Intelligent warehouse management and resource optimization method

Abstract

The invention discloses an intelligent warehouse management and resource optimization method. The method comprises the steps of S1, multi-source data acquisition and fusion processing; s2, performing real-time analysis and dynamic prediction on the inventory state; s3, intelligently generating storage tasks and scheduling priorities; s4, collaborative path planning and resource optimization allocation are carried out; s5, task execution monitoring and real-time dynamic re-optimization are carried out; s6, performing task closed-loop analysis and strategy optimization; and S7, performing storage layout and resource configuration simulation optimization. According to the method, the storage data is collected in real time through the Internet of Things equipment and the sensor, intelligent collaborative optimization of the storage resources and the operation tasks is achieved in combination with the dynamic analysis model, the inventory state and the equipment operation condition can be monitored in real time, the storage tasks are automatically generated and adjusted, the equipment path and resource allocation are optimized, and the operation efficiency is improved. And the warehousing operation efficiency and the resource utilization rate are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of warehouse management system technology, and more specifically to intelligent warehouse management and resource optimization methods. Background Technology

[0002] With the rapid development of the logistics industry, traditional warehouse management methods are no longer sufficient to meet the demands for efficient and precise operations. Existing technologies largely rely on manual experience or static rules for inventory management and task scheduling, leading to inefficient warehousing operations, unreasonable resource allocation, and slow response times. Especially when orders fluctuate significantly and there is a wide variety of goods, existing methods often cannot adjust strategies in real time, easily causing warehouse congestion, inventory backlogs, or idle equipment. Although some systems have introduced automated equipment, the lack of global collaboration and dynamic optimization capabilities means that overall warehouse operating costs remain high and fail to meet the flexibility and complexity requirements of modern logistics. Summary of the Invention

[0003] To address this, the present invention provides an intelligent warehouse management and resource optimization method to solve the problem that the overall warehouse operation cost remains high and is difficult to adapt to the flexibility and complexity requirements of modern logistics due to the lack of global coordination and dynamic optimization capabilities in the existing technology.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] Intelligent warehouse management and resource optimization methods include the following steps:

[0006] S1. Multi-source data acquisition and fusion processing: Through the IoT sensor network, RFID reading and writing devices and business data interfaces of the warehouse management system deployed in the warehouse area, real-time data on warehouse environment, inventory status, equipment operation and order flow are collected; the above data is preprocessed by cleaning, noise reduction and timestamp alignment through edge computing nodes, and then transmitted to the central database for standardized storage and fusion to form a unified real-time digital image of the warehouse.

[0007] S2. Real-time inventory status analysis and dynamic forecasting: Based on the inventory status data and order flow data fused from the central database, the system analyzes the inventory change rate, static storage time, and dynamic turnover rate of each storage area and each category of goods in real time. By comparing with the preset inventory threshold model, it identifies low-turnover stagnant goods, urgently needed replenishment goods below safety stock, and loosely distributed goods. At the same time, combining historical sales data and external demand forecast information, the system uses a time series forecasting model to make rolling forecasts of the inbound and outbound flow of each category of goods within a preset period in the future.

[0008] S3. Intelligent generation and priority scheduling of warehousing tasks: Based on the real-time received order instructions, the inventory anomalies identified in step S2, and the traffic prediction results, a set of work tasks is dynamically generated; task metadata describing its content, target location, associated goods, and constraints is generated for each task; the task generation logic comprehensively considers the urgency of the order, the validity period of the goods, potential conflicts in the task execution path, and the real-time load of the equipment resource pool, and adopts a hybrid decision-making mechanism based on rules and weighted scoring to calculate a dynamic priority score for each task in the task set, and performs initial sorting and scheduling accordingly;

[0009] S4. Collaborative Path Planning and Resource Optimization Allocation: For the scheduled task set generated in step S3, with the optimization goal of maximizing global operation efficiency and minimizing overall energy consumption, the resource optimization allocation engine is invoked; based on the physical operation requirements of the task and the equipment capability profile, the task is optimally matched and bound with handling equipment, picking workstations and operators, generating an executable operation instruction package containing detailed path points, operation sequence and resource assignment information.

