Intelligent warehouse management method

Abstract

The invention discloses an intelligent warehouse management method, which is applied to a multi-layer shuttle vehicle intensive stereoscopic warehouse, and comprises the steps: predicting warehouse-out orders in a future preset time period according to historical order data, and obtaining order prediction information; according to the order prediction information and a warehouse model of the multi-layer shuttle vehicle intensive stereoscopic warehouse, predicting a cargo storage position of a cargo in a warehousing order and a corresponding carrying equipment path; constructing a plurality of cargo handling distribution schemes based on the handling equipment path, the load balance degree of the handling equipment and the emergency degree of the task; according to the cargo storage position, the corresponding carrying equipment path and the cargo carrying distribution scheme, utilizing an optimization algorithm to calculate a globally optimal cargo storage position, a cargo carrying distribution scheme and a carrying equipment path; and carrying out warehousing scheduling on the goods in the warehousing order according to the globally optimal goods storage position, the corresponding goods carrying distribution scheme and the carrying equipment path.

Description

Technical Field

[0001] The present invention relates to the technical field of warehousing and logistics, and more specifically to an intelligent warehousing management method. Background Art

[0002] In recent years, with the vigorous development of global e-commerce and the growing market demand, traditional warehouse management methods have been unable to meet the requirements of the modern logistics industry for high efficiency, low cost and high reliability. Efficient, accurate and economical warehouse management has become one of the key factors in improving corporate competitiveness. At present, most of the goods in multi-layer shuttle intensive stereoscopic warehouses are stored in one area according to the same type, and the shelves are replenished with the same type of goods after leaving the warehouse. However, due to the randomness and uncertainty of orders, the goods stored in the innermost part may frequently enter and exit the warehouse, which consumes resources and seriously affects the efficiency of warehouse scheduling and operating costs. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide an intelligent warehouse management method to achieve the warehouse management and scheduling of goods in a multi-layer shuttle vehicle-intensive stereoscopic warehouse, thereby improving the scheduling efficiency.

[0004] In order to solve the above technical problems, the present invention provides an intelligent storage management method, which is applied to a multi-layer shuttle vehicle intensive stereoscopic warehouse, and comprises: According to the historical order data, the outbound orders within the preset time period in the future are predicted to obtain the order prediction information; Predicting the cargo storage location and corresponding handling equipment path of the cargo in the incoming order based on the order prediction information and the warehouse model of the multi-layer shuttle intensive stereoscopic warehouse; Construct multiple cargo handling allocation plans based on the handling equipment paths, load balance of the handling equipment, and the urgency of the task; According to the cargo storage location, the corresponding handling equipment path and the cargo handling allocation plan, the optimization algorithm is used to calculate the global optimal cargo storage location, cargo handling allocation plan and handling equipment path; The goods in the warehousing order are scheduled for warehousing according to the globally optimal goods storage location and the corresponding goods handling allocation plan and handling equipment path.

[0005] Its further technical solution is: predicting the outbound orders within a preset time period in the future based on the historical order data specifically includes: Use the ARIMA model to perform autocorrelation analysis and partial autocorrelation analysis on historical order data to obtain the time trend and seasonal fluctuation of historical order data, and obtain an order delivery forecasting model that can be used to predict future demand for goods; Use the outbound order forecasting model to predict outbound orders within a preset time period in the future based on seasons and / or market activities.

[0006] Its further technical solution is: predicting the outbound orders within a preset time period in the future based on the historical order data specifically includes: The decision tree algorithm is used to train the order time, goods type, quantity and goods information in the historical order data to obtain an order delivery prediction model that can be used to classify goods and predict future demand for goods; The order outbound forecasting model is used to predict outbound orders within a preset time period in the future.

[0007] Its further technical solution is: the optimization algorithm is a genetic algorithm, a simulated annealing algorithm, a particle swarm optimization algorithm and a multi-objective ant colony algorithm.

