Goods dispatching and distribution system based on stocker storage vertical warehouse

Abstract

The invention relates to a goods dispatching and distribution system based on a stocker storage three-dimensional warehouse, and the system comprises a cloud cooperative dispatching center which is in communication connection with a goods aging prediction module which is provided with a goods aging prediction model and is used for accessing e-commerce promotion activity calendar data, and according to the historical warehouse-out data of warehouse-in goods, the goods aging prediction module is used for predicting the goods aging; predicting and generating a warehouse-out time interval of the warehouse-in goods; the cargo information management module is used for managing warehouse-in and warehouse-out information of cargos; the dynamic partition management module is used for dividing the stereoscopic warehouse into corresponding physical storage areas according to the warehouse-out time interval labels and dynamically expanding the physical storage capacity of a top-speed area in response to a great promotion activity demand; and the load balancing module is used for monitoring the operation load state of the stacker in each physical storage area in real time and dynamically allocating the stacker resources according to the load state. According to the invention, the dispatching efficiency of the stereoscopic warehouse for peak logistics can be effectively improved.

Description

Technical Field

[0001] This application relates to the technical field of goods scheduling, and in particular to a goods scheduling and distribution system based on a stacker storage vertical warehouse, which is particularly suitable for efficient scheduling management during peak periods of goods inbound and outbound, such as e-commerce promotions and shopping festivals. Background Art

[0002] With the continuous development and growth of Internet e-commerce, the resulting demand for goods distribution in e-commerce transactions is also increasing. The huge number of goods poses a great challenge to the implementation of scheduling and distribution. Especially during promotional activities on e-commerce platforms, the scheduling pressure of the stacker storage vertical warehouse reaches its peak. The huge number of goods places higher requirements on the scheduling and distribution system. How to optimize the warehousing efficiency during peak periods has become the core pain point of the industry.

[0003] Chinese Patent with Publication No. CN117635027B discloses a goods scheduling and distribution system based on a stacker storage vertical warehouse, which relates to the technical field of goods scheduling. It includes a cloud management platform, which is communicatively connected to a goods control module, a data interaction module, and a user terminal module; the goods control module is provided with an information entry unit, a stacker control unit, and a goods storage unit; through the information entry unit, information of a number of goods is entered, through the stacker control unit, a number of stackers are controlled to perform pick-up and placement operations on the goods, and through the goods storage unit, a number of goods partitions are set for stacking goods in partitions; through the data interaction module, data interaction of a number of goods partitions is carried out, and the interaction environment is monitored in real time; through the user terminal module, user identity verification is carried out. When the verification is passed, a demand definition is generated to generate a corresponding demand text, and demand interaction is carried out through the demand text. When the verification fails, the user is marked as a blacklist user. This application adopts a fixed goods partition strategy. According to the expected delivery time and reminder frequency of users, the goods scheduling priority is set to determine the goods scheduling order, and the stacker control unit performs pick-up and placement operations in the order of priority. However, such operations, although they can meet the scheduling needs of users to a certain extent usually, are prone to having popular goods with high-frequency outbound and slow-moving goods with long-term storage mixed in fixed partitions during scheduling peaks. The stacker needs to travel back and forth between the deep layer of the vertical warehouse and the outbound port, resulting in a long pick-up path and low efficiency. And only setting the scheduling priority according to user needs is prone to causing the scheduling needs of popular goods that need to be quickly inbound and outbound to compete for resources with regular tasks, easily leading to uneven equipment utilization, resulting in chaotic and inefficient goods scheduling.

[0004] Regarding the above related technologies, the inventor believes that the goods scheduling of existing stacker storage vertical warehouses cannot meet the requirements of e-commerce peak logistics scheduling. Summary of the Invention

[0005] To solve the above problems, the present application provides a goods scheduling and distribution system based on a stacker storage vertical warehouse.