[0010] S5. Task Execution Monitoring and Real-time Dynamic Re-optimization: The executable operation instruction package is sent to the corresponding warehouse equipment control system and personnel terminal to drive the equipment and personnel to execute the instructions. Throughout the entire task execution cycle, the actual location of the equipment, task progress, and changes in the warehouse environment are continuously monitored through the sensor network, and the actual execution data is fed back to the central database in real time. A dynamic re-optimization trigger is established. When equipment failure, significant delay in task execution, insertion of urgent high-priority orders, or sudden path congestion are detected, a local or global re-optimization process is immediately triggered. Based on the latest system status, the paths, priorities, and resource allocation of subsequent tasks are quickly replanned and adjusted, and the adjusted instructions are synchronized to the affected equipment and personnel in real time.

[0011] Preferably, the warehousing environment data includes real-time temperature and humidity, light intensity, and security monitoring video streams for each storage zone; the inventory status data includes the goods identification, quantity, storage and retrieval timestamps, and expected storage period for each storage location; the equipment operation data includes the real-time location coordinates, working status, battery level, and fault alarm information for each AGV, stacker crane, and conveyor line; and the order flow data includes real-time received inbound orders, outbound orders, inventory count instructions, and inventory transfer instructions.

[0012] Preferably, in step S2, the specific process of identifying abnormal inventory conditions further includes: for identified stagnant goods, automatically generating cross-regional or cross-warehouse transfer suggestions based on their storage location, product attributes, and associated predicted demand information; for urgently needed replenishment goods, generating suggestions for replenishing to the picking area or issuing an alert to the purchasing system; the generation of the transfer and replenishment suggestions is achieved through a lightweight machine learning model, which uses historical movement frequency of goods, seasonal factors, associated sales data, and the current overall inventory distribution of the warehouse as input features, and calculates the results with the optimization objective of minimizing the expected future stockout rate and overall handling costs.

[0013] Preferably, in step S3, the calculation of the dynamic priority score is specifically as follows: configurable weight coefficients are assigned to multiple dimensions, including the urgency of order deadline, the urgency of goods shelf life, the length of task execution path, the scarcity of required equipment, and the contribution of the task to alleviating the current warehouse congestion. The weighted sum is then obtained after normalizing the scores of each task on each dimension.

[0014] Preferably, in step S4, the resource optimization allocation engine, based on the real-time refreshed warehouse digital map, equipment status information, and task metadata, adopts an improved collaborative path planning algorithm that integrates time window constraints and dynamic traffic control strategies to plan the optimal operation path sequence with no conflict or minimum conflict cost for each available AGV and stacker crane. The improved collaborative path planning algorithm is a multi-agent path planning algorithm that combines conflict search and time window optimization. It plans smooth paths that are staggered in time and collision-free in space for multiple mobile devices by predicting and resolving conflicts on an abstract spatiotemporal map.

[0015] Preferably, it also includes S6, task closed-loop analysis and strategy optimization: after a single task or a batch of tasks is completed, the system automatically collects the actual completion time of the task, the actual energy consumption of the equipment, the path deviation and the operation accuracy data, and compares and analyzes them with the preset benchmark value or the simulation expected value; based on the comparison results, the task priority weight coefficient in step S3 and the path planning in step S4 are adaptively adjusted.

[0016] Preferably, the task closed-loop analysis in step S6 establishes a multi-dimensional performance evaluation model, which includes time and energy efficiency indicators, equipment utilization balance, and hot zone task dispersion.

[0017] Preferably, in step S1, the preprocessing performed by the edge computing node further includes: real-time analysis of data from video surveillance to identify safety risk events such as whether personnel have entered dangerous areas or whether cargo stacks have tilted, and immediately sending an alarm to the central system, while simultaneously controlling nearby equipment to enter a slow-down or paused state.

[0018] Preferably, the edge computing node can autonomously decide and control the start and stop of intelligent ventilation, air conditioning or dehumidification equipment in its area while uploading the real-time analyzed temperature and humidity data.

[0019] Preferably, it also includes S7, warehouse layout and resource allocation simulation optimization: the system periodically or as needed starts the simulation optimization module, which constructs a digital twin simulation model of warehouse operation based on long-term accumulated historical operating data, equipment failure records and future business growth forecasts; by simulating the operation effect of different storage location layout schemes, different equipment configuration quantities and scheduling strategies on the digital twin simulation model, the system automatically evaluates the cost, efficiency and robustness of each scheme, and finally outputs suggestions for adjusting the physical layout of the warehouse and optimization schemes for key equipment resource allocation.