[0008] A further technical solution is: before predicting the storage location and corresponding handling equipment path of the goods in the incoming order based on the order prediction information and the warehouse model of the multi-layer shuttle intensive stereoscopic warehouse, it also includes: Constructing a warehouse model according to the shelf arrangement, storage location distribution and handling equipment distribution of the multi-layer shuttle intensive stereoscopic warehouse; A multidimensional cell array is generated according to the storage location, the pressure bearing capacity of the storage location, the distance from the storage location to the storage entrance, the storage location space, the distance from the storage location to the storage exit, and the storage location stability; wherein the multidimensional cell array is used to characterize the storage location characteristics of the storage location in the warehouse model.

[0009] A further technical solution is: predicting the storage location of the goods in the incoming order and the corresponding handling equipment path according to the order prediction information and the warehouse model of the multi-layer shuttle intensive warehouse, specifically including: Calculate and predict the storage location of the goods in the incoming order based on the quantity, type, time of inbound and outbound goods, goods information and the location characteristics of the locations in the warehouse model in the order forecast information; The corresponding transport equipment path is predicted based on the predicted distance between the cargo storage location and the transport equipment, as well as the congestion situation and the energy consumption of the transport equipment.

[0010] A further technical solution is: the calculation and prediction of the storage location of the goods in the incoming order is performed based on the quantity, type, time of inbound and outbound goods, goods information and the location characteristics of the storage location in the warehouse model in the order prediction information, specifically including: Check whether the goods in the incoming order are within the order forecast information; If so, the in-and-out quantity, in-and-out time and goods information of the goods in the order forecast information are compared and matched with the corresponding values ​​in the location characteristics of the location in the warehouse model to calculate the predicted storage location of the goods; If not, extend the duration of the future preset time period, and re-predict the outbound orders within the future preset time period based on the historical order data until the goods in the inbound order appear in the predicted outbound order, and then compare and match the inbound and outbound quantity, inbound and outbound time and goods information of the goods in the predicted outbound order with the corresponding values ​​in the location characteristics of the location in the warehouse model to calculate the predicted storage location of the goods.

[0011] Its further technical solution is: before predicting the outbound orders within a preset time period in the future based on historical order data to obtain order forecast information, it also includes: querying whether the incoming order is bound to an outbound order, and if not, executing the step of predicting the outbound orders within a preset time period in the future based on historical order data to obtain order forecast information.

[0012] Its further technical solution is: before predicting the outbound orders within a preset time period in the future based on the historical order data to obtain the order prediction information, it also includes: obtaining the historical order data, and performing data cleaning on the historical order data, and performing data standardization on the cleaned historical order data.

[0013] Its further technical solution is: scheduling the goods in the warehousing order according to the global optimal goods storage location and the corresponding goods handling allocation plan and handling equipment path, and then also includes: real-time monitoring of the operation status of the handling equipment, and dynamically adjusting the warehousing scheduling of the goods according to the operation status of the handling equipment.

[0014] Compared with the prior art, the present invention predicts outbound orders within a preset time period in the future according to historical order data to obtain order prediction information, and predicts the cargo storage location and corresponding handling equipment path of the cargo in the inbound order according to the order prediction information and the warehouse model of the multi-layer shuttle intensive stereoscopic warehouse, constructs multiple cargo handling distribution plans based on the handling equipment path and the load balancing degree of the handling equipment and the urgency of the task, and calculates the global optimal cargo storage location, cargo handling distribution plan and handling equipment path according to the cargo storage location, the corresponding handling equipment path and the cargo handling distribution plan by using an optimization algorithm, and finally performs inbound scheduling of the cargo in the inbound order according to the global optimal cargo storage location and the corresponding cargo handling distribution plan and handling equipment path. It can be seen that the present invention deeply mines the rules in the historical order data, predicts the outbound orders within a preset time period in the future, and combines the outbound order prediction with the warehouse model, and uses the optimization algorithm to comprehensively plan the cargo storage location, cargo handling distribution plan and handling equipment path of the cargo in the inbound order with the goal of optimal overall efficiency, thereby improving the efficiency of warehouse management scheduling, thereby ensuring that the cargo in and out of the warehouse is more efficient and economical. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of a flow chart of an intelligent warehouse management method provided by an embodiment of the present invention; Figure 2 A schematic diagram of sub-processes of the intelligent warehouse management device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0018] It should also be understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0019] See also Figure 1 , Figure 11 is a flow chart of the intelligent warehouse management method provided by an embodiment of the present invention. The intelligent warehouse management method of the present invention can be applied to a multi-layer shuttle vehicle-intensive stereoscopic warehouse to manage and dispatch goods in incoming orders. As shown in the figure, the method includes the following steps S110-150.