[0006] In a first aspect, the present application provides a goods scheduling and distribution system based on a stacker storage vertical warehouse, adopting the following technical solutions: A goods scheduling and distribution system based on a stacker storage vertical warehouse includes a cloud collaborative scheduling center, which is communicatively connected to: A goods timeliness prediction module, configured with a goods timeliness prediction model, for accessing e-commerce promotion activity calendar data and predicting and generating an outbound time interval of inbound goods according to the historical outbound data of the inbound goods. The outbound time interval at least includes: an express outbound interval, a fast outbound interval, and a long-term storage interval; the goods timeliness prediction model is an iterative training obtained by a machine learning model through historical inbound and outbound data and e-commerce promotion data; A goods information management module, for managing the inbound and outbound information of goods and assigning a predicted outbound time interval label to the inbound goods; A dynamic partition management module, for dividing the stereoscopic warehouse into corresponding physical storage areas according to the outbound time interval label, including an express area, a fast area, and a long-term storage area, and dynamically expanding the physical storage capacity of the express area in response to the needs of major promotion activities; A goods scheduling execution module, for controlling the stacker to execute the inbound and outbound operations of goods; A load balancing module, for real-time monitoring of the operating load status of the stacker in each physical storage area and dynamically allocating stacker resources according to the load status.

[0007] Preferably, the express area divided by the dynamic partition management module is located at the closest end to the inbound and outbound ports of the stereoscopic warehouse; the long-term storage area is located in the depth area of the stereoscopic warehouse.

[0008] Preferably, the dynamic partition management module includes multiple groups of movable shelf modules and electric tracks laid on the floor of the stereoscopic warehouse. When the occupancy rate of the inventory capacity in the express area is greater than a preset first inventory threshold, the cloud collaborative scheduling center controls the electric tracks to drive multiple groups of movable shelf modules to move to the side of the express area close to the inbound and outbound ports of the stereoscopic warehouse to expand the express area and update the inventory layout in real time.

[0009] Preferably, the electric tracks are grid-shaped electric tracks, and multiple groups of movable shelf modules are located in the long-term storage area. The cloud collaborative scheduling center performs dynamic partitioning on multiple groups of movable shelf modules based on the inventory capacity of each physical storage area in the stereoscopic warehouse.

[0010] Preferably, the cloud collaborative scheduling center performing dynamic partitioning on multiple groups of movable shelf modules based on the inventory capacity of each physical storage area in the stereoscopic warehouse specifically includes the following steps: A1. The cloud collaborative scheduling center determines in real time whether the occupancy rate of the inventory capacity in the express area is greater than a preset first inventory threshold; A2. If it is greater, control the electric track to drive multiple groups of movable rack modules to move to one side of the express area near the inlet / outlet of the stereoscopic warehouse to expand the express area; A3. If it is not greater, the cloud collaborative scheduling center obtains the in / outbound orders in real time to predict whether there is a peak moment when the occupancy rate of the inventory capacity in the express area exceeds the first inventory threshold in the next storage cycle; A4. If there is, control the electric track to drive multiple groups of movable rack modules to move to one side of the express area near the inlet / outlet of the stereoscopic warehouse to expand the express area; A5. If there is not, the cloud collaborative scheduling center determines in real time whether the occupancy rate of the inventory capacity in the long-term storage area is greater than a preset second inventory threshold; A6. If it is not greater, keep multiple groups of movable rack modules in a fixed state and designate them as a standby area without storing goods; A7. If it is greater, the cloud collaborative scheduling center accesses the e-commerce promotion activity calendar data to determine whether there is an e-commerce promotion activity in the next storage cycle; A8. If there is not, divide groups of movable rack modules into the long-term storage area in sequence until the occupancy rate of the inventory capacity in the new long-term storage area is less than the preset second inventory threshold, and keep the remaining movable rack modules in a fixed state and designate them as a standby area without storing goods; A9. If there is, the cloud collaborative scheduling center determines in real time whether the occupancy rate of the inventory capacity in the long-term storage area exceeds a preset third inventory threshold, where the third inventory threshold is greater than the second inventory threshold; A10. If it exceeds, divide groups of movable rack modules into the long-term storage area in sequence until the occupancy rate of the inventory capacity in the new long-term storage area is less than the preset second inventory threshold, and keep the remaining movable rack modules in a fixed state and designate them as a standby area without storing goods; A11. If it does not exceed, keep multiple groups of movable rack modules in a fixed state and designate them as a standby area without storing goods.

[0011] Preferably, steps A8 and A10 further include: if all groups of movable rack modules are divided into the long-term storage area and the occupancy rate of the inventory capacity in the long-term storage area is still not less than the preset second inventory threshold, generate an alarm message indicating insufficient number of movable rack modules and send it to the management personnel.

[0012] Preferably, steps A8 and A10 further include: when the occupancy rate of the inventory capacity in the long-term storage area is less than a preset fourth inventory threshold, the cloud collaborative scheduling center generates a clearance instruction to control the stacker in the long-term storage area to run during idle time, relocate the goods stored on the movable rack module in the long-term storage area to the fixed rack in the long-term storage area, and move the movable rack module in the long-term storage area out of the long-term storage area after the relocation is completed.