[0020] The present invention has the following advantages: The present invention collects warehouse data in real time through IoT devices and sensors, and combines it with a dynamic analysis model to realize intelligent collaborative optimization of warehouse resources and operation tasks. Moreover, the method can monitor the inventory status and equipment operation in real time, automatically generate and adjust warehouse tasks, optimize equipment paths and resource allocation, and significantly improve warehouse operation efficiency and resource utilization.

[0021] Through anomaly detection and adaptive adjustment mechanisms, the system can respond quickly to order fluctuations or equipment failures, reducing manual intervention, lowering operating costs, and improving the accuracy and reliability of warehouse management. Attached Figure Description

[0022] To more intuitively illustrate the prior art and this application, exemplary drawings are provided below. It should be understood that the specific shapes and structures shown in the drawings should not generally be regarded as limiting conditions for implementing this application; for example, based on the technical concept disclosed in this application and the exemplary drawings, those skilled in the art are able to easily make conventional adjustments or further optimizations to the addition / reduction / classification, specific shapes, positional relationships, connection methods, size ratios, etc. of certain units (components).

[0023] Figure 1 A flowchart illustrating the intelligent warehouse management and resource optimization method provided in this application embodiment. Detailed Implementation

[0024] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these embodiments are merely for further explanation of the present invention and should not be construed as limiting the scope of protection of the present invention. Technical engineers in the field can make some non-essential improvements and adjustments to the present invention based on the above-described content. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Please see Figure 1 The intelligent warehouse management and resource optimization method includes the following steps:

[0026] S1. Multi-source data acquisition and fusion processing: Through the IoT sensor network, RFID reading and writing devices and business data interfaces of the warehouse management system deployed in the warehouse area, real-time data on warehouse environment, inventory status, equipment operation and order flow are collected; the above data is preprocessed by cleaning, noise reduction and timestamp alignment through edge computing nodes, and then transmitted to the central database for standardized storage and fusion to form a unified real-time digital image of the warehouse.

[0027] S2. Real-time inventory status analysis and dynamic prediction: Based on the inventory status data and order flow data fused from the central database, the system analyzes the inventory change rate, static storage time, and dynamic turnover rate of each storage area and each category of goods in real time. By comparing with the preset inventory threshold model, it identifies low-turnover stagnant goods, urgently needed replenishment goods below safety stock, and loosely distributed goods. At the same time, combining historical sales data and external demand forecast information, the system uses a time series forecasting model to make rolling predictions of the inbound and outbound flow of each category of goods within a preset period in the future, providing a forward-looking basis for task generation.

[0028] S3. Intelligent Generation and Priority Scheduling of Warehouse Tasks: Based on the real-time received order instructions, the inventory anomalies identified in step S2, and the traffic prediction results, a set of composite warehousing operation tasks, including shelving, picking, replenishment, inventory counting, and transfer, is dynamically generated. Metadata describing the content, target location, associated goods, and constraints of each task is generated. The task generation logic comprehensively considers the urgency of the order, the expiration date of the goods, potential conflicts in the task execution path, and the real-time load of the equipment resource pool. A hybrid decision-making mechanism based on rules and weighted scoring is adopted to calculate a dynamic priority score for each task in the task set, and to perform initial sorting and scheduling accordingly.

[0029] S4. Collaborative Path Planning and Resource Optimization Allocation: For the scheduled task set generated in step S3, the resource optimization allocation engine is invoked with the comprehensive optimization goal of maximizing global operational efficiency and minimizing overall energy consumption. This engine, based on the real-time updated warehouse digital map, equipment status information, and task metadata, adopts an improved collaborative path planning algorithm that integrates time window constraints and dynamic traffic control strategies to plan the optimal operation path sequence with no conflict or minimum conflict cost for each available AGV, stacker crane, and other mobile equipment. At the same time, based on the physical operation requirements of the task and the equipment capability profile, the task is optimally matched and bound with handling equipment, picking workstations, and operators to generate an executable operation instruction package containing detailed path points, operation sequence, and resource assignment information.

[0030] S5. Task Execution Monitoring and Real-time Dynamic Re-optimization: The executable operation instruction package is sent to the corresponding warehouse equipment control system and personnel terminal to drive the equipment and personnel to execute the instructions. Throughout the entire task execution cycle, the actual location of the equipment, task progress, and changes in the warehouse environment are continuously monitored through the sensor network, and the actual execution data is fed back to the central database in real time. A dynamic re-optimization trigger is established. When equipment failure, significant delay in task execution, insertion of urgent high-priority orders, or sudden path congestion are detected, a local or global re-optimization process is immediately triggered. Based on the latest system status, the paths, priorities, and resource allocation of subsequent tasks are quickly replanned and adjusted, and the adjusted instructions are synchronized to the affected equipment and personnel in real time.