[0020] S110. Forecast outbound orders within a preset time period in the future based on historical order data to obtain order forecast information.

[0021] In this step, the historical order data includes data in different time periods and different business scenarios, including order timestamps, specific types and quantities of goods, and other relevant attributes; the future preset time period can be the future day, the future week, or the future month, and can be set according to user needs; and the order forecast information may include the expected distribution of the quantity, type, and time of goods entering and leaving the warehouse within the future preset time period, and may further include goods information, including the weight, size, correlation between different types of goods, and classification of goods; the correlation between goods refers to whether they are sold together, such as toothbrushes and toothpaste; classification of goods refers to classification according to the shape, stackability, or packaging form of the goods, for example, the goods can be classified according to the stackability of the goods, and can be divided into single-layer storage and multi-layer storage according to non-stackable and multi-layer stackable goods.

[0022] Specifically, the prediction of outbound orders within a preset time period in the future based on historical order data includes: using the ARIMA model to perform autocorrelation analysis and partial autocorrelation analysis on the historical order data to obtain the time trend and seasonal fluctuations of the historical order data, and obtain an order outbound prediction model that can be used to predict future demand for goods; using the order outbound prediction model to predict outbound orders within a preset time period in the future based on seasons and / or market activities, so as to accurately capture the time trend and seasonal fluctuations of the outbound order data, and deeply explore the trend and seasonality of the outbound orders in the time dimension; further, in this embodiment, it can also include: using a decision tree algorithm According to the order time, goods type, quantity and goods information (such as weight, size, correlation between different types of goods, etc.) in the historical order data, training is performed to obtain an order outbound forecasting model that can be used for goods classification and prediction of future goods demand; the order outbound forecasting model is used to predict the outbound orders within the future preset time period; the order outbound forecasting model constructed based on the ARIMA model is used to predict the outbound orders within the future preset time period, and the order outbound forecasting model constructed based on the decision tree algorithm is used to predict the outbound orders within the future preset time period. The forecast information is obtained by predicting the outbound orders within the future preset time period, thereby obtaining order forecasting information. In this embodiment, the ARIMA model and the decision tree algorithm are combined to predict the outbound orders within the future preset time period according to the historical order data, so as to obtain more accurate order forecasting information. It is understandable that the deep learning algorithm can be further combined to learn the long-term dependency relationship of the time series data of the outbound orders, and output more accurate order forecasting information. It is understandable that in some other embodiments, the ARIMA model or the decision tree algorithm can also be used alone to predict the outbound orders within the future preset time period, thereby obtaining order forecasting information.

[0023] In this step, the order delivery prediction model is deduced based on the historical order data to predict the delivery orders. During training, the model can be corrected by actual delivery orders. In actual use, the model can also be corrected according to different industry types. For example, the fresh food industry needs to import major festivals, seasonal, public opinion dishes, and the time change curve of the value of goods for deduction. The fast-moving consumer goods industry is related to information such as big promotions and new products. It can be corrected according to the characteristics of different industries to improve the prediction accuracy of the model.

[0024] In the present invention, before predicting the outbound orders within the preset time period in the future according to the historical order data to obtain the order prediction information, it also includes: obtaining the historical order data, performing data cleaning on the historical order data, and performing data standardization on the cleaned historical order data. Historical order data covering different time periods and multiple business scenarios can be widely collected from various business systems of the enterprise (such as ERP, WMS, etc.), and then the collected historical order data is cleaned, for example, obviously erroneous data (such as negative quantity, irregular time format, etc.) is removed, and the missing data is reasonably filled (statistical methods such as mean and median can be used), and then data standardization is performed to make data of different dimensions comparable.

[0025] S120: predicting the goods storage location and corresponding handling equipment path of the goods in the incoming order based on the order prediction information and the warehouse model of the multi-layer shuttle intensive stereoscopic warehouse.

[0026] In the present invention, the cargo storage location is the location coordinates of the storage location in the shelf, the handling equipment is a handling robot, etc., the handling equipment path is the path for the handling equipment to move to the cargo storage location, and both the cargo storage location and the handling equipment path can be multiple.