[0013] Preferably, the goods information management module includes: An information collection unit, including an RFID reader / writer and a barcode scanner, for real-time collecting the identity and status information of goods; A label processing unit: for receiving the outbound time interval label generated by the goods aging prediction module and writing the label into the RFID chip of the goods or associating it with the goods database entry; An inventory mapping unit, for establishing a dynamic association database of the goods position information including three-dimensional coordinates and the outbound time interval label, and its data model is: [Goods ID, shelf coordinates (X, Y, Z), outbound interval label, storage environment parameters, last operation timestamp]; A status monitoring unit: by deploying an Internet of Things sensor array on the shelf, real-time monitoring the storage environment parameters of the goods; A data storage unit, for establishing a basic goods database for storing the identity of goods and the inbound and outbound order information; And the basic goods database and the dynamic association database form a master-slave architecture, and a two-way real-time synchronization channel is established between the two databases through the goods ID.

[0014] Preferably, when the load balancing module performs resource allocation, the following strategy is adopted: When it is detected that the busy / idle rate of the stacker in any physical storage area is greater than 85% or the backlog length of the conveyor belt exceeds a preset threshold L, a cross-region device support mechanism is triggered, and the idle stackers with a preset support quantity are retrieved in the order of the long-term storage area, the fast area, and the extreme speed area to cooperate in performing the distribution tasks in the triggered area.

[0015] Preferably, the cooperation of the idle stackers in performing the distribution tasks in the triggered area specifically includes: after the cloud collaborative scheduling center triggers the cross-region device support mechanism, reallocating the distribution tasks in the triggered area based on the shortest job first algorithm, and planning the stacker paths based on the distribution tasks of each stacker by using the Dijkstra algorithm and sending them to the goods scheduling execution module to control the stacker to perform the inbound and outbound operations of the goods.

[0016] In summary, the present application includes at least one of the following beneficial technical effects: 1. Integrate five major modules through the cloud collaborative scheduling center to build a data-driven intelligent warehousing closed-loop system. Based on the machine learning model, fuse historical data and promotion calendars to accurately predict the outbound time of goods. At the same time, dynamically divide the three-dimensional warehouse into regions, so that high-frequency goods are always located at the optimal storage and retrieval positions, greatly reducing the average picking time, which can greatly relieve the scheduling pressure of the three-dimensional warehouse facing peak logistics and effectively improve the scheduling efficiency of the three-dimensional warehouse facing peak logistics; 2. Trigger support based on the dual indicators of busyness rate and conveyor belt backlog, and preferentially call equipment in low-priority areas for cross-regional support, further improving the anti-pressure and fault-tolerant rate of the extreme zone of the three-dimensional warehouse facing the peak of goods scheduling, and realizing the maximization of the reuse of stacker equipment; 3. Form a stepped response strategy based on the first / second / third / fourth inventory thresholds to balance real-time demand and future predictions, constituting an elastic defense system, and realizing the intelligent dynamic allocation of the storage capacity of the three-dimensional warehouse; according to the inventory capacity of the extreme zone, automatically trigger expansion, dynamically schedule the expandable mobile rack module to expand the extreme zone, and at the same time, schedule the module in advance according to order prediction to avoid sudden out-of-stock during major promotions; when the inventory in the extreme zone is stable, according to the inventory capacity of the long-term storage area, combine with e-commerce promotions to adapt to the promotion cycle, dynamically adjust the capacity of the long-term area, and balance the daily storage and major promotion stocking requirements; realize the fine dynamic allocation of storage resources, and effectively improve the stability and universality of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a system block diagram of a goods scheduling and distribution system based on a stacker storage vertical warehouse in Embodiment 1 of the present application; Figure 2 is a system block diagram of the goods information management module in Embodiment 1 of the present application; Figure 3 is a system block diagram of the dynamic partition management module in Embodiment 1 of the present application; Figure 4 is a method flow chart for dynamically partitioning multiple groups of movable rack modules in Embodiment 2 of the present application.