[0031] S1. Multi-source data acquisition and fusion processing: Comprehensive collection of environmental, inventory, equipment, and order data via IoT sensors, RFID, and system interfaces. After edge computing preprocessing, this data is fused to form a unified real-time digital image of the warehouse, solving the problems of data silos and information lag in traditional systems, providing an accurate and consistent decision-making basis for global optimization. Next, S2. Real-time inventory status analysis and dynamic prediction: Not only does it monitor inventory and turnover in real time and identify abnormal inventory, but it also combines predictive models to estimate future flow, transforming traditional passive response into proactive predictive management, effectively preventing stagnation and stockouts, and optimizing inventory structure. S3. Intelligent warehouse task generation and priority scheduling: Dynamically generating composite tasks and calculating dynamic priorities, comprehensively considering multiple factors such as orders, inventory, forecasts, and resource load, ensuring the rationality and flexibility of task scheduling, and significantly enhancing the ability to cope with order fluctuations. Finally, S4. Collaborative path planning and resource optimization allocation: With global efficiency and energy consumption as comprehensive goals, advanced algorithms are used to plan conflict-free optimal paths for multiple devices and match resources, solving the problems of inter-device conflicts and high idle rates, and significantly improving the efficiency of collaborative equipment operations and resource utilization. Finally, S5, task execution monitoring, and real-time dynamic re-optimization continuously monitor task execution and trigger dynamic replanning based on unexpected situations such as failures, delays, and urgent orders, ensuring the system's stability and real-time responsiveness in the face of uncertainty. This entire approach forms a complete closed loop from data perception, intelligent decision-making, precise execution to dynamic adjustment, realizing a transformation in warehouse operations from static, experience-driven to dynamic, data-driven intelligence-driven, comprehensively improving the efficiency, flexibility, and cost-effectiveness of warehouse operations.

[0032] The warehousing environment data includes real-time temperature and humidity, light intensity, and security monitoring video streams for each storage zone; the inventory status data includes the goods identification, quantity, storage and retrieval timestamps, and expected storage period for each storage location; the equipment operation data includes the real-time location coordinates, working status, battery level, and fault alarm information for each AGV, stacker crane, and conveyor line; and the order flow data includes real-time received inbound orders, outbound orders, inventory count instructions, and inventory transfer instructions.

[0033] When obsolete items are identified, the system analyzes their location and attributes, intelligently suggesting their relocation to warehouses or areas with higher demand. This activates inventory, reduces capital tied up in inventory, and frees up prime storage space. For urgently needed items below safety stock levels, the system automatically triggers replenishment instructions to the picking area or sends an alert to the purchasing system, effectively preventing order fulfillment delays due to stockouts. This feature directly links inventory status analysis with actionable operational instructions, reducing the time and potential for errors in manual analysis and decision-making, and significantly improving inventory turnover and order fulfillment rates.

[0034] In step S2, the specific process of identifying abnormal inventory conditions further includes: for identified obsolete items, automatically generating cross-regional or cross-warehouse transfer suggestions based on their storage location, product attributes, and associated predicted demand information; for items urgently needing replenishment, generating suggestions for replenishing to the picking area or issuing an alert to the purchasing system; the generation of the transfer and replenishment suggestions is achieved through a lightweight machine learning model, which uses historical movement frequency of goods, seasonal factors, associated sales data, and the current overall inventory distribution in the warehouse as input features, and calculates the results with the optimization objective of minimizing the expected future stockout rate and overall handling costs.

[0035] The lightweight machine learning model comprehensively considers multiple dimensions of characteristics, such as the historical frequency of goods movement, seasonality, related sales, and global inventory distribution. It calculates with the goal of minimizing future stockout risks and overall handling costs. The resulting allocation or replenishment suggestions are no longer based on simple rules, but on data-driven intelligent prediction and global optimization. This allows the solution to more accurately balance inventory levels between different warehouses or regions, reduce overall logistics costs, and make replenishment timing and quantity more scientific. It avoids the high costs caused by excessive inventory or emergency transfers, thus achieving more refined management and optimization of inventory costs at a deeper level.

