Cross-border e-commerce intelligent warehouse logistics optimization scheduling method and system

Abstract

The invention relates to the technical field of machine learning, and discloses a cross-border e-commerce intelligent warehouse logistics optimization scheduling method and system, and the method comprises the steps: constructing a multi-dimensional dynamic data pool of a warehouse; analyzing customer values, commodity shortage degrees and timeliness urgency degrees of order commodities corresponding to the historical order data, and determining dynamic priority labels of the order commodities; constructing a three-dimensional digital twinborn model of the warehouse, and performing state intelligent mapping on the three-dimensional digital twinborn model to obtain a state mapping twinborn model; the method comprises the following steps: establishing a multi-target mixed integer programming model of a warehouse, outputting warehouse logistics optimization parameters of the warehouse by using the multi-target mixed integer programming model, analyzing abnormal nodes under the warehouse logistics optimization parameters, and analyzing real-time risk values of the abnormal nodes to adjust the warehouse logistics optimization parameters to obtain target warehouse logistics optimization parameters. And warehouse logistics optimization scheduling of the warehouse is executed based on the target warehouse logistics optimization parameters. According to the invention, the operation efficiency of cross-border e-commerce warehouse logistics can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to a cross-border e-commerce intelligent warehouse logistics optimization scheduling method and system, belonging to the technical field of machine learning. BACKGROUND

[0002] Warehouse logistics optimization scheduling refers to a process of making global collaborative decisions on warehouse operations (such as warehousing, storage, picking, packaging, sorting, and warehouse-out) and logistics transportation (such as vehicle routing, loading optimization, and cross-border customs coordination) through intelligent technology means, to achieve optimal operation under the constraints of time efficiency, cost, resource utilization, and other multi-objective constraints, and to achieve a series of optimization goals such as efficient utilization of warehouse resources, minimization of logistics operation cost, maximization of order fulfillment efficiency, and improvement of customer satisfaction.

[0003] The traditional way of cross-border e-commerce intelligent warehouse logistics optimization scheduling mainly relies on manual experience and static rules for decision-making, for example, workers develop fixed picking paths and transportation plans based on historical experience, or use simple Excel spreadsheets for inventory management and order allocation. This approach often lacks flexibility and is difficult to cope with fluctuations in order volume, changes in product categories, and the complex logistics environment of cross-border e-commerce, resulting in low operational efficiency of warehouse logistics. SUMMARY

[0004] The present application provides a cross-border e-commerce intelligent warehouse logistics optimization scheduling method and system, which aims to improve the operational efficiency of cross-border e-commerce warehouse logistics.

[0005] To achieve the above-mentioned purpose, the present application provides a cross-border e-commerce intelligent warehouse logistics optimization scheduling method, which comprises: Collecting warehouse logistics data and historical order data of a warehouse, and based on the warehouse logistics data, constructing a multi-dimensional dynamic data pool of the warehouse; According to the historical order data, analyzing the customer value, product shortage degree, and time urgency of the order goods corresponding to the historical order data, and determining the dynamic priority label of the order goods according to the customer value, product shortage degree, and time urgency; Constructing a three-dimensional digital twin model of the warehouse, and intelligently mapping the state of the three-dimensional digital twin model based on the multi-dimensional dynamic data pool to obtain a state mapping twin model; Establishing multiple optimization objectives of the warehouse to establish a multi-objective mixed integer programming model of the warehouse under a predetermined multi-constraint condition, and based on the dynamic priority label and the state mapping twin model, outputting the warehouse logistics optimization parameters of the warehouse using the multi-objective mixed integer programming model, wherein the warehouse logistics optimization parameters include transportation optimization mode and transportation optimization path; analyzing the abnormal node under the warehouse logistics optimization parameter, to analyze a real-time risk value of the abnormal node, adjusting the warehouse logistics optimization parameter based on the real-time risk value, obtaining a target warehouse logistics optimization parameter, and performing warehouse logistics optimization scheduling of the warehouse based on the target warehouse logistics optimization parameter.

[0006] Optionally, the constructing the multi-dimensional dynamic data pool of the warehouse based on the warehouse logistics data comprises: preprocessing the warehouse logistics data to obtain preprocessed warehouse logistics data; establishing a snowflake storage structure of the preprocessed warehouse logistics data; mining derivative indexes of the preprocessed warehouse logistics data; associating the derivative indexes with the preprocessed warehouse logistics data to obtain an association data set; storing the association data set into the snowflake storage structure to obtain the multi-dimensional dynamic data pool of the warehouse.

[0007] Optionally, the analyzing the customer value, the commodity tightness and the time urgency of the order commodity corresponding to the historical order data comprises: removing invalid orders in the historical order data to obtain target orders; extracting customer dimension features, commodity dimension features and cross-border dimension features of the target orders; calculating order customer scores of order customers corresponding to the order commodities based on the customer dimension features to determine the customer value of the order customers; analyzing the demand degree of the order commodities according to the commodity dimension features to determine the commodity tightness of the order commodities; determining the time urgency level of the order commodities through the cross-border dimension features; calculating the time urgency of the order commodities according to the time urgency level.

[0008] Optionally, the determining the dynamic priority label of the order commodity according to the customer value, the commodity tightness and the time urgency comprises: normalizing the customer value, the commodity tightness and the time urgency to obtain normalized customer value, normalized commodity tightness and normalized time urgency; determining customer value weights, commodity tightness weights and time urgency weights of the normalized customer value, the normalized commodity tightness and the normalized time urgency; Based on the normalized customer value, the normalized commodity shortage degree, the normalized time urgency degree, the customer value weight, the commodity shortage degree weight, and the time urgency degree weight, a priority coefficient of the order commodity is calculated by using the following formula:

[0009] wherein, represents the priority coefficient of the order commodity, represents the normalized customer value, represents the customer value weight of the normalized customer value, represents the normalized commodity shortage degree, represents the commodity shortage degree weight of the normalized commodity shortage degree, represents the time urgency degree, represents the time urgency degree weight of the time urgency degree, represents an exponential function; According to the priority coefficient, a dynamic priority label of the order commodity is determined.