[0018] Description of reference numerals: 1. Goods aging prediction module; 2. Goods information management module; 21. Information collection unit; 22. Label processing unit; 23. Inventory mapping unit; 24. Status monitoring unit; 25. Data storage unit; 3. Dynamic partition management module; 31. Movable rack module; 32. Electric track; 4. Goods scheduling execution module; 5. Load balancing module; 6. Cloud collaborative scheduling center. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The following will be further described in detail with reference to the accompanying Figures 1-4 drawings to further illustrate the present application.

[0020] Embodiment 1

[0021] This application embodiment discloses a goods scheduling and distribution system based on a stacker storage vertical warehouse. Refer to Figure 1 , a goods scheduling and distribution system based on a stacker storage vertical warehouse, includes a cloud collaborative scheduling center 6. The cloud collaborative scheduling center 6 is a cloud management platform. The cloud collaborative scheduling center 6 is communicatively connected to: A goods timeliness prediction module 1, configured with a goods timeliness prediction model, for accessing e-commerce promotion activity calendar data, and predicting and generating an outbound time interval of inbound goods according to the historical outbound data of the inbound goods. The outbound time interval at least includes: an express outbound interval, a fast outbound interval, and a long-term storage interval; the goods timeliness prediction model is an iterative training obtained by a machine learning model through historical inbound and outbound data and e-commerce promotion data; it should be noted that the specific training steps of the machine learning model are prior art and will not be elaborated here; in addition, the machine learning model preferably uses an LSTM neural network model, and due to its modeling ability for long sequence dependence relationships, it becomes a preferred solution for goods timeliness prediction; A goods information management module 2, for managing the inbound and outbound information of goods, and assigning a predicted outbound time interval label to the inbound goods; A dynamic partition management module 3, for dividing the three-dimensional warehouse into corresponding physical storage areas according to the outbound time interval label, including an express area, a fast area, and a long-term storage area, and dynamically expanding the physical storage capacity of the express area in response to the needs of large promotion activities; A goods scheduling execution module 4, for controlling the stacker to execute the inbound and outbound operations of goods; A load balancing module 5, for real-time monitoring of the operating load status of the stacker in each physical storage area, and dynamically allocating stacker resources according to the load status. Through the cloud collaborative scheduling center 6 integrating the five modules, a data-driven intelligent warehousing closed-loop system is constructed. Based on the machine learning model, historical data and promotion calendars are fused to accurately predict the outbound timeliness of goods (express / fast / long-term). At the same time, the three-dimensional warehouse is dynamically partitioned, so that high-frequency goods are always located at the optimal access positions, greatly reducing the average picking time, and being able to greatly relieve the scheduling pressure of the three-dimensional warehouse facing peak logistics, and effectively improving the scheduling efficiency of the three-dimensional warehouse facing peak logistics.

[0022] Refer to Figure 1 and Figure 2 , the goods information management module 2 includes: An information collection unit 21, including an RFID reader / writer, a barcode scanner, and other auxiliary devices capable of collecting goods, for real-time collecting goods identity and status information; Label processing unit 22: It is used to receive the outbound time interval label generated by the goods timeliness prediction module 1 and write the label into the RFID chip of the goods or associate it with the goods database entry; Inventory mapping unit 23, which is used to establish a dynamic association database of the goods location information including three-dimensional coordinates and the outbound time interval label, and its data model is: [Goods ID, shelf coordinates (X, Y, Z), outbound interval label, storage environment parameters, last operation timestamp]; Status monitoring unit 24: Through the Internet of Things sensor array deployed on the shelves, it monitors the storage environment parameters of the goods in real time; Data storage unit 25, which is used to establish a basic goods database for storing the identity of the goods and the inbound and outbound order information; and the basic goods database and the dynamic association database form a master-slave architecture, and the two databases establish a two-way real-time synchronization channel through the goods ID. Through the setting of the goods information management module 2, on the basis of efficiently collecting and managing the inbound and outbound goods information, forming labels based on the outbound time interval generated by the goods timeliness prediction module 1, adopting a dual-database separation design, decoupling the static attributes and dynamic states of the goods, and the two-way synchronization of the basic database and the dynamic association database to ensure the consistency of inventory data, improve the query efficiency, reduce the scheduling errors caused by information lag, and at the same time, the separation of static and dynamic realizes incremental synchronization instead of full polling, reducing 70% of the synchronization traffic.