[0027] In some embodiments, before step S120, it also includes: building a warehouse model according to the shelf arrangement, storage location distribution and handling equipment distribution of the multi-layer shuttle intensive stereoscopic warehouse; generating a multidimensional cell array according to the storage location location, storage location pressure bearing capacity, the distance from the storage location to the storage entrance, the storage location space, the distance from the storage location to the storage exit and the storage location stability; wherein the multidimensional cell array is used to characterize the storage location characteristics of the storage location in the warehouse model, that is, the multidimensional cell array includes the storage location location, storage location pressure bearing capacity, the distance from the storage location to the storage entrance, the storage location space, the distance from the storage location to the storage exit and the storage location stability. The array of elements such as space, distance from the storage location to the exit, and storage location stability can be represented by F{x, y, z, r, t, n}, where x is the storage location's pressure bearing capacity, y is the storage location space, the storage location space is the available space of the storage location, z is the storage location stability, r is the storage location position (position coordinate), t is the distance from the storage location to the entrance, which is used to represent the time cost of placing goods. The longer the distance, the greater the time cost. n is the distance from the storage location to the exit, which is used to represent the energy consumption cost required for handling equipment to pick up goods when leaving the warehouse. The longer the distance, the greater the energy consumption cost. Understandably, the pressure-bearing capacity of the storage location can be obtained based on finite element analysis simulation calculations, and the acceptable pressure-bearing weight can be represented by a grade value. For example, it can be divided into 10 grades (Grade 10: the acceptable pressure-bearing capacity of the storage location is 100-90kg; Grade 9: the acceptable pressure-bearing capacity is 90-80kg; ... Grade 1: below 10kg). Similarly, the distance from the storage location to the storage entrance, the storage location space, the distance from the storage location to the storage exit, and the stability of the storage location can all be represented by grade values.

[0028] In this embodiment, the method of predicting the goods storage location and the corresponding handling equipment path of the goods in the incoming order based on the order prediction information and the warehouse model of the multi-layer shuttle intensive stereoscopic warehouse includes the following steps: S121. Calculate and predict the storage location of the goods in the incoming order based on the quantity, type, time of inbound and outbound goods, goods information and location characteristics of the locations in the warehouse model in the order prediction information.

[0029] This step may specifically include: detecting whether the goods in the incoming order are in the order prediction information; if so, comparing and matching the in-and-out quantity, in-and-out time and goods information of the goods in the order prediction information with the corresponding values ​​in the location characteristics of the location in the warehouse model, and calculating and predicting the storage location of the goods; understandably, during the comparison and matching, the difference between each value of the goods and the corresponding value in the location characteristics may be less than a preset threshold value as the target predicted storage location of the goods, and if there are too many predicted storage locations of the goods, the difference may be weightedly calculated to take the weighted average value less than the preset range of the goods storage location as the final predicted storage location of the goods The weights can be reconfigured according to the actual situation during weighted calculation. For example, during busy hours, the weight of the time cost of placing goods is the highest, which improves the overall efficiency, and the weight of the energy consumption cost during idle hours is increased to save the energy consumption of handling equipment. If not, the length of the future preset time period is extended, and the outbound orders within the future preset time period are re-predicted based on the historical order data until the goods in the inbound order appear in the predicted outbound order, and then the inbound and outbound quantity, inbound and outbound time and goods information of the goods in the predicted outbound order are compared and matched with the corresponding values ​​in the location characteristics of the location in the warehouse model to calculate the predicted storage location of the goods.

[0030] S122: predicting a corresponding transport equipment path according to the predicted distance between the cargo storage location and the transport equipment, the congestion situation, and the energy consumption of the transport equipment.

[0031] In this step, the congestion situation refers to the queuing and congestion of the handling equipment when the handling equipment is handling goods in the multi-layer shuttle-intensive stereoscopic warehouse. The congestion level can be represented by numbers 1, 2, ..., 8, etc., and the larger the number, the more serious the congestion situation; the energy consumption of the handling equipment refers to the energy consumption level comprehensively derived from factors such as the operating speed, load and slope of the path of the handling equipment. The energy consumption level can be represented by numbers 1, 2, ..., 8, etc. For example, if the operating speed is 3 meters per minute, the load exceeds 50KG, and the slope of the path is 30°, the energy consumption level can be 4.

