (which can be a personal computer, server, or network device, specifically a processor in the computer device) to execute some or all of the steps of the method described in each embodiment of the present invention.

[0130] In the embodiments provided herein, it should be understood that "B corresponding to A" means that B is associated with A and B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B based solely on A; B can also be determined based on A and / or other information.

[0131] Those skilled in the art will appreciate that some or all of the steps in the various methods of the embodiments may be accomplished by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, including a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0132] The above is a detailed introduction to the multi-axis intelligent warehouse management method, system, electronic device and storage medium disclosed in the embodiments of the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A multi-axis-based intelligent warehouse management method, characterized in that: include: Obtain the business operation data of each warehouse item, including warehouse entry information, warehouse transfer information, and warehouse exit information, and determine the time information corresponding to each business operation data through the timestamp module integrated in the warehouse management system; Obtaining an inventory turnover time series based on the time information corresponding to each business operation data, wherein the inventory turnover time series includes an inbound quantity series, an outbound quantity series, and an inventory turnover rate series arranged in chronological order; The pre-processed inventory turnover time series data and business operation data are used as inputs to the LSTM model to obtain the goods turnover trend within a set time range. The goods turnover trend includes the incoming and outgoing quantities and inventory turnover rate in the next time period. By calculating the ratio of the volume of space occupied by stored items in the warehouse to the volume of available space in the warehouse at the current moment, the three-dimensional space occupancy rate of each area of ​​the warehouse is calculated in real time. Whether to trigger the corresponding partition adjustment strategy is determined based on the cargo flow trend of each area and the space occupancy of each area predicted by LSTM, and the corresponding area data is updated according to the partition adjustment strategy.

2. The multi-axis intelligent warehouse management method according to claim 1, characterized in that: The method of determining whether to trigger a corresponding partition adjustment strategy based on the cargo flow trends of each area and the space occupancy of each area predicted by LSTM, and updating the corresponding area data according to the partition adjustment strategy, includes: If the cargo flow trend and space occupancy of the corresponding area match the first set condition, a first partition adjustment instruction is triggered to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume and spatial attributes of the corresponding area; If the cargo flow trend and space occupancy of the corresponding area match the second set condition, a second partition adjustment instruction is triggered to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume and spatial attributes of the corresponding area; If the cargo flow trend and space occupancy of the corresponding area match the third set condition, a third partition adjustment instruction is triggered to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume, and spatial attributes of the corresponding area; If the cargo flow trend and space occupancy of the corresponding area match the fourth set condition, a fourth partition adjustment instruction is triggered to perform a space adjustment operation on the corresponding area to update the spatial position, spatial volume and spatial attributes of the corresponding area.

3. The multi-axis intelligent warehouse management method according to claim 1, characterized in that: The pre-processed inventory turnover time series data and business operation data are used as inputs to the LSTM model to obtain the goods turnover trend within a set time range, including: The pre-processed data is encoded using spatiotemporal data fusion coding and event-driven coding. The spatiotemporal data fusion coding converts spatial data into fixed-length vectors through one-hot encoding or embedded encoding, and then concatenates them with time series data and business operation data according to time steps to form an input feature vector containing spatiotemporal correlations. The event-driven coding uses binary vectors to indicate whether a key warehouse event has occurred, and encodes the event's occurrence time and duration as numerical features. The pre-processed warehouse spatiotemporal data and business data are encoded and then input into the constructed LSTM model. The LSTM model learns the spatiotemporal correlation and long-term dependency in the data through the gating mechanism and network structure to predict the cargo flow trend. The LSTM model is constructed as follows: An LSTM model is constructed, which includes an input layer, a hidden layer and an output layer, wherein the input layer receives encoded data; the hidden layer adopts an LSTM hidden layer structure with hierarchical functional division and a bidirectional LSTM structure, wherein the hidden layer includes an upper hidden layer and a lower hidden layer, the upper hidden layer is used to mine long-term dependencies and complex patterns, and the lower hidden layer is used to learn cargo flow patterns and short-term dependencies. During the training process, the number of neurons in each hidden layer is determined by hyperparameter tuning, and residual connections are added between the hidden layers; the output layer sets the number of neurons according to the prediction target and outputs the cargo flow trend prediction result.