[0010] Optionally, the three-dimensional digital twin model of the warehouse is constructed, including: identifying the geometric structure, the warehouse equipment, and the warehouse environment of the warehouse; performing entity modeling on the geometric structure, the warehouse equipment, and the warehouse environment to obtain a warehouse physical model; analyzing the logistics behavior, the equipment behavior, and the personnel behavior of the warehouse; performing behavior modeling on the logistics behavior, the equipment behavior, and the personnel behavior to obtain a warehouse behavior model; integrating the warehouse physical model and the warehouse behavior model to obtain the three-dimensional digital twin model of the warehouse.

[0011] Optionally, the multi-objective mixed integer programming model of the warehouse under the preset multi-constraint condition is established, including: based on the multi-constraint condition, determining the warehouse priority weight, the channel cost, and the channel standard transportation time urgency of the warehouse; establishing a warehouse operation efficiency function of the warehouse according to the warehouse priority weight; establishing a transportation cost function of the warehouse through the channel cost; based on the channel standard transportation time urgency, a time urgency compliance rate function of the warehouse is constructed by using the following formula, wherein the time urgency compliance rate function:

[0012] wherein, represents the time urgency compliance rate function, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th warehouse of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes a binary variable, the order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, denotes the i-th order of the warehouse, According to the warehouse operation efficiency function, the transportation cost function and the time limit compliance rate function, a multi-objective function of the warehouse is constructed. Through the multi-objective function, a multi-objective mixed integer programming model of the warehouse under a preset multi-constraint condition is established.

[0013] Optionally, the multi-objective function of the warehouse is constructed according to the warehouse operation efficiency function, the transportation cost function and the time limit compliance rate function, comprising: determining warehouse operation efficiency weight coefficients, transportation cost weight coefficients and time limit compliance rate weight coefficients of the warehouse operation efficiency function, the transportation cost function and the time limit compliance rate function; Based on the warehouse operation efficiency function, the transportation cost function, the time limit compliance rate function, the warehouse operation efficiency weight coefficients, the transportation cost weight coefficients and the time limit compliance rate weight coefficients, the multi-objective function of the warehouse is calculated and constructed by using the following formula, wherein the multi-objective function:

[0014] wherein, denotes a multi-objective function, denotes a warehouse operation efficiency function, denotes a transportation cost function, denotes a time limit compliance rate function, denotes a warehouse operation efficiency weight coefficient, denotes a transportation cost weight coefficient, denotes a time limit compliance rate weight coefficient, denotes a minimum value of warehouse operation efficiency, denotes a maximum value of warehouse operation efficiency, representing the average value of transportation cost, representing the target value of time limit compliance rate.

[0015] Optionally, the analysis of the abnormal node under the warehouse logistics optimization parameter comprises: collecting order transportation data of the target commodity under the warehouse logistics optimization parameter; based on the order transportation data, analyzing the current state of the target commodity, wherein the current state comprises the physical state of the commodity, the location state of the commodity and the transportation state of the commodity; determining the abnormal value of the target commodity according to the physical state of the commodity, the location state of the commodity and the transportation state of the commodity; based on the abnormal value, determining the abnormal node of the target commodity.

[0016] Optionally, the analysis of the real-time risk value of the abnormal node comprises: determining the abnormal factor of the abnormal node; defining the risk index of the abnormal factor; establishing a three-dimensional risk analysis matrix of the risk index; determining the real-time risk value of the abnormal node according to the three-dimensional risk analysis matrix.

[0017] In order to solve the above problems, the application also provides a cross-border e-commerce intelligent warehouse logistics optimization scheduling system, the system comprises: a data pool construction module for collecting warehouse logistics data and historical order data of a warehouse, and constructing a multi-dimensional dynamic data pool of the warehouse based on the warehouse logistics data; a priority label analysis module for analyzing customer value, commodity shortage degree and time limit urgency of order commodities corresponding to the historical order data according to the historical order data, and determining dynamic priority labels of the order commodities according to the customer value, commodity shortage degree and time limit urgency; a twin model construction module for constructing a three-dimensional digital twin model of the warehouse, and intelligently mapping the three-dimensional digital twin model based on the multi-dimensional dynamic data pool to obtain a state mapping twin model; an optimization parameter determination module for establishing multiple optimization targets of the warehouse, establishing a multi-objective mixed integer programming model of the warehouse under a plurality of predetermined constraint conditions, and outputting warehouse logistics optimization parameters of the warehouse by using the multi-objective mixed integer programming model based on the dynamic priority labels and the state mapping twin model, wherein the warehouse logistics optimization parameters comprise transportation optimization mode and transportation optimization path; The logistics optimization scheduling module is configured to analyze an abnormal node under the warehouse logistics optimization parameter, analyze a real-time risk value of the abnormal node, adjust the warehouse logistics optimization parameter based on the real-time risk value, obtain a target warehouse logistics optimization parameter, and perform warehouse logistics optimization scheduling of the warehouse based on the target warehouse logistics optimization parameter.

[0018] By implementing the above warehouse logistics optimization scheduling method and system, the overall efficiency and operation level of warehouse logistics can be significantly improved. Firstly, the multi-dimensional dynamic data pool and the three-dimensional digital twin model are constructed, which realizes real-time and accurate mapping of the warehouse state, and provides a reliable data basis for optimization scheduling. Secondly, by analyzing historical order data, the dynamic priority label of the order goods is determined, which ensures that high-value, scarce or urgent orders can be processed preferentially, and waste of resources is avoided. Thirdly, the warehouse logistics optimization parameters output by the multi-objective mixed integer programming model, including transportation optimization mode and transportation optimization path, can significantly improve warehouse operation efficiency, reduce in-warehouse time, speed up order processing speed, and reduce operation cost. In addition, the system adjusts the warehouse logistics optimization parameter by analyzing the abnormal node and the real-time risk value, effectively reduces the additional cost caused by abnormal conditions, such as delay, loss, rework, etc., and enhances the risk control capability. Finally, the optimization scheduling performed based on the target warehouse logistics optimization parameter can ensure that the warehouse realizes multiple optimization goals under the premise of meeting multiple constraint conditions, improves customer satisfaction, and improves the overall benefit and competitiveness of warehouse logistics. The application of this method and system makes the warehouse logistics management more intelligent and refined. Therefore, the present application can improve the operation efficiency of cross-border e-commerce warehouse logistics. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 A flowchart of a cross-border e-commerce intelligent warehouse logistics optimization scheduling method provided by an embodiment of the present application is shown. Figure 2 A three-level risk threshold construction diagram for implementing the cross-border e-commerce intelligent warehouse logistics optimization scheduling method provided by an embodiment of the present application is shown. Figure 3 A module diagram for implementing the cross-border e-commerce intelligent warehouse logistics optimization scheduling method provided by an embodiment of the present application is shown.