[0032] It can be understood that when calculating the predicted storage location of goods, if the goods in the incoming order are in the outgoing order within the predicted future preset time period, the ingoing and outgoing quantity, ingoing and outgoing time and goods information of the goods in the order prediction information are compared and matched with the values ​​of the corresponding elements in the location characteristics of the storage location in the warehouse model, and the predicted storage location of the outgoing goods is calculated. When calculating the prediction, the goods in the same outgoing order are preferentially placed in the same regional storage location. If the goods in the incoming order are not in the outgoing order within the predicted future preset time period, the preset time period is extended and the prediction is made again until the goods in the incoming order appear in the predicted outgoing order, and the outgoing time, quantity and other information of the goods in the incoming order are obtained, so as to compare and match the values ​​of the corresponding elements in the location characteristics according to the above information. Understandably, goods in incoming orders that are frequently in and out in recent times are preferentially stored in shelf locations close to the entrances and exits. The weight and size of the goods, as well as the pressure-bearing capacity and space of the shelf locations, are considered when placing them. The storage area can also be further subdivided according to the distribution of their entry and exit times. For example, if a certain type of goods is frequently in and out of the warehouse in the morning, it will be stored in the shelf area close to the entrance and exit where operations are more convenient in the morning.

[0033] When predicting the corresponding transport equipment path, the A* algorithm combined with the collision detection mechanism can be used to implement path planning from the transport equipment to the storage location, so as to calculate a shorter and feasible transport equipment path. At the same time, the transport equipment path is adjusted according to the congestion situation and the energy consumption of the transport equipment on the path, so that the congestion situation and the energy consumption of the transport equipment on the path are respectively less than the preset levels during the scheduling process. That is, when a certain area in the transport equipment path shows signs of congestion (such as equipment waiting in line), the path planning is adjusted in time to guide the transport equipment to avoid the congested area. In the process of path planning, not only the shortest path should be pursued, but also the lowest energy consumption of the equipment should be taken into account. For example, under the same distance, the path with a smaller slope is given priority to reduce the energy consumption of the equipment.

[0034] Furthermore, before predicting the outbound orders within a preset time period in the future based on the historical order data to obtain the order prediction information, it also includes: querying whether the incoming order is bound to the outbound order, if not, executing step S110, and if so, predicting the cargo storage location and corresponding handling equipment path of the cargo in the incoming order based on the information of the bound outbound order and the warehouse model of the multi-layer shuttle intensive warehouse, and then performing prediction and calculation of the cargo storage location, cargo handling allocation plan and handling equipment path, thereby realizing efficient scheduling management of the cargo in the incoming order.

[0035] S130: construct multiple cargo handling allocation plans based on the handling equipment paths, the load balance of the handling equipment, and the urgency of the task.

[0036] In this step, a cargo handling allocation plan for allocating handling tasks to the handling equipment is constructed by comprehensively considering the handling equipment path, the load balancing degree of the handling equipment, and the urgency of the task. The load balancing degree of the handling equipment refers to whether the workload of each handling equipment in the warehouse is balanced. The workload includes the amount of completed tasks, the amount of tasks being executed, and the remaining carrying capacity of the equipment. When constructing the cargo handling allocation plan, tasks are allocated to equipment with lighter loads first to ensure that the workload of each equipment is roughly balanced to avoid the situation where some equipment is overworked while other equipment is idle. In addition, tasks are divided into different priorities (such as urgent, ordinary, deferrable, etc.) according to the urgency of the task. When allocating tasks, urgent tasks are given priority to ensure that urgent goods can be put in and out of the warehouse in time. For deferrable tasks, they can be flexibly arranged according to the idle time of the handling equipment.

[0037] S140, using an optimization algorithm to calculate the global optimal cargo storage location, cargo handling allocation plan and handling equipment path according to the cargo storage location, the corresponding handling equipment path and the cargo handling allocation plan.