[0036] In step S3, the calculation of the dynamic priority score is specifically as follows: configurable weight coefficients are assigned to multiple dimensions such as the urgency of the order deadline, the urgency of the goods' shelf life, the length of the task execution path, the scarcity of required equipment, and the contribution of the task to alleviating the current warehouse congestion. The score is obtained by normalizing the scores of each task on each dimension and then summing them by weight.

[0037] By assigning weights to multiple dimensions such as time urgency, shelf life, path length, equipment scarcity, and contribution to congestion relief, and then applying weighted scoring, the system achieves refined and quantitative management of task priorities. This overcomes the limitations of traditional single-dimensional (such as "first-come, first-served") scheduling, and can intelligently adjust the task execution order based on real-time warehouse operational status (such as congestion or the busyness of certain types of equipment). For example, it might prioritize a non-urgent warehouse relocation task that can alleviate main road congestion, thereby improving overall efficiency.

[0038] In step S4, the improved cooperative path planning algorithm is a multi-agent path planning algorithm that combines conflict search and time window optimization. It plans smooth paths that are staggered in time and collision-free in space for multiple mobile devices by performing conflict prediction and resolution on an abstract spatiotemporal map.

[0039] A multi-agent path planning algorithm combining conflict search and time window optimization is employed. This algorithm extrapolates on an abstract spatiotemporal dimension, predicting in advance when and where conflicts might occur between different device paths. It proactively adjusts the departure time or path details of each device (i.e., allocating time windows) to plan a smooth set of staggered, collision-free paths. This effectively solves traffic deadlock and congestion problems in multi-device parallel operations, ensuring smooth and safe workflows. Compared to traditional methods that independently plan the shortest path for each device, this method significantly reduces idle time caused by devices waiting for each other, thereby greatly improving the overall throughput of multi-device systems in complex operational scenarios.

[0040] It also includes step S6: task closed-loop analysis and strategy optimization, which specifically includes: after a single task or a batch of tasks is completed, the system automatically collects the actual completion time of the task, the actual energy consumption of the equipment, the path deviation and the operation accuracy data, and compares and analyzes them with the preset benchmark value or the simulation expected value; based on the comparison results, the task priority weight coefficient in step S3 and the path planning cost function parameter in step S4 are adaptively fine-tuned to continuously optimize the generation and allocation strategy of subsequent tasks.

[0041] By collecting actual task execution data and comparing it with expectations, the system can objectively evaluate the effectiveness of current strategies (such as priority weights and path cost parameters). This makes the system no longer a rigid program with fixed parameters, but rather it can automatically fine-tune its internal decision parameters based on actual operational feedback. For example, if it finds that a certain piece of equipment is frequently delayed due to excessively long path planning, the system can automatically adjust the cost function, favoring the allocation of shorter paths in subsequent planning. This adaptive mechanism enables the warehouse management system to continuously adapt to changes in business models and the aging of its own equipment, maintaining optimal or near-optimal operational performance in the long term, and reducing the cost of later manual maintenance and optimization.

[0042] The task closed-loop analysis in step S6 establishes a multi-dimensional performance evaluation model. This model not only includes time and energy efficiency indicators, but also introduces indicators such as equipment utilization balance and hot zone task dispersion to measure the overall operational health of the warehouse. The system regularly generates performance analysis reports and displays the changing trends of key indicators through a visual dashboard to support management decisions.

[0043] Traditional efficiency analyses often focus solely on "completion speed" and "power consumption." This method, however, considers the overall robustness of the system, assessing the fairness of resource usage and the rationality of workload distribution. This helps prevent systemic risks and avoids problems caused by the long-term overuse of a few pieces of equipment or concentrated operations in a particular area. Consequently, it extends equipment lifespan, reduces failure rates, and balances usage intensity across different areas of the warehouse. Regularly generated performance reports and visual dashboards facilitate managers' understanding of the on-site situation.

[0044] The preprocessing performed by the edge computing node also includes: real-time analysis of data from video surveillance to identify safety risk events such as whether personnel have entered dangerous areas or whether cargo stacks have tilted, and immediately sending alarms to the central system. At the same time, it can also control nearby equipment to enter a slow-down or paused state, thereby greatly improving the safety level of warehousing operations, preventing possible personnel injuries and equipment collision accidents, and reducing operational interruptions and property losses caused by safety accidents.