[0020] The object implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0022] The embodiment of the present application provides a cross-border e-commerce intelligent warehousing logistics optimization scheduling method. The execution subject of the cross-border e-commerce intelligent warehousing logistics optimization scheduling method includes but is not limited to at least one of electronic devices such as a server, a terminal and the like which can be configured to execute the method provided by the embodiment of the present application. In other words, the cross-border e-commerce intelligent warehousing logistics optimization scheduling method can be executed by software or hardware installed in a terminal device or a server device. The server includes but is not limited to a single server, a server cluster, a cloud server or a cloud server cluster and the like.

[0023] Embodiment 1 Referring to Figure 1 Fig. 1 is a flowchart of a cross-border e-commerce intelligent warehousing logistics optimization scheduling method provided by an embodiment of the present application. In the embodiment, the cross-border e-commerce intelligent warehousing logistics optimization scheduling method includes the following steps. S1, collecting warehousing logistics data and historical order data of a warehouse, and constructing a multi-dimensional dynamic data pool of the warehouse based on the warehousing logistics data.

[0024] It should be explained that the warehouse refers to a physical facility or warehouse used by cross-border e-commerce to store goods, the warehousing logistics data refers to various data related to the operation and management of the cross-border e-commerce warehouse, such as inventory data, order data, transportation data, equipment data and the like, and the historical order data refers to the order records processed by the cross-border e-commerce platform or the warehousing system in the past period of time, such as order basic information, commodity information, order status information and the like.

[0025] The present application constructs a multi-dimensional dynamic data pool of the warehouse based on the warehousing logistics data, which can construct a multi-dimensional dynamic data pool with clear structure, rich content, real-time update and reliable quality, and provide strong data support for cross-border e-commerce intelligent warehousing logistics optimization scheduling.

[0026] In detail, the step of constructing a multi-dimensional dynamic data pool of the warehouse based on the warehousing logistics data includes the following steps. preprocessing the warehousing logistics data to obtain preprocessed warehousing logistics data; establishing a snowflake storage structure of the preprocessed warehousing logistics data; mining derivative indexes of the preprocessed warehousing logistics data; associating the derivative indexes with the preprocessed warehousing logistics data to obtain an association data set; storing the association data set into the snowflake storage structure to obtain the multi-dimensional dynamic data pool of the warehouse.

[0027] The preprocessing warehouse logistics data refers to data extracted from warehouse logistics data (such as WMS, TMS, IoT sensor, etc.) and subjected to a series of cleaning, conversion, standardization and other operations to meet the data quality and subsequent analysis requirements, and the snowflake storage structure refers to a data warehouse storage structure, which is a variant of a star schema. On the basis of the star schema, the dimension table is further standardized, and the dimension table is split into smaller and more atomic tables and related by primary and foreign keys. The derived index refers to the data index associated with the preprocessing warehouse logistics data, such as warehouse turnover rate and order fulfillment time. The associated data set refers to the association of the derived index and the original data after preprocessing through common keys (such as order ID, product ID, timestamp, etc.) to form a new data set. The multi-dimensional dynamic data pool refers to the storage of the associated data set in the snowflake storage structure to form a data storage environment that integrates multiple dimensions, is updated in real time and is scalable.

[0028] Optionally, the establishment of the snowflake storage structure of the preprocessing warehouse logistics data can be based on the star schema, and the dimension table is further standardized and split into smaller tables to reduce data redundancy through data modeling technology.

[0029] Optionally, the derived index of the preprocessing warehouse logistics data can be a new index generated by calculation, statistics, aggregation and the like.

[0030] S2, according to the historical order data, analyzing the customer value, commodity tightness and time urgency of the order goods corresponding to the historical order data, and determining the dynamic priority label of the order goods according to the customer value, commodity tightness and time urgency.

[0031] According to the historical order data, the application analyzes the customer value, commodity tightness and time urgency of the order goods corresponding to the historical order data, which can improve the reliability of the later warehouse optimization.

[0032] In detail, the analysis of the customer value, commodity tightness and time urgency of the order goods corresponding to the historical order data according to the historical order data comprises: Removing invalid orders in the historical order data to obtain target orders; Extracting customer dimension features, commodity dimension features and cross-border dimension features of the target orders; Based on the customer dimension features, calculating the order customer score of the order goods corresponding to the order customer to determine the customer value of the order customer; According to the commodity dimension features, analyzing the demand degree of the order goods to determine the commodity tightness of the order goods; determine a time-sensitive level of the order item through the cross-border dimension feature; calculate a time urgency of the order item according to the time-sensitive level.

[0033] The target order refers to an effective order after data cleaning, excluding returned, canceled, repeated or incomplete data orders, the customer dimension feature refers to a feature related to customer behavior and value extracted from the order, including customer level (VIP / ordinary), registration time, region (domestic / cross-border), the item dimension feature refers to a feature describing the properties and inventory status of the item, such as current inventory, safety inventory threshold, replenishment cycle, inventory turnover rate, the cross-border dimension feature refers to a feature related to cross-border logistics and customs clearance, such as transportation mode (air / sea / railway), carrier time commitment, the order customer score refers to a numerical score calculated by the customer dimension feature, the customer value refers to the normalized order customer score (0~1), the item shortage degree refers to an index quantifying the immediate supply pressure of the item (0~1), the higher the value, the more scarce, the time-sensitive level refers to a discrete level (such as 1~5 levels) divided according to the cross-border logistics channel, the higher the level, the more stringent the time, and the time urgency refers to a continuous index (0~1) based on the time-sensitive level and the remaining time, reflecting the time pressure of order delivery.

[0034] Optionally, the order customer score of the order item corresponding to the order customer based on the customer dimension feature can be calculated by determining the score index of the order customer through the customer dimension feature, analyzing the dynamic weight of the score index, and performing weighted calculation on the score index and the dynamic weight. Wherein, the score index includes purchase time, purchase frequency and the like.

