Digital cloud warehouse supply chain space intelligent regulation method and system

Abstract

The application relates to the technical field of supply chain regulation, and particularly discloses a supply chain space intelligent regulation method and system of a digital cloud warehouse, which comprises the following steps: acquiring a supply point of a product and passenger flow information of the product, and creating a supply layer according to the supply point and the passenger flow information; acquiring order information of the product, inserting a demand point into the supply layer according to the order information, and obtaining a supply-demand feature map; generating a buffer node based on the supply-demand feature map of different time periods, and generating a supply chain regulation instruction pointing to the buffer node; the supply information and the demand information are subjected to two-dimensional processing, the supply-demand feature map is constructed, supply-demand analysis is carried out in the supply-demand feature map, the buffer point is determined, the buffer point is taken as a new temporary supply point, a supply chain update request is synchronously generated, a new supply chain control instruction is determined, and thus the regulation process is realized, so that the overall supply efficiency is greatly improved.

Description

Technical Field

[0001] This invention relates to the field of supply chain control technology, specifically a method and system for intelligent spatial control of the supply chain in a digital cloud warehouse. Background Technology

[0002] Digital cloud warehouses are flexible supply chain service models that rely on technologies such as cloud computing, big data, the Internet of Things, and artificial intelligence to integrate distributed physical warehousing resources and take a cloud-based intelligent management platform as the core to achieve full-chain digital collaboration in warehousing, orders, and logistics. Essentially, they transform warehousing and logistics capabilities into cloud services that can be called upon on demand; in layman's terms, they are intelligent supply chain management systems.

[0003] Existing supply chain architectures are mostly generated based on existing supply points, and the generation process is extremely sophisticated. However, demand information itself changes in real time, and at certain times existing supply points are unable to meet the corresponding demand. Therefore, how to optimize and adjust the supply chain architecture to match real-time changing demand information is the technical problem that this invention aims to solve. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for intelligent control of supply chain space in digital cloud warehouses, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for intelligent spatial control of the supply chain in a digital cloud warehouse, the method comprising: Obtain product supply locations and their customer flow information, and create a supply layer based on the supply locations and their customer flow information; the supply locations include online locations and offline locations; Obtain product order information, insert demand points into the supply layer based on the order information, and obtain a supply and demand feature map; wherein, the supply and demand feature map corresponds one-to-one with the order information in time; Buffer nodes are generated based on supply and demand characteristic maps of different time periods, and supply chain control instructions pointing to the buffer nodes are generated. Obtain environmental parameters and modify supply chain control instructions based on the environmental parameters.

[0006] As a further aspect of the present invention: the step of obtaining the product's supply locations and customer flow information, and creating a supply layer based on the supply locations and customer flow information, includes: Obtain the online supply point location and its supply vector for the product, and determine the online radiation layer based on the online supply point location and its supply vector; wherein, the direction of the supply vector represents the supply direction, and the magnitude of the supply vector represents the supply speed; Obtain the location of the product's offline supply points and their customer traffic, and determine the offline radiation layer based on the location of the offline supply points and their customer traffic; By overlaying the offline radiation layer and the online radiation layer, a supply layer is obtained.

[0007] As a further aspect of the present invention: the step of obtaining product order information and inserting demand points into the supply layer based on the order information to obtain a supply and demand feature map includes: Based on preset permissions, order information is obtained in real time, and time information, order location, and order quantity are extracted from the order information; the time information includes the order placement time and the receipt time. The time axis is binned according to a preset time step to obtain time intervals; For any time interval, the matching time information is included in the order information within that time interval. The order location is determined in the supply layer based on the order location, and the order feature value of the order location is determined based on the order quantity. The order feature value contains an adjustment coefficient, and the adjustment coefficient matched to the delivery time is greater than the adjustment coefficient matched to the order placement time. When the matching is complete, the demand layer corresponding to the time interval is obtained. The demand layer and the supply layer are treated as two independent channels and merged to obtain the supply and demand feature map.

[0008] As a further aspect of the present invention: the step of generating buffer nodes based on supply and demand characteristic maps of different time periods, and generating supply chain control instructions pointing to the buffer nodes, includes: Read all supply and demand feature graphs within a preset time period, sort them according to time order, and obtain a graph sequence; For any supply and demand feature map, cluster the locations in the demand layer of the supply and demand feature map to obtain demand regions labeled with time intervals; where the number of clusters is a preset value, and the number of clusters is the same for all supply and demand feature maps; Based on the supply layer of the supply and demand feature map, the satisfaction rate of each demand region is determined, and a satisfaction rate matrix is ​​constructed according to the location relationship of the demand regions. Based on the sequential statistical satisfaction rate matrix of the graph sequence, buffer nodes are determined based on the satisfaction rate matrix, and supply chain control instructions pointing to the buffer nodes are generated.

[0009] As a further aspect of the present invention: the steps of sequentially statistically analyzing the satisfaction rate matrix based on the graph sequence, determining buffer nodes based on the satisfaction rate matrix, and generating supply chain control instructions pointing to the buffer nodes include: Based on the sequential statistical satisfaction rate matrix of the graph sequence, a matrix sequence is obtained; Extract the satisfaction rate at each time point corresponding to each row and column position to obtain the satisfaction rate sequence; The satisfaction rate sequence is analyzed, and the row and column positions are marked; the analysis process includes mean analysis and deviation analysis. The buffer node is determined based on the row and column position of the marker, and supply chain control instructions pointing to the buffer node are generated.