[0045] Based on real-time analysis of temperature and humidity data, the edge computing node can autonomously decide and control the start and stop of intelligent ventilation, air conditioning or dehumidification equipment in its area while uploading the data, so as to achieve precise local environmental regulation and optimize the overall energy consumption of the warehouse while ensuring the storage conditions of goods.

[0046] This method allows for on-demand adjustments to areas with different storage requirements (such as refrigerated and dry areas), avoiding energy waste. Furthermore, because decisions are made instantly at edge computing nodes without waiting for cloud instructions, environmental adjustments respond faster and are better able to handle sudden environmental changes (such as temperature rises due to door opening), ensuring the quality of pharmaceuticals, food, and other goods sensitive to storage conditions and reducing the risk of damage due to improper storage.

[0047] It also includes step S7: simulation optimization of warehouse layout and resource allocation. Specifically, the system periodically or as needed starts the simulation optimization module. This module constructs a digital twin simulation model of warehouse operation based on long-term accumulated historical operating data, equipment failure records and future business growth forecasts. By simulating the operating effects of different storage location layout schemes, different equipment configuration quantities and scheduling strategies on this model, the system automatically evaluates the cost, efficiency and robustness of each scheme, and finally outputs suggestions for adjusting the physical layout of the warehouse and optimization schemes for key equipment resource allocation, which are used to guide the long-term renovation and expansion planning of the warehouse.

[0048] Step S7 uses historical data to build a high-fidelity virtual warehouse model, and simulates and tests the long-term operating effects of different layout schemes, equipment configurations and strategies on this model, providing a scientific and low-risk decision-making basis for the physical transformation and strategic investment of the warehouse.

[0049] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent warehouse management and resource optimization method, characterized in that, Includes the following steps: S1. Multi-source data acquisition and fusion processing: Through the IoT sensor network, RFID reading and writing devices and business data interfaces of the warehouse management system deployed in the warehouse area, real-time data on warehouse environment, inventory status, equipment operation and order flow are collected; the above data is preprocessed by cleaning, noise reduction and timestamp alignment through edge computing nodes, and then transmitted to the central database for standardized storage and fusion to form a unified real-time digital image of the warehouse. S2. Real-time inventory status analysis and dynamic forecasting: Based on the inventory status data and order flow data fused from the central database, the system analyzes the inventory change rate, static storage time, and dynamic turnover rate of each storage area and each category of goods in real time. By comparing with the preset inventory threshold model, it identifies low-turnover stagnant goods, urgently needed replenishment goods below safety stock, and loosely distributed goods. At the same time, combining historical sales data and external demand forecast information, the system uses a time series forecasting model to make rolling forecasts of the inbound and outbound flow of each category of goods within a preset period in the future. S3. Intelligent generation and priority scheduling of warehousing tasks: Based on the real-time received order instructions, the inventory anomalies identified in step S2, and the traffic prediction results, dynamically generate a set of work tasks; generate task metadata for each task, describing its content, target location, associated goods, and constraints. The task generation logic comprehensively considers the urgency of orders, the validity period of goods, potential conflicts in task execution paths, and the real-time load of the equipment resource pool. It adopts a hybrid decision-making mechanism based on rules and weighted scoring to calculate a dynamic priority score for each task in the task set and perform initial sorting and scheduling accordingly. S4. Collaborative Path Planning and Resource Optimization Allocation: For the scheduled task set generated in step S3, with the optimization goal of maximizing global operation efficiency and minimizing overall energy consumption, the resource optimization allocation engine is invoked; based on the physical operation requirements of the task and the equipment capability profile, the task is optimally matched and bound with handling equipment, picking workstations and operators, generating an executable operation instruction package containing detailed path points, operation sequence and resource assignment information. S5. Task Execution Monitoring and Real-time Dynamic Re-optimization: The executable operation instruction package is sent to the corresponding warehouse equipment control system and personnel terminal to drive the equipment and personnel to execute the instructions. Throughout the entire task execution cycle, the actual location of the equipment, task progress, and changes in the warehouse environment are continuously monitored through the sensor network, and the actual execution data is fed back to the central database in real time. A dynamic re-optimization trigger is established. When equipment failure, significant delay in task execution, insertion of urgent high-priority orders, or sudden path congestion are detected, a local or global re-optimization process is immediately triggered. Based on the latest system status, the paths, priorities, and resource allocation of subsequent tasks are quickly replanned and adjusted, and the adjusted instructions are synchronized to the affected equipment and personnel in real time.