[0035] According to the customer value, item shortage degree and time urgency, the application determines the dynamic priority label of the order item, which provides a basis for dynamic optimization of the late warehouse.

[0036] In detail, the dynamic priority label of the order item is determined according to the customer value, item shortage degree and time urgency, comprising: normalizing the customer value, item shortage degree and time urgency to obtain normalized customer value, normalized item shortage degree and normalized time urgency; determining the customer value weight, item shortage degree weight and time urgency weight of the normalized customer value, normalized item shortage degree and normalized time urgency; Based on the normalized customer value, the normalized commodity shortage degree, the normalized time urgency degree, the customer value weight, the commodity shortage degree weight, and the time urgency degree weight, the priority coefficient of the order commodity is calculated by using the following formula:

[0037] wherein, P represents the priority coefficient of the order commodity, V represents the normalized customer value, W represents the customer value weight of the normalized customer value, S represents the normalized commodity shortage degree, W represents the commodity shortage degree weight of the normalized commodity shortage degree, T represents the time urgency degree, W represents the time urgency degree weight of the time urgency degree, exp represents the exponential function; According to the priority coefficient, the dynamic priority label of the order commodity is determined.

[0038] wherein, the normalized customer value refers to scaling the customer value data to a unified dimension according to certain rules, the normalized commodity shortage degree refers to scaling the commodity shortage degree data to a unified dimension according to certain rules, the normalized time urgency degree refers to scaling the time urgency degree data to a unified dimension according to certain rules, the customer value weight refers to the relative importance of the normalized customer value in calculating the priority coefficient of the order commodity, the commodity shortage degree weight refers to the relative importance of the normalized commodity shortage degree in calculating the priority coefficient of the order commodity, the time urgency degree weight refers to the relative importance of the normalized time urgency degree in calculating the priority coefficient of the order commodity, and the dynamic priority label refers to dynamically and real-timely assigning a priority level to the order commodity according to the value of the priority coefficient (P) of the order commodity.

[0039] In detail, the priority coefficient calculation formula adopts a nonlinear weighting algorithm, wherein, the high-value customer differentiation degree can be amplified, the commodity shortage degree extreme value fluctuation can be alleviated, it is a time-constrained stepwise amplification effect.

[0040] Optionally, the normalization of the customer value, the commodity shortage degree, and the time urgency degree to obtain the normalized customer value, the normalized commodity shortage degree, and the normalized time urgency degree can be normalized by the minimum-maximum normalization method.

[0041] S3, a three-dimensional digital twin model of the warehouse is constructed, and state intelligent mapping is performed on the three-dimensional digital twin model based on the multi-dimensional dynamic data pool to obtain a state mapping twin model.

[0042] The three-dimensional digital twin model of the warehouse constructed by the application can construct a powerful, real-time updated three-dimensional digital twin model of the warehouse, thereby providing strong support for warehouse operation and management.

[0043] In detail, the construction of the three-dimensional digital twin model of the warehouse comprises: identifying the geometric structure, warehouse equipment and warehouse environment of the warehouse; performing entity modeling on the geometric structure, warehouse equipment and warehouse environment to obtain a warehouse physical model; analyzing the logistics behavior, equipment behavior and personnel behavior of the warehouse; performing behavior modeling on the logistics behavior, equipment behavior and personnel behavior to obtain a warehouse behavior model; integrating the warehouse physical model and the warehouse behavior model to obtain the three-dimensional digital twin model of the warehouse.

[0044] The geometric structure refers to the physical space form and layout of the warehouse, including the shape of the warehouse, internal space division, shelf arrangement mode, passage design, etc. The warehouse equipment refers to physical equipment used for handling, storage, sorting and other operations in the warehouse, such as conveyors, forklifts and other equipment. The warehouse environment refers to the environmental factors inside the warehouse, such as light, temperature and other environments. The warehouse physical model refers to the entity modeling result of the geometric structure, equipment and environment of the warehouse, which is a static digital model describing physical entities. The logistics behavior refers to the business processes related to goods storage and handling in the warehouse, such as warehousing, warehousing, storage and other behaviors. The equipment behavior refers to the running state of various devices in the warehouse when performing tasks, such as the driving path and speed of the forklift. The personnel behavior refers to the actions of the workers in the warehouse when performing tasks, such as the picking path and efficiency of the picker. The warehouse behavior model refers to a dynamic digital model describing business processes and operations. The three-dimensional digital twin model refers to the result of integrating the warehouse physical model and the warehouse behavior model, which is a virtual model that comprehensively and real-time reflects the state and behavior of the physical warehouse.

[0045] Further, the behavior modeling of the logistics behavior, equipment behavior and personnel behavior to obtain the warehouse behavior model can use simulation software (such as AnyLogic, FlexSim, Arena, etc.) to simulate logistics processes, equipment operation and personnel operation.

[0046] The application maps the state of the three-dimensional digital twin model based on the multi-dimensional dynamic data pool to obtain a state mapping twin model, realizes real-time and intelligent mapping of the state of the warehouse library, and better understands the running state of the warehouse library. The state mapping twin model refers to the current running state of the warehouse library model obtained by mapping the current state of the corresponding goods, equipment and environment of the warehouse library to the three-dimensional digital twin model through the multi-dimensional dynamic data pool. In detail, the three-dimensional digital twin model uses a physical-digital mapping algorithm (such as Kalman filtering) to synchronize real-time data to a virtual model, ensuring that the digital twin and the physical warehouse state are consistent.

[0047] S4, a multi-optimization target of the warehouse library is established to establish a multi-objective mixed integer programming model of the warehouse library under a plurality of preset constraint conditions, and a warehouse logistics optimization parameter of the warehouse library is output by using the multi-objective mixed integer programming model based on the dynamic priority label and the state mapping twin model, wherein the warehouse logistics optimization parameter includes a transportation optimization mode and a transportation optimization path.

[0048] It should be explained that the multi-optimization target refers to multiple, possibly conflicting targets that need to be considered and optimized at the same time, including cost targets, warehouse operation efficiency targets, and time efficiency compliance rate targets.

[0049] The multi-objective mixed integer programming model established under the plurality of preset constraint conditions can improve the logistics efficiency and reliability of the goods.