[0023] Refer to Figure 1 and Figure 3 , the express area divided by the dynamic partition management module 3 is located at the closest end of the inbound and outbound ports of the automated warehouse, and a high-density shelf layout is adopted; the long-term storage area is located in the depth area of the automated warehouse, and the rest is divided into the fast area. Among them, the inventory ratios of the express area, the fast area and the long-term storage area in the automated warehouse are set by the management personnel. In this embodiment, the ratios of the express area, the fast area and the long-term storage area are set to: 3:2:4. The dynamic partition management module 3 partitions the automated warehouse through the differential design of the spatial location, significantly shortening the access path of high-frequency goods, enabling fast in and out for popular electric products, and being able to greatly relieve the scheduling pressure of the automated warehouse facing peak logistics, so as to effectively improve the scheduling efficiency of the automated warehouse facing peak logistics. While low-frequency goods are stored in the remote area (the depth of the automated warehouse) to reduce the interference to the core operation area, improve the space utilization rate, and reduce the overall operation conflict rate of the automated warehouse.

[0024] The dynamic partition management module 3 includes multiple groups of movable rack modules 31 and electric tracks 32 laid on the floor of the automated warehouse. When the occupancy rate of the inventory capacity in the extreme speed area is greater than a preset first inventory threshold (set to 80% in this example), the cloud collaborative scheduling center 6 controls the electric tracks 32 to drive multiple groups of movable rack modules 31 to move to one side of the extreme speed area near the inlet and outlet of the automated warehouse to expand the extreme speed area, and updates the inventory layout in real time. Through the setting of the movable rack modules 31 and the electric tracks 32, flexible expansion of the extreme speed area is achieved. According to the inventory capacity occupancy rate threshold, expansion is automatically triggered, avoiding delays in manual intervention, and greatly improving the anti-pressure fault tolerance rate of the automated warehouse in the face of peak cargo scheduling.

[0025] When the above load balancing module 5 performs resource allocation, the following strategy is adopted: When it is detected that the busy rate of any stacker in a physical storage area is greater than 85% or the backlog length of the conveyor belt exceeds a preset threshold L, a cross-region equipment support mechanism is triggered, and idle stackers with preset support amounts are retrieved in the order of the long-term storage area, the fast area, and the extreme speed area to cooperate in performing the distribution tasks in the triggered area. Based on the dual indicators of the busy rate (>85%) and the conveyor belt backlog, support is triggered, and equipment in the low-priority area is preferentially called (long-term storage area → fast area → extreme speed area), further improving the anti-pressure fault tolerance rate of the extreme speed area of the automated warehouse in the face of peak cargo scheduling and achieving the maximum reuse of stacker equipment.

[0026] It can be expanded that when facing an automated warehouse with a high-level architecture, a multi-layer shuttle car subsystem can be additionally set up to cooperate with the stacker to achieve more efficient cargo storage and retrieval, further improving the cargo scheduling efficiency.

[0027] In addition, the specific process of the idle stackers cooperating in performing the distribution tasks in the triggered area includes: after the cloud collaborative scheduling center 6 triggers the cross-region equipment support mechanism, it reallocates the distribution tasks in the triggered area based on the shortest job first algorithm (SJF), and uses the Dijkstra algorithm to plan the stacker paths based on the distribution tasks of each stacker and send them to the cargo scheduling execution module 4 to control the stacker to perform the inbound and outbound operations of the cargo. For the tasks in the triggered area, especially for high-frequency small-piece orders in e-commerce, through the innovation of algorithm combination (SJF + Dijkstra), the leap from "local optimization" to "global intelligence" in warehousing task scheduling is achieved, which not only meets the "fast response" requirement during the e-commerce peak period, but also reduces the operation cost through path optimization, achieving the efficient operation effect of a high-density automated warehouse.

[0028] Embodiment 2

[0029] In this preferred embodiment: The electric track 32 is a grid-shaped electric track 32. Multiple groups of movable rack modules 31 are located in the long-term storage area. The cloud collaborative scheduling center 6 performs dynamic zoning on multiple groups of movable rack modules 31 based on the inventory capacity of each physical storage area in the automated warehouse. The grid-shaped electric track 32 improves the moving flexibility of the movable rack modules 31, enabling the dynamic zoning management module 3 to support dynamic zoning in complex scenarios. The idle modules are pre-stored in the long-term storage area and used as backup resources during off-peak periods to avoid hardware waste.