[0038] In this step, the optimization algorithm is an intelligent optimization algorithm, which can be a genetic algorithm, a simulated annealing algorithm, a particle swarm optimization algorithm and a multi-objective ant colony algorithm. The optimization goal is to balance the cargo storage location, the cargo handling allocation plan and the handling equipment path, so that the cargo storage is reasonable, the handling path is efficient, the cargo handling allocation is balanced, and the overall optimization is achieved. The predicted cargo storage location, the corresponding handling equipment path and the cargo handling allocation plan are input into the pre-trained genetic algorithm model / simulated annealing algorithm model / particle swarm optimization algorithm model / multi-objective ant colony algorithm model for multi-objective optimization iteration to obtain the comprehensive global optimal cargo storage location, cargo handling allocation plan and handling equipment path.

[0039] S150: Performing warehousing scheduling for the goods in the warehousing order according to the globally optimal goods storage location and the corresponding goods handling allocation plan and handling equipment path.

[0040] In this step, the goods in the incoming order are scheduled for incoming goods according to the global optimal cargo storage location calculated by the optimization algorithm, the corresponding handling equipment path, and the cargo handling allocation plan. During the execution of the incoming goods scheduling, the handling equipment operates efficiently according to the handling equipment path to ensure that the goods are accurately placed in the designated cargo storage location, and the cargo handling allocation plan is executed evenly and efficiently among the various devices.

[0041] Furthermore, in the present invention, after executing step S150, it also includes: real-time monitoring of the operation status of the handling equipment, and dynamically adjusting the warehousing scheduling of goods according to the operation status of the handling equipment, that is, by setting sensors at key locations (such as shelf entrances and exits, key nodes of handling equipment, etc.), real-time data collection (such as the position of the handling equipment, the movement status of goods, the equipment operation speed, etc.), and it can also be combined with the camera equipment in the multi-layer shuttle car intensive three-dimensional warehouse to monitor the operation status of the handling equipment in real time, and compare and analyze with the predetermined cargo handling distribution plan, the handling equipment path and the cargo storage location. When it is found that the operation status deviates from the predetermined plan (such as equipment failure, temporary change of order, etc.), the dynamic adjustment mechanism is immediately started, and the optimization calculation is re-performed according to the real-time operation status, and a new scheduling plan is quickly generated to ensure the efficient execution of the management scheduling method.

[0042] In summary, the intelligent warehouse management method of the present invention deeply mines the rules in historical order data, predicts the outbound orders within a preset time period in the future, and combines the outbound order prediction with the warehouse model, and uses the intelligent optimization algorithm to comprehensively plan the goods storage location, goods handling distribution plan and handling equipment path of the goods in the inbound order with the goal of optimal overall efficiency, so that the goods storage location is more reasonable, and the goods closer to the outbound time are placed on the shelf closer to the outbound port, and the placed goods are found after matching with the location characteristics of the warehouse location. The handling equipment path is more scientific, which can avoid congestion, the goods handling distribution plan is balanced, and the workload of each handling equipment is relatively balanced, avoiding overload or long-term idleness, thereby improving the overall efficiency of warehouse management and scheduling, thereby ensuring that the goods in and out of the warehouse are more efficient and economical.

[0043] It should be noted that the steps in the method of the embodiment of the present invention can be adjusted in order, combined and deleted according to actual needs. The above is only a specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. An intelligent warehouse management method, applied to a multi-layer shuttle-intensive stereoscopic warehouse, characterized in that: The intelligent warehouse management method comprises: According to the historical order data, the outbound orders within the preset time period in the future are predicted to obtain the order prediction information; Predicting the cargo storage location and corresponding handling equipment path of the cargo in the incoming order based on the order prediction information and the warehouse model of the multi-layer shuttle intensive stereoscopic warehouse; Construct multiple cargo handling allocation plans based on the handling equipment paths, load balance of the handling equipment, and the urgency of the task; According to the cargo storage location, the corresponding handling equipment path and the cargo handling allocation plan, the optimization algorithm is used to calculate the global optimal cargo storage location, cargo handling allocation plan and handling equipment path; The goods in the warehousing order are scheduled for warehousing according to the globally optimal goods storage location and the corresponding goods handling allocation plan and handling equipment path.