[0050] In detail, the establishment of the multi-objective mixed integer programming model of the warehouse library under the plurality of preset constraint conditions includes: Based on the plurality of constraint conditions, the warehouse priority weight, the channel cost, and the channel standard transportation time efficiency of the warehouse library are determined; According to the warehouse priority weight, a warehouse operation efficiency function of the warehouse library is established; Through the channel cost, a transportation cost function of the warehouse library is established; Based on the channel standard transportation time efficiency, the following formula is used to construct a time efficiency compliance rate function of the warehouse library, wherein the time efficiency compliance rate function:

[0051] Wherein, The time efficiency compliance rate function is represented by, The first commodity order of the warehouse library is represented by, The first warehouse of the warehouse library is represented by, The first commodity order of the warehouse library is represented by, a transportation channel, denotes a binary variable, a commodity order from a warehouse through a transportation channel transportation time is 1, denotes the task completion time of the warehouse, denotes the channel standard transportation time of the transportation channel, denotes the promised delivery time of the commodity order of the warehouse; According to the warehouse operation efficiency function, the transportation cost function and the time limit compliance rate function, a multi-objective function of the warehouse is constructed; Through the multi-objective function, a multi-objective mixed integer programming model of the warehouse under a preset multi-constraint condition is established.

[0052] , wherein the warehouse priority weight refers to a relative weight value given to each warehouse in the warehouse network according to its importance, location, capacity, processing capacity, customer service capacity and other factors, the channel cost refers to the unit cost generated by transporting goods from the warehouse to the destination through a specific transportation channel (such as express, less than truckload, full truck, air transport, etc.), the channel standard transportation time refers to the standard or average transportation time expected to be spent from the warehouse to the destination through a specific transportation channel, the warehouse operation efficiency function refers to a function for quantifying the efficiency of completing order operations inside the warehouse (from receiving to picking, packaging, and warehouse-out), and the specific quantification can adopt “the number of orders completed per unit time”, the transportation cost function refers to a total cost function for calculating the total cost of transporting all orders from the warehouse to the destination through the selected transportation channel, and the transportation cost is mainly composed of variable cost (fees directly related to transportation volume, such as transportation fee) and possible fixed cost (fees related to single shipment or batch), the time limit compliance rate function refers to a function for calculating the proportion of orders in which the actual transportation time (warehouse task completion time + channel standard time) does not exceed the promised delivery time, and the time limit compliance rate function is used to calculate the proportion of orders in which the actual transportation time does not exceed the promised delivery time, the promised delivery time refers to the final time point at which the warehouse or logistics service provider promises to deliver goods to the customer for a specific customer order, the multi-objective function refers to a set composed of the warehouse operation efficiency function, the transportation cost function and the time limit compliance rate function in a multi-objective optimization problem, and the multi-objective mixed integer programming model refers to a mathematical optimization model containing multiple objective functions established under a series of constraint conditions (such as warehouse capacity, transportation capacity, time window, etc.).

[0053] Optionally, the establishing the warehouse operation efficiency function of the warehouse according to the warehouse priority weight can be constructed by a weighted summation formula.

[0054] Further, the constructing the multi-objective function of the warehouse according to the warehouse operation efficiency function, the transportation cost function and the time efficiency compliance rate function comprises: determining warehouse operation efficiency weight coefficient, transportation cost weight coefficient and time efficiency compliance rate weight coefficient of the warehouse operation efficiency function, the transportation cost function and the time efficiency compliance rate function; calculating and constructing the multi-objective function of the warehouse based on the warehouse operation efficiency function, the transportation cost function, the time efficiency compliance rate function, the warehouse operation efficiency weight coefficient, the transportation cost weight coefficient and the time efficiency compliance rate weight coefficient, wherein the multi-objective function is:

[0055] wherein, represents the multi-objective function, represents the warehouse operation efficiency function, represents the transportation cost function, represents the time efficiency compliance rate function, represents the warehouse operation efficiency weight coefficient, represents the transportation cost weight coefficient, represents the time efficiency compliance rate weight coefficient, represents the minimum value of the warehouse operation efficiency, represents the maximum value of the warehouse operation efficiency, represents the average value of the transportation cost, represents the target value of the time efficiency compliance rate.

[0056] wherein the warehouse operation efficiency weight coefficient refers to the importance of the warehouse operation efficiency relative to other targets (transportation cost, time efficiency compliance rate) in the overall optimization target, the transportation cost weight coefficient refers to the importance of the transportation cost relative to other targets in the overall optimization target, the time efficiency compliance rate weight coefficient refers to the importance of the time efficiency compliance rate relative to other targets in the overall optimization target, and the transportation cost average value refers to an average value of the transportation cost in all possible transportation schemes or historical data considered.

[0057] The physical intelligent optimization of the warehouse can be realized by using the warehouse logistics optimization parameter output by the multi-objective mixed integer programming model based on the dynamic priority label and the state mapping twin model. The transportation optimization mode refers to the transportation mode analyzed according to the multi-objective mixed integer programming model. Specifically, the transportation optimization mode generates a transportation mode decision table (such as preferentially using a combination of "drone + ground vehicle") by inputting the real-time state (such as transportation vehicle load rate, road congestion index) output by the state mapping twin model and the dynamic priority label (such as order urgency) into the mixed integer programming model, using an improved genetic algorithm. The transportation optimization path refers to the commodity transportation path planned according to the multi-objective mixed integer programming model. Specifically, the transportation optimization path generation includes a dynamic map (containing real-time road conditions, temporary no-entry areas) updated in real time based on the twin model, using the A* algorithm combined with distance, congestion index, and transfer frequency constraints, and introducing a rolling horizon control (RHC), and generating a two-dimensional GIS map label every 30 minutes: green arrow = optimal path, red dashed line = avoidance area and path predicted time consumption.

[0058] S5, analyze the abnormal node under the warehouse logistics optimization parameter, analyze the real-time risk value of the abnormal node, adjust the warehouse logistics optimization parameter based on the real-time risk value, obtain the target warehouse logistics optimization parameter, and execute the warehouse logistics optimization scheduling of the warehouse based on the target warehouse logistics optimization parameter.