[0030] Referring Figure 4 , the cloud collaborative scheduling center 6 performing dynamic zoning on multiple groups of movable rack modules 31 based on the inventory capacity of each physical storage area in the automated warehouse specifically includes the following steps: A1. The cloud collaborative scheduling center 6 continuously determines whether the occupancy rate of the inventory capacity in the express area is greater than a preset first inventory threshold; the first inventory threshold is set by the management personnel according to the actual situation of the automated warehouse. For example, for an automated warehouse with a large storage capacity, its error tolerance is high, and the first inventory threshold can be set higher. For a small-scale automated warehouse with a small storage capacity, in order to ensure the stability of the express area, it should be set lower. In this embodiment, the first inventory threshold is 80%; A2. If it is greater, control the electric track 32 to drive multiple groups of movable rack modules 31 to move to one side of the express area close to the loading and unloading opening of the automated warehouse to expand the express area; A3. If it is not greater, the cloud collaborative scheduling center 6 continuously obtains whether there is a peak moment when the occupancy rate of the inventory capacity in the express area exceeds the first inventory threshold in the next storage cycle for the incoming / outgoing orders; A4. If there is, control the electric track 32 to drive multiple groups of movable rack modules 31 to move to one side of the express area close to the loading and unloading opening of the automated warehouse to expand the express area; A5. If there is not, the cloud collaborative scheduling center 6 continuously determines whether the occupancy rate of the inventory capacity in the long-term storage area is greater than a preset second inventory threshold; as described above, the second inventory threshold is also set by the management personnel according to the actual situation of the automated warehouse. In this embodiment, the second inventory threshold is 85%; A6. If it is not greater, keep multiple groups of movable rack modules 31 in a fixed state and classify them as a standby area without storing goods; A7. If it is greater, the cloud collaborative scheduling center 6 accesses the e-commerce promotion activity calendar data to determine whether there is an e-commerce promotion activity in the next storage cycle; A8. If there is not, successively classify one group of movable rack modules 31 into the long-term storage area until the occupancy rate of the inventory capacity in the new long-term storage area is less than the preset second inventory threshold, and keep the remaining movable rack modules 31 in a fixed state and classify them as a standby area without storing goods; A9. If so, the cloud collaborative scheduling center 6 determines in real time whether the inventory capacity occupancy rate of the long-term storage area exceeds a preset third inventory threshold, where the third inventory threshold is greater than the second inventory threshold; as described above, the third inventory threshold is also set by the management personnel according to the actual situation of the stereoscopic warehouse. In this embodiment, the third inventory threshold is 95%. A10. If it exceeds, the movable rack modules 31 are successively divided into the long-term storage area until the inventory capacity occupancy rate of the new long-term storage area is less than the preset second inventory threshold, and the remaining movable rack modules 31 are kept fixed and designated as a spare area without storing goods. A11. If it does not exceed, the multiple groups of movable rack modules 31 are kept fixed and designated as a spare area without storing goods. Through the above steps, a stepped response strategy is formed based on the first / second / third / fourth inventory thresholds to balance the real-time demand and future prediction, constituting an elastic defense system, and realizing the intelligent dynamic allocation of the storage capacity of the stereoscopic warehouse; according to the inventory capacity of the extreme speed area, the expansion is automatically triggered, and the movable rack modules are dynamically scheduled to expand the extreme speed area. At the same time, the modules are scheduled in advance according to the order prediction to avoid sudden out-of-stock during peak sales; when the inventory in the extreme speed area is stable, according to the inventory capacity of the long-term storage area, the promotion cycle is adapted in combination with e-commerce promotions, and the capacity of the long-term area is dynamically adjusted to balance the daily storage and peak sales stocking requirements; realizing the fine dynamic allocation of storage resources, and achieving the effect of effectively improving the system stability and universality.

[0031] Among them, the above steps A8 and A10 also include: if all the movable rack modules 31 are divided into the long-term storage area and the inventory capacity occupancy rate of the long-term storage area is still not less than the preset second inventory threshold, an alarm message indicating insufficient number of movable rack modules 31 is generated and sent to the management personnel. When the expansion demand exceeds the hardware resources, a real-time alarm is given to prompt manual intervention to avoid order delays and ensure the goods scheduling efficiency of the system.