2. The intelligent warehouse management and resource optimization method according to claim 1, characterized in that, The warehousing environment data includes real-time temperature and humidity, light intensity, and security monitoring video streams for each storage zone; the inventory status data includes the goods identification, quantity, storage and retrieval timestamps, and expected storage period for each storage location; the equipment operation data includes the real-time location coordinates, working status, battery level, and fault alarm information for each AGV, stacker crane, and conveyor line; and the order flow data includes real-time received inbound orders, outbound orders, inventory count instructions, and inventory transfer instructions.

3. The intelligent warehouse management and resource optimization method according to claim 2, characterized in that, In step S2, the specific process of identifying abnormal inventory conditions further includes: for identified stagnant items, automatically generating cross-regional or cross-warehouse transfer suggestions based on their storage location, product attributes, and associated predicted demand information; for items urgently needing replenishment, generating suggestions to replenish the picking area or to issue an alert to the purchasing system; the generation of the transfer and replenishment suggestions is achieved through a lightweight machine learning model, which uses historical movement frequency of goods, seasonal factors, associated sales data, and the current overall inventory distribution in the warehouse as input features, and calculates the results with the optimization objective of minimizing the expected future stockout rate and overall handling costs.

4. The intelligent warehouse management and resource optimization method according to claim 1, characterized in that, In step S3, the calculation of the dynamic priority score is specifically as follows: configurable weight coefficients are assigned to multiple dimensions, including the urgency of order deadlines, the urgency of goods shelf life, the length of task execution paths, the scarcity of required equipment, and the contribution of the task to alleviating the current warehouse congestion. The score is obtained by normalizing the scores of each task on each dimension and then summing them by weight.

5. The intelligent warehouse management and resource optimization method according to claim 4, characterized in that, In step S4, the resource optimization allocation engine, based on the real-time refreshed warehouse digital map, equipment status information, and task metadata, adopts an improved collaborative path planning algorithm that integrates time window constraints and dynamic traffic control strategies to plan the optimal operation path sequence with no conflict or minimum conflict cost for each available AGV and stacker crane. The improved collaborative path planning algorithm is a multi-agent path planning algorithm that combines conflict search and time window optimization. It plans smooth paths that are staggered in time and collision-free in space for multiple mobile devices by predicting and resolving conflicts on an abstract spatiotemporal map.

6. The intelligent warehouse management and resource optimization method according to claim 1, characterized in that, It also includes S6, task closed-loop analysis and strategy optimization: After a single task or a batch of tasks is completed, the system automatically collects the actual completion time of the task, the actual energy consumption of the equipment, the path deviation and the operation accuracy data, and compares and analyzes them with the preset benchmark value or the simulation expected value; based on the comparison results, the task priority weight coefficient in step S3 and the path planning in step S4 are adaptively adjusted.

7. The intelligent warehouse management and resource optimization method according to any one of claims 6, characterized in that, The task closed-loop analysis in step S6 establishes a multi-dimensional performance evaluation model, which includes time and energy efficiency indicators, equipment utilization balance, and hot zone task dispersion.

8. The intelligent warehouse management and resource optimization method according to claim 1, characterized in that, In step S1, the preprocessing performed by the edge computing node also includes: real-time analysis of data from video surveillance to identify safety risk events such as whether personnel have entered dangerous areas or whether cargo stacks have tilted, and immediately sending an alarm to the central system. At the same time, it can also control nearby devices to enter a slow-down or paused state.

9. The intelligent warehouse management and resource optimization method according to claim 8, characterized in that, Based on real-time analyzed temperature and humidity data, the edge computing node can autonomously decide and control the start and stop of intelligent ventilation, air conditioning, or dehumidification equipment in its area while uploading the data.

10. The intelligent warehouse management and resource optimization method according to claim 1, characterized in that, It also includes S7, warehouse layout and resource allocation simulation optimization: The system periodically or as needed starts the simulation optimization module. Based on long-term accumulated historical operating data, equipment failure records and future business growth forecasts, the simulation optimization module builds a digital twin simulation model of warehouse operation; by simulating the operation effect of different storage location layout schemes, different equipment configuration quantities and scheduling strategies on the digital twin simulation model, it automatically evaluates the cost, efficiency and robustness of each scheme, and finally outputs suggestions for adjusting the physical layout of the warehouse and optimization schemes for key equipment resource allocation.

Images