[0059] The analysis of the abnormal node under the warehouse logistics optimization parameter can capture the abnormal state that may exist in real time, thereby improving the reliability of the warehouse logistics.

[0060] Specifically, the analysis of the abnormal node under the warehouse logistics optimization parameter comprises: Collecting order transportation data of a target commodity under the warehouse logistics optimization parameter; Based on the order transportation data, analyzing the current state of the target commodity, wherein the current state includes the physical state of the commodity, the location state of the commodity, and the transportation state of the commodity; According to the physical state of the commodity, the location state of the commodity and the transportation state of the commodity, determine the abnormal value of the target commodity; Based on the abnormal value, determine the abnormal node of the target commodity.

[0061] The target commodity refers to a commodity transported by the warehouse logistics system, the order transportation data refers to various data records of an order related to the target commodity in the transportation process, the commodity physical state refers to a physical condition of the target commodity in the transportation process, for example, a perfect, damaged or the like condition, the commodity location state refers to a current location of the target commodity in the warehouse logistics system, the commodity transportation state refers to a state of the target commodity in the transportation process, the abnormal value refers to a value not conforming to a normal range, which is identified according to the commodity physical state, the commodity location state and the commodity transportation state by comparison with a preset normal value or threshold, and the abnormal node refers to a specific link causing the abnormal value in the warehouse logistics system.

[0062] Optionally, the collection of the order transportation data of the target commodity under the warehouse logistics optimization parameter can be realized by using a radio frequency identification (RFID) technology, a sensor and GPS positioning in combination.

[0063] The analysis of the real-time risk value of the abnormal node can analyze the risk caused by the abnormal node, so that risk optimization can be performed in time and the efficiency of the warehouse physical is improved.

[0064] In detail, the analysis of the real-time risk value of the abnormal node comprises: determining an abnormal factor of the abnormal node; defining a risk index of the abnormal factor; establishing a three-dimensional risk analysis matrix of the risk index; determining the real-time risk value of the abnormal node according to the three-dimensional risk analysis matrix.

[0065] The abnormal factor refers to a root variable causing the warehouse logistics node to deviate from the normal state, including equipment failure type (conveyor downtime, forklift failure frequency), operation error type (picking error rate, packaging damage rate), external interference type (transportation delay time, weather influence level), wherein the equipment failure type is calculated by recording the failure time / frequency through sensors (such as vibration, temperature sensors) or maintenance logs, the operation error type is calculated by recording the number of error orders through WMS (warehouse management system), and the external interference type is obtained by transportation delay data through the logistics platform API, the risk index refers to a measurable parameter quantifying the damage degree of the abnormal factor, in detail, the risk index is calculated by multiplying the abnormal factor and the abnormal factor influence coefficient, the three-dimensional risk analysis matrix refers to a three-dimensional model for evaluating the risk from three dimensions of severity (Severity), probability of occurrence (Probability), and controllability (Controllability), exemplary scoring standards: severity (S): 1 (mild) ~ 5 (catastrophic); probability of occurrence (P): 1 (extremely low) ~ 5 (extremely high); controllability (C): 1 (completely uncontrollable) ~ 5 (completely controllable), the real-time risk value refers to a normalized risk evaluation value output by the three-dimensional matrix, and a weighted geometric mean method is used to integrate the three-dimensional scores.

[0066] Exemplarily, the abnormal node is a scene that an AGV frequently loses connection in the cross-border picking area of a bonded warehouse, the abnormal factor is a positioning signal interference and an HS code recognition error, the risk index refers to MTBF=45min (<standard value 120min), and customs compliance=65% (<threshold value 85%), then the three-dimensional risk analysis matrix constructed has severity=0.8, probability of occurrence=0.7, and controllability=0.4, and the real-time risk value is sqrt{0.8^2 + 0.7^2 + 0.6^2} 1.2=1.25{more than 1.0 needs emergency treatment}.

[0067] Alternatively, the three-dimensional risk analysis matrix for establishing the risk index can be realized by a fault tree analysis method.

[0068] Based on the real-time risk value, the warehouse logistics optimization parameters are adjusted to obtain target warehouse logistics optimization parameters, which can further optimize the warehouse logistics to improve the overall cross-border warehouse logistics efficiency. The target warehouse logistics optimization parameters refer to a final decision variable set dynamically adjusted based on the real-time risk value, which is used to guide the actual warehouse logistics operation, including inventory management related parameters, order fulfillment related parameters, transportation and distribution related parameters and the like. In detail, based on the real-time risk value, the three-level risk threshold of the target commodity corresponding to the warehouse logistics optimization parameter is determined, Referring to Figure 2 As shown in the figure, a three-level risk threshold construction diagram for implementing the cross-border e-commerce intelligent warehouse logistics optimization scheduling method is provided in an embodiment of the present application: wherein the three-level risk threshold includes low risk: automatic recording, medium risk: early warning prompt, high risk: interrupt process, wherein the automatic recording refers to when the system detects that the real-time risk value is lower than the low risk threshold, the system will automatically record the current running data and state information, including but not limited to device running parameters, order processing progress, inventory level, environmental data, etc., the early warning prompt refers to when the real-time risk value is between the low risk threshold and the medium risk threshold, the system will trigger the early warning mechanism, and send early warning information to relevant personnel (for example, warehouse managers, dispatchers, etc.). Early warning information usually includes risk level, risk source, possible impact and suggested countermeasures, the interrupt process refers to when the real-time risk value is higher than the high risk threshold, the system will immediately interrupt the current logistics process and stop the related job activities.

[0069] By implementing the above warehouse logistics optimization scheduling method and system, the overall efficiency and operation level of warehouse logistics can be significantly improved. First, the multi-dimensional dynamic data pool and three-dimensional digital twin model are constructed, which realizes real-time and accurate mapping of the warehouse state, providing a reliable data foundation for optimization scheduling. Second, by analyzing historical order data, the dynamic priority label of order goods is determined to ensure that high-value, scarce or urgent orders can be processed first, avoiding waste of resources. Third, the warehouse logistics optimization parameters output by the multi-objective mixed integer programming model, including transportation optimization mode and transportation optimization path, can significantly improve warehouse operation efficiency, reduce warehouse time, speed up order processing, and reduce operating costs. In addition, the system adjusts the warehouse logistics optimization parameters by analyzing abnormal nodes and real-time risk values, effectively reducing additional costs such as delays, losses, rework, etc., enhancing risk control capabilities. Finally, based on the target warehouse logistics optimization parameters, the optimization scheduling can ensure that the warehouse meets multiple constraint conditions and achieves multiple optimization goals, improving customer satisfaction and overall benefits and competitiveness of warehouse logistics. The application of this method and system makes warehouse logistics management more intelligent and refined. Therefore, the present application can improve the operation efficiency of cross-border e-commerce warehouse logistics.