[0032] In addition, the above steps A8 and A10 also include: when the inventory capacity occupancy rate of the long-term storage area is less than the preset fourth inventory threshold, the cloud collaborative scheduling center 6 generates a clearing area instruction to control the stacker in the long-term storage area to operate during idle time (when there is no task), relocate the goods stored on the movable rack modules 31 in the long-term storage area to the fixed racks in the long-term storage area, and move out the movable rack modules 31 in the long-term storage area after the relocation is completed. The fourth inventory threshold is set by the management personnel, and in this implementation, it is set to 60%. When the inventory occupancy rate in the long-term area is lower than the fourth threshold, the goods are concentrated on the fixed racks, and the released movable modules can be used as a general resource pool to support the expansion of the extreme speed area or the fast area at any time, further improving the overall space utilization rate of the stereoscopic warehouse and ensuring the anti-pressure performance of the system facing peak logistics.

[0033] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than limiting the protection scope of the invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on these embodiments, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art can still, without conflict and without creative efforts, combine, add or delete the features in the embodiments of the present invention according to the circumstances or make other adjustments, so as to obtain different technical solutions that essentially do not depart from the concept of the present invention, and these technical solutions also belong to the scope of protection of the present invention.

Claims

1. A goods scheduling and distribution system based on a storage vertical warehouse with a stacker, including a cloud collaborative scheduling center (6), characterized in that, The cloud collaborative scheduling center (6) is communicatively connected to: A goods timeliness prediction module (1), configured with a goods timeliness prediction model, for accessing e-commerce promotion activity calendar data, and predicting and generating an outbound time interval for inbound goods based on the historical outbound data of the inbound goods. The outbound time interval at least includes: an express outbound interval, a fast outbound interval, and a long-term storage interval; the goods timeliness prediction model is an iterative training model obtained by a machine learning model through historical inbound and outbound data and e-commerce promotion data; A goods information management module (2), for managing the inbound and outbound information of goods, and assigning a predicted outbound time interval label to the inbound goods; A dynamic partition management module (3), for dividing the three-dimensional warehouse into corresponding physical storage areas according to the outbound time interval label, including an express area, a fast area, and a long-term storage area, and dynamically expanding the physical storage capacity of the express area in response to the demand of large promotion activities; A goods scheduling execution module (4), for controlling the stacker to execute the inbound and outbound operations of goods; A load balancing module (5), for real-time monitoring of the operating load status of the stacker in each physical storage area, and dynamically allocating stacker resources according to the load status.

2. The goods scheduling and distribution system based on the stacker storage vertical warehouse according to claim 1, wherein, The express area divided by the dynamic partition management module (3) is located at the closest end to the inlet and outlet of the three-dimensional warehouse; the long-term storage area is located in the depth area of the three-dimensional warehouse.

3. The goods scheduling and distribution system based on the stacker storage vertical warehouse according to claim 2, characterized in that: The dynamic partition management module (3) includes multiple groups of movable shelf modules (31) and an electric track (32) laid on the floor of the three-dimensional warehouse. When the occupancy rate of the inventory capacity in the express area is greater than a preset first inventory threshold, the cloud collaborative scheduling center (6) controls the electric track (32) to drive the multiple groups of movable shelf modules (31) to move to one side of the express area close to the inlet and outlet of the three-dimensional warehouse to expand the express area, and updates the inventory layout in real time.

4. The goods scheduling and distribution system based on the stacker storage vertical warehouse according to claim 3, characterized in that: The electric track (32) is a grid-shaped electric track (32), and multiple groups of movable shelf modules (31) are located in the long-term storage area. The cloud collaborative scheduling center (6) dynamically partitions the multiple groups of movable shelf modules (31) based on the inventory capacity of each physical storage area in the three-dimensional warehouse.