2. The intelligent warehouse management method according to claim 1, characterized in that: The forecasting of outbound orders within a preset time period in the future based on historical order data specifically includes: Use the ARIMA model to perform autocorrelation analysis and partial autocorrelation analysis on historical order data to obtain the time trend and seasonal fluctuation of historical order data, and obtain an order delivery forecasting model that can be used to predict future demand for goods; Use the outbound order forecasting model to predict outbound orders within a preset time period in the future based on seasons and / or market activities.

3. The intelligent warehouse management method according to claim 1 or 2, characterized in that: The forecasting of outbound orders within a preset time period in the future based on historical order data specifically includes: The decision tree algorithm is used to train the order time, goods type, quantity and goods information in the historical order data to obtain an order delivery prediction model that can be used to classify goods and predict future demand for goods; The order outbound forecasting model is used to predict outbound orders within a preset time period in the future.

4. The intelligent warehouse management method according to claim 1, characterized in that: The optimization algorithms are genetic algorithm, simulated annealing algorithm, particle swarm optimization algorithm and multi-objective ant colony algorithm.

5. The intelligent warehouse management method according to claim 1, characterized in that: Before predicting the storage location and corresponding handling equipment path of the goods in the incoming order based on the order prediction information and the warehouse model of the multi-layer shuttle intensive stereoscopic warehouse, the method further includes: Constructing a warehouse model according to the shelf arrangement, storage location distribution and handling equipment distribution of the multi-layer shuttle intensive stereoscopic warehouse; A multidimensional cell array is generated according to the storage location, the pressure bearing capacity of the storage location, the distance from the storage location to the storage entrance, the storage location space, the distance from the storage location to the storage exit, and the storage location stability; wherein the multidimensional cell array is used to characterize the storage location characteristics of the storage location in the warehouse model.

6. The intelligent warehouse management method according to claim 5, characterized in that: The predicting of the goods storage location and the corresponding handling equipment path of the goods in the incoming order based on the order prediction information and the warehouse model of the multi-layer shuttle intensive stereoscopic warehouse specifically includes: Calculate and predict the storage location of the goods in the incoming order based on the quantity, type, time of inbound and outbound goods, goods information and the location characteristics of the locations in the warehouse model in the order forecast information; The corresponding transport equipment path is predicted based on the predicted distance between the cargo storage location and the transport equipment, as well as the congestion situation and the energy consumption of the transport equipment.

7. The intelligent warehouse management method according to claim 6, characterized in that: The calculating and predicting of the storage location of the goods in the incoming order according to the quantity, type, time of inbound and outbound goods, goods information and the location characteristics of the locations in the warehouse model in the order forecast information specifically includes: Check whether the goods in the incoming order are within the order forecast information; If so, the in-and-out quantity, in-and-out time and goods information of the goods in the order forecast information are compared and matched with the corresponding values ​​in the location characteristics of the location in the warehouse model to calculate the predicted storage location of the goods; If not, extend the duration of the future preset time period, and re-predict the outbound orders within the future preset time period based on the historical order data until the goods in the inbound order appear in the predicted outbound order, and then compare and match the inbound and outbound quantity, inbound and outbound time and goods information of the goods in the predicted outbound order with the corresponding values ​​in the location characteristics of the location in the warehouse model to calculate the predicted storage location of the goods.

8. The intelligent warehouse management method according to claim 1, characterized in that: Before predicting the outbound orders within a preset time period in the future based on the historical order data to obtain the order forecast information, it also includes: querying whether the incoming order is bound to an outbound order, and if not, executing the step of predicting the outbound orders within a preset time period in the future based on the historical order data to obtain the order forecast information.

9. The intelligent warehouse management method according to claim 1, characterized in that: Before predicting the outbound orders within a preset time period in the future based on the historical order data to obtain the order prediction information, it also includes: obtaining the historical order data, cleaning the historical order data, and normalizing the cleaned historical order data.

10. The intelligent warehouse management method according to claim 1, characterized in that: After the goods in the warehousing order are scheduled for entry according to the global optimal cargo storage location and the corresponding cargo handling allocation plan and the handling equipment path, it also includes: real-time monitoring of the operation status of the handling equipment, and dynamically adjusting the warehousing scheduling of the goods according to the operation status of the handling equipment.

Images