[0070] Embodiment 2: As Figure 3 shown, it is a function module diagram of a cross-border e-commerce intelligent warehouse logistics optimization scheduling system of the present application.

[0071] The cross-border e-commerce intelligent warehouse logistics optimization scheduling system 300 can be installed in an electronic device. According to the functions implemented, the cross-border e-commerce intelligent warehouse logistics optimization scheduling system can include a data pool construction module 301, a priority label analysis module 302, a twin model construction module 303, an optimization parameter determination module 304, and a logistics optimization scheduling module 305. The modules in the present application can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete a fixed function, which are stored in the memory of the electronic device.

[0072] In the embodiments of the present application, the functions of each module / unit are as follows: The data pool construction module 301 is configured to collect warehouse logistics data and historical order data of a warehouse, and construct a multi-dimensional dynamic data pool of the warehouse based on the warehouse logistics data. The priority label analysis module 302 is configured to analyze customer value, commodity tightness, and time urgency of order goods corresponding to the historical order data according to the historical order data, and determine a dynamic priority label of the order goods according to the customer value, commodity tightness, and time urgency. The twin model construction module 303 is configured to construct a three-dimensional digital twin model of the warehouse, and perform state intelligent mapping on the three-dimensional digital twin model based on the multi-dimensional dynamic data pool to obtain a state mapping twin model. The optimization parameter determination module 304 is configured to establish multiple optimization objectives of the warehouse, to establish a multi-objective mixed integer programming model of the warehouse under a plurality of predetermined constraint conditions, and to output warehouse logistics optimization parameters of the warehouse by using the multi-objective mixed integer programming model based on the dynamic priority label and the state mapping twin model, wherein the warehouse logistics optimization parameters include a transportation optimization mode and a transportation optimization path. The logistics optimization scheduling module 305 is configured to analyze abnormal nodes under the warehouse logistics optimization parameters, to analyze real-time risk values of the abnormal nodes, to adjust the warehouse logistics optimization parameters based on the real-time risk values to obtain target warehouse logistics optimization parameters, and to perform warehouse logistics optimization scheduling of the warehouse based on the target warehouse logistics optimization parameters.

[0073] In detail, the modules in the cross-border e-commerce intelligent warehouse logistics optimization scheduling system 300 in the embodiments of the present application use the same technical means as the cross-border e-commerce intelligent warehouse logistics optimization scheduling method described in the above Figure 1 , and can produce the same technical effects, which will not be described here again.

[0074] It is apparent for a person skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, but that the present application can be implemented in other concrete forms without departing from the spirit or essential characteristics of the present application.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not to limit the present application, and although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the present application.

Claims

1. A cross-border e-commerce intelligent warehousing logistics optimization scheduling method, characterized in that: The method comprises: Collecting warehouse logistics data and historical order data of the warehouse, and building a multi-dimensional dynamic data pool of the warehouse based on the warehouse logistics data; Analyzing, based on the historical order data, customer value, product scarcity, and timeliness of order products corresponding to the historical order data, and determining a dynamic priority tag for the order products based on the customer value, product scarcity, and timeliness; Constructing a three-dimensional digital twin model of the warehouse, and performing state intelligent mapping on the three-dimensional digital twin model based on the multi-dimensional dynamic data pool to obtain a state mapping twin model; Establishing multiple optimization objectives for the warehouse to establish a multi-objective mixed integer programming model for the warehouse under preset multi-constraint conditions; based on the dynamic priority label and the state mapping twin model, using the multi-objective mixed integer programming model to output the warehouse logistics optimization parameters of the warehouse, wherein the warehouse logistics optimization parameters include the transportation optimization mode and the transportation optimization path; Analyze the abnormal nodes under the warehousing and logistics optimization parameters to analyze the real-time risk values ​​of the abnormal nodes, adjust the warehousing and logistics optimization parameters based on the real-time risk values, obtain target warehousing and logistics optimization parameters, and execute warehousing and logistics optimization scheduling of the warehouse based on the target warehousing and logistics optimization parameters.

2. The cross-border e-commerce intelligent warehousing logistics optimization and scheduling method according to claim 1 is characterized in that: The multi-dimensional dynamic data pool of the warehouse is constructed based on the warehouse logistics data, including: Preprocessing the warehousing logistics data to obtain preprocessed warehousing logistics data; Establishing a snowflake storage structure for the pre-processed warehousing logistics data; mining derivative indicators of the pre-processed warehouse logistics data; Associating the derived index with the pre-processed warehousing logistics data to obtain an associated data set; The associated data set is stored in the Snowflake storage structure to obtain a multi-dimensional dynamic data pool of the warehouse.

3. The cross-border e-commerce intelligent warehousing logistics optimization scheduling method according to claim 2 is characterized in that: Analyzing the customer value, commodity scarcity, and timeliness of the order commodities corresponding to the historical order data based on the historical order data includes: Removing invalid orders from the historical order data to obtain target orders; Extracting customer dimension features, product dimension features, and cross-border dimension features of the target order; Based on the customer dimension features, calculating the order customer score of the order customer corresponding to the order product to determine the customer value of the order customer; Analyze the demand for the ordered goods based on the dimensional characteristics of the goods to determine the scarcity of the ordered goods; Determine the timeliness level of the ordered goods based on the cross-border dimension features; The urgency of the order's timeliness is calculated based on the timeliness level.