5. The goods scheduling and distribution system based on a stacker storage vertical warehouse according to claim 4, wherein The cloud collaborative scheduling center (6) dynamically partitioning the multiple groups of movable shelf modules (31) based on the inventory capacity of each physical storage area in the three-dimensional warehouse specifically includes the following steps: A1. The cloud collaborative scheduling center (6) determines in real time whether the occupancy rate of the inventory capacity in the express area is greater than a preset first inventory threshold; A2. If it is greater, control the electric track (32) to drive the multiple groups of movable shelf modules (31) to move to one side of the express area close to the inlet and outlet of the three-dimensional warehouse to expand the express area; A3. If it is not greater, the cloud collaborative scheduling center (6) obtains in real time whether there is a peak moment when the occupancy rate of the inventory capacity in the express area exceeds the first inventory threshold in the predicted next storage cycle of the inbound / outbound order; A4. If there is, control the electric track (32) to drive the multiple groups of movable shelf modules (31) to move to one side of the express area close to the inlet and outlet of the three-dimensional warehouse to expand the express area; A5. If not, the cloud collaborative scheduling center (6) determines in real time whether the inventory capacity occupancy rate of the long-term storage area is greater than a preset second inventory threshold; A6. If not greater, keep the multi-group movable rack modules (31) in a fixed state and designate them as a standby area without storing goods; A7. If greater, the cloud collaborative scheduling center (6) accesses the e-commerce promotion activity calendar data to determine whether there is an e-commerce promotion activity in the next storage cycle; A8. If not, successively divide groups of movable rack modules (31) into the long-term storage area until the inventory capacity occupancy rate of the new long-term storage area is less than the preset second inventory threshold, and keep the remaining movable rack modules (31) in a fixed state and designate them as a standby area without storing goods; A9. If there is, the cloud collaborative scheduling center (6) determines in real time whether the inventory capacity occupancy rate of the long-term storage area exceeds a preset third inventory threshold, where the third inventory threshold is greater than the second inventory threshold; A10. If it exceeds, successively divide groups of movable rack modules (31) into the long-term storage area until the inventory capacity occupancy rate of the new long-term storage area is less than the preset second inventory threshold, and keep the remaining movable rack modules (31) in a fixed state and designate them as a standby area without storing goods; A11. If it does not exceed, keep the multi-group movable rack modules (31) in a fixed state and designate them as a standby area without storing goods.

6. The goods scheduling and distribution system based on a storage vertical warehouse with a stacker according to claim 5, characterized in that, The steps A8 and A10 further include: if all the multi-group movable rack modules (31) are divided into the long-term storage area and the inventory capacity occupancy rate of the long-term storage area is still not less than the preset second inventory threshold, generate an alarm message indicating insufficient quantity of movable rack modules (31) and send it to the management personnel.

7. A goods scheduling and distribution system based on a stacker storage vertical warehouse according to claim 5, characterized in that, The steps A8 and A10 further include: when the inventory capacity occupancy rate of the long-term storage area is less than a preset fourth inventory threshold, the cloud collaborative scheduling center (6) generates a zone clearing instruction to control the stacker in the long-term storage area to operate during idle time, relocate the goods stored on the movable rack modules (31) in the long-term storage area to the fixed racks in the long-term storage area, and remove the movable rack modules (31) in the long-term storage area from the long-term storage area after the relocation is completed.

8. A goods scheduling and distribution system based on a stacker storage vertical warehouse according to claim 1, characterized in that: The goods information management module (2) includes: An information collection unit (21), including an RFID reader and a barcode scanner, for collecting the identity and status information of goods in real time; A label processing unit (22): for receiving the outbound time interval label generated by the goods aging prediction module (1) and writing the label into the RFID chip of the goods or associating it with the goods database entry; An inventory mapping unit (23), for establishing a dynamic association database of goods position information including three-dimensional coordinates and the outbound time interval label, and its data model is: [Goods ID, rack coordinates (X, Y, Z), outbound interval label, storage environment parameters, last operation timestamp]; A status monitoring unit (24): through an Internet of Things sensor array deployed on the racks, real-time monitoring of the storage environment parameters of the goods; A data storage unit (25), for establishing a basic goods database storing the identity of goods and the inbound and outbound order information; Moreover, the basic goods database and the dynamic association database form a master-slave architecture, and a two-way real-time synchronization channel is established between the two databases through the goods ID.

9. The goods scheduling and distribution system based on the stacker storage vertical warehouse according to claim 1, characterized in that, When the load balancing module (5) executes resource allocation, the following strategy is adopted: When it is detected that the busy rate of any stacker in the physical storage area is greater than 85% or the backlog length of the conveyor belt exceeds the preset threshold L, a cross-region equipment support mechanism is triggered, and idle stackers with preset support amounts are retrieved in the order of the long-term storage area, the fast area, and the extreme speed area to jointly execute the distribution tasks in the triggered area.

10. A goods scheduling and distribution system based on a stacker storage vertical warehouse according to claim 9, characterized in that, The specific process of the idle stackers jointly executing the distribution tasks in the triggered area includes: after the cross-region equipment support mechanism is triggered, the cloud collaborative scheduling center (6) reallocates the distribution tasks in the triggered area based on the shortest job first algorithm, and uses the Dijkstra algorithm to plan the stacker paths based on the distribution tasks of each stacker and sends them to the goods scheduling execution module (4) to control the stacker to perform the inbound and outbound operations of the goods.

Images