4. The multi-axis intelligent warehouse management method according to claim 1, characterized in that: The three-dimensional space occupancy rate of each area of ​​the warehouse is calculated in real time by counting the ratio of the space volume occupied by the stored items in the warehouse at the current moment to the volume of each available space in the warehouse, including: Counting the volume of space currently occupied by stored items in the warehouse, where the warehouse space is divided into multiple three-dimensional grid cells, each of which represents a minimum storage unit. A three-dimensional array or hash table is established to store status information for each three-dimensional grid cell, including whether it is occupied, the type of goods stored, and the storage time. The overall three-dimensional space occupancy rate of the warehouse is calculated according to the overall space occupancy formula, which is: Among them, V i represents the volume of the i-th occupied grid cell, V tot is the total available space volume of the warehouse, T o is the overall occupancy rate, m is the number of occupied grid cells; The regional three-dimensional space occupancy rate is calculated according to the regional space occupancy formula, and the regional space occupancy formula is: Among them, T k is the area occupancy rate, V j represents the volume of the jth occupied grid cell, V reg It represents the total available space volume of the corresponding area, and n is the number of occupied grids in the area.

5. The multi-axis intelligent warehouse management method according to claim 1, characterized in that: The method of determining whether to trigger a corresponding partition adjustment strategy based on the cargo flow trends of each area and the space occupancy of each area predicted by LSTM includes: When the space occupancy of the corresponding area remains unchanged, if the cargo turnover rate drops by more than the first set threshold for multiple consecutive statistical periods, the partition expansion operation is triggered; if the cargo turnover rate rises by more than the second set threshold for multiple consecutive statistical periods, the partition compression operation is triggered; the first set threshold and the second set threshold are determined based on the fluctuations in the historical turnover rate.

6. The multi-axis intelligent warehouse management method according to claim 1, characterized in that: The warehouse management method further includes: Divide the storage area into a first storage area, a second storage area, and a third storage area, and arrange high-turnover items, medium-turnover items, and low-turnover items in the first storage area, the second storage area, and the third storage area according to different item arrangement ratios; Determine the sorting time for the corresponding area based on the working parameters of the sorting robot, the type of items in each storage area, and the item configuration ratio, and determine the congestion level of the corresponding path based on the sorting time; If the degree of congestion exceeds a set value, the item configuration ratio or the area space occupancy rate of the corresponding storage area is adjusted.

7. The multi-axis intelligent warehouse management method according to claim 6, characterized in that: The warehouse management method further includes: Processing the received front-end promotion data to determine association information between the items, the front-end promotion data including promotion combination information; generating a corresponding first shift operation if the position between the corresponding items exceeds a first set value; The co-occurrence frequency between the products is calculated based on the historical order data. If the co-occurrence frequency exceeds 80%, it is determined that there is a strong correlation between the two products. If the position between the two items in the storage area exceeds a second set value, a corresponding second shift operation is generated.

8. An intelligent warehouse management system based on multi-axis level, characterized in that: include: Acquisition module: used to obtain the business operation data of each warehouse item, including the information of warehouse entry, warehouse transfer and warehouse exit, and determine the time information corresponding to each business operation data through the timestamp module integrated in the warehouse management system; Calculation module: used to obtain the inventory turnover time series based on the time information corresponding to each business operation data. The inventory turnover time series includes the inventory inflow quantity series, the inventory outflow quantity series, and the inventory turnover rate series arranged in chronological order. Model input module: This module uses pre-processed inventory turnover time series data and business operation data as input to the LSTM model to obtain the goods turnover trend within a set time range. The goods turnover trend includes the incoming and outgoing quantities and inventory turnover rate in the next time period. Statistics module: used to calculate the three-dimensional space occupancy rate of each area of ​​the warehouse in real time by counting the ratio of the space volume occupied by the stored items in the warehouse at the current moment to the volume of each available space in the warehouse; Update module: used to determine whether to trigger the corresponding partition adjustment strategy based on the cargo flow trend of each area and the space occupancy of each area predicted by LSTM, and update the corresponding area data according to the partition adjustment strategy.

9. An electronic device, characterized in that: include: a memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the multi-axis-based intelligent warehouse management method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program enables a computer to execute the multi-axis-based intelligent warehouse management method according to any one of claims 1 to 7.

Images