4. The cross-border e-commerce intelligent warehousing logistics optimization scheduling method according to claim 3 is characterized in that: Determining the dynamic priority tag of the ordered product based on the customer value, product scarcity, and time urgency includes: Normalizing the customer value, product scarcity, and timeliness urgency to obtain normalized customer value, normalized product scarcity, and normalized timeliness urgency; Determining the customer value weights, product scarcity weights, and timeliness urgency weights of the normalized customer value, normalized product scarcity, and normalized timeliness urgency; Based on the normalized customer value, normalized product scarcity, normalized time urgency, customer value weight, product scarcity weight, and time urgency weight, the priority coefficient of the order product is calculated using the following formula: in, Indicates the priority coefficient of the order item. represents the normalized customer value, The customer value weight representing the normalized customer value, represents the normalized commodity scarcity, The commodity scarcity weight representing the normalized commodity scarcity, Indicates time urgency. The timeliness urgency weight indicating the timeliness urgency, represents the exponential function; A dynamic priority label of the order item is determined according to the priority coefficient.

5. The cross-border e-commerce intelligent warehousing logistics optimization scheduling method according to claim 4 is characterized in that: The construction of the three-dimensional digital twin model of the warehouse includes: Identifying the warehouse's geometric structure, warehouse equipment, and warehouse environment; Performing entity modeling on the geometric structure, storage equipment, and storage environment to obtain a physical model of the storage; Analyze the logistics behavior, equipment behavior, and personnel behavior of the warehouse; Conducting behavioral modeling on the logistics behavior, equipment behavior, and personnel behavior to obtain a warehouse behavior model; The physical model of the warehouse and the behavioral model of the warehouse are integrated to obtain a three-dimensional digital twin model of the warehouse.

6. The cross-border e-commerce intelligent warehousing logistics optimization scheduling method according to claim 5 is characterized in that: The multi-objective mixed integer programming model for the warehouse is established under preset multi-constraint conditions, including: Based on the multiple constraints, determining the warehouse priority weight, channel cost, and channel standard transportation time efficiency of the warehouse; Establishing a warehousing operation efficiency function of the warehouse according to the warehouse priority weight; Establishing the transportation cost function of the warehouse through the channel cost; Based on the channel standard transportation time efficiency, the following formula is used to construct the time efficiency compliance rate function of the warehouse, wherein the time efficiency compliance rate function is: in, represents the timeliness compliance rate function, Indicates the warehouse Product orders, Indicates the warehouse warehouses, Indicates the warehouse Product order transportation channels, Represents a binary variable, product order From the warehouse Through transportation channels 1 for transport, Indicates the The task completion time of each warehouse, Indicates the The standard transportation time of each transportation channel, Indicates the warehouse The promised delivery time of each product order; Constructing a multi-objective function of the warehouse based on the warehouse operation efficiency function, the transportation cost function and the timeliness compliance rate function; Through the multi-objective function, a multi-objective mixed integer programming model of the warehouse under preset multi-constraint conditions is established.

7. The cross-border e-commerce intelligent warehousing logistics optimization scheduling method according to claim 6, characterized in that: The multi-objective function of the warehouse is constructed based on the warehousing operation efficiency function, the transportation cost function and the timeliness compliance rate function, including: Determining a warehousing operation efficiency weight coefficient, a transportation cost weight coefficient, and a timeliness compliance rate weight coefficient for the warehousing operation efficiency function, the transportation cost function, and the timeliness compliance rate function; Based on the warehousing operation efficiency function, the transportation cost function, the timeliness compliance rate function, the warehousing operation efficiency weight coefficient, the transportation cost weight coefficient, and the timeliness compliance rate weight coefficient, the following formula is used to calculate and construct the multi-objective function of the warehouse, wherein the multi-objective function is: in, represents a multi-objective function, represents the warehouse operation efficiency function, represents the transportation cost function, represents the timeliness compliance rate function, represents the weight coefficient of warehousing operation efficiency, represents the transportation cost weight coefficient, represents the weight coefficient of the timeliness compliance rate, Indicates the minimum value of warehousing operation efficiency, Indicates the maximum efficiency of warehousing operations, represents the average transportation cost, Indicates the target value of the timeliness achievement rate.

8. The cross-border e-commerce intelligent warehousing logistics optimization scheduling method according to claim 7, characterized in that: The analyzing of abnormal nodes under the warehouse logistics optimization parameters includes: Collecting order transportation data corresponding to the target product under the warehouse logistics optimization parameters; Analyzing the current status of the target product based on the order transportation data, wherein the current status includes the physical status of the product, the location status of the product, and the transportation status of the product; Determining an abnormal value of the target commodity based on the physical state, location state, and transportation state of the commodity; Based on the outlier value, an outlier node of the target product is determined.

9. The cross-border e-commerce intelligent warehousing logistics optimization and scheduling method according to claim 8, characterized in that: The analyzing the real-time risk value of the abnormal node includes: Determining an abnormal factor of the abnormal node; Defining risk indicators for the abnormal factors; Establishing a three-dimensional risk analysis matrix of the risk indicators; According to the three-dimensional risk analysis matrix, a real-time risk value of the abnormal node is determined.

10. A cross-border e-commerce intelligent warehousing logistics optimization and scheduling system, characterized by: The system comprises: A data pool construction module is used to collect the warehouse logistics data and historical order data of the warehouse, and build a multi-dimensional dynamic data pool of the warehouse based on the warehouse logistics data; a priority tag analysis module for analyzing, based on the historical order data, the customer value, product scarcity, and timeliness of the order products corresponding to the historical order data, and determining a dynamic priority tag for the order products based on the customer value, product scarcity, and timeliness; A twin model construction module is used to construct a three-dimensional digital twin model of the warehouse, and perform state intelligent mapping on the three-dimensional digital twin model based on the multi-dimensional dynamic data pool to obtain a state mapping twin model; An optimization parameter determination module is used to establish multiple optimization objectives for the warehouse, so as to establish a multi-objective mixed integer programming model for the warehouse under preset multi-constraint conditions, and based on the dynamic priority label and the state mapping twin model, use the multi-objective mixed integer programming model to output the warehouse logistics optimization parameters of the warehouse, wherein the warehouse logistics optimization parameters include the transportation optimization mode and the transportation optimization path; The logistics optimization and scheduling module is used to analyze the abnormal nodes under the warehouse logistics optimization parameters to analyze the real-time risk value of the abnormal nodes, adjust the warehouse logistics optimization parameters based on the real-time risk value, obtain the target warehouse logistics optimization parameters, and execute the warehouse logistics optimization scheduling of the warehouse based on the target warehouse logistics optimization parameters.

Images