[0010] As a further aspect of the present invention: the step of acquiring environmental parameters and modifying supply chain control instructions based on the environmental parameters includes: The process of determining the buffer nodes is constructed as a stochastic process; The random process is executed cyclically to obtain a preset number of buffer node schemes; Obtain environmental parameters and determine the supply speed of each buffer node in each buffer node scheme based on the environmental parameters; Update the online radiation layer based on each buffer node and its supply speed; The final buffer node scheme is determined based on the updated online radiation layer.

[0011] The present invention also provides a supply chain spatial intelligent control system for digital cloud warehouses, the system comprising: The supply layer determination module is used to obtain the product supply locations and their customer flow information, and create a supply layer based on the supply locations and their customer flow information; the supply locations include online locations and offline locations; The supply and demand characteristic analysis module is used to obtain product order information and insert demand points into the supply layer based on the order information to obtain a supply and demand characteristic map; wherein, the supply and demand characteristic map and the order information correspond one-to-one in time; The regulation instruction generation module is used to generate buffer nodes based on the supply and demand characteristic maps of different time periods, and generate supply chain regulation instructions pointing to the buffer nodes. The control command correction module is used to acquire environmental parameters and correct the supply chain control commands based on the environmental parameters.

[0012] As a further aspect of the present invention: the supply layer determination module includes: The online analysis unit is used to obtain the online supply point location and its supply vector of the product, and to determine the online radiation layer based on the online supply point location and its supply vector; wherein, the direction of the supply vector represents the supply direction, and the magnitude of the supply vector represents the supply speed; The offline analysis unit is used to obtain the location of the product's offline supply points and their customer traffic, and to determine the offline radiation layer based on the location of the offline supply points and their customer traffic. The layer overlay unit is used to overlay the offline radial layer and the online radial layer to obtain the supply layer.

[0013] As a further aspect of the present invention: the supply and demand characteristic analysis module includes: The information extraction unit is used to obtain order information in real time based on preset permissions, and extract time information, order location and order quantity from the order information; the time information includes the order placement time and the receipt time; The time axis binning unit is used to divide the time axis into bins according to a preset time step to obtain time intervals; The order analysis unit is used to match order information whose time information is contained within any given time interval, determine the order location in the supply layer based on the order location, and determine the order feature value of the order location based on the order quantity; wherein, the order feature value contains an adjustment coefficient, and the adjustment coefficient matched to the delivery time is greater than the adjustment coefficient matched to the order placement time; The merged execution unit is used to obtain the demand layer corresponding to the time interval when the matching is completed. The demand layer and the supply layer are treated as two independent channels and merged to obtain the supply and demand feature map.

[0014] As a further aspect of the present invention: the control command generation module includes: The graph sequence generation unit is used to read all supply and demand feature graphs within a preset time period, sort them according to time order, and obtain a graph sequence. The location clustering unit is used to cluster locations in the demand layer of any supply and demand feature map to obtain demand regions labeled with time intervals; the number of clusters is a preset value, and the number of clusters is the same for all supply and demand feature maps; The satisfaction rate determination unit is used to determine the satisfaction rate of each demand region based on the supply layer of the supply and demand feature map, and to construct a satisfaction rate matrix according to the location relationship of the demand regions. The satisfaction rate analysis unit is used to sequentially calculate the satisfaction rate matrix based on the graph sequence, determine the buffer node based on the satisfaction rate matrix, and generate supply chain control instructions pointing to the buffer node.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention performs two-dimensional processing on supply and demand information to construct a supply and demand feature map. Supply and demand analysis is performed on the feature map to determine buffer points. These buffer points are then used as new temporary supply points. Simultaneously, supply chain update requests are generated, and new supply chain control instructions are determined, thereby realizing the regulation process. Since this process involves creating new supply points, the actual cost may only be the selection of a warehouse as a temporary storage center, but it can greatly improve the overall supply efficiency. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention.

[0017] Figure 1 The overall flowchart of the intelligent supply chain spatial control method for digital cloud warehouses is shown.

[0018] Figure 2 The diagram shows the structure of the intelligent supply chain spatial control system for digital cloud warehouses. Detailed Implementation

[0019] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0020] Figure 1 This invention provides a flowchart of a method and system for intelligent supply chain spatial control in a digital cloud warehouse. In this embodiment, a method for intelligent supply chain spatial control in a digital cloud warehouse includes: Step S100: Obtain the product's supply locations and customer flow information, and create a supply layer based on the supply locations and customer flow information; wherein, the supply locations include online locations and offline locations; This invention is described from a product perspective. When multiple products exist (in fact, multiple products exist), this invention can be applied multiple times. For any product, the supply locations are obtained, and the customer flow information of the supply locations is queried. Based on the supply locations and their customer flow information, the supply situation can be determined. In the above content, the corresponding supply layer is the two-dimensional numerical form of the supply situation. It should be noted that the supply locations include online locations and offline locations. Online locations correspond to some online stores, and offline locations correspond to some offline stores.

[0021] Step S200: Obtain product order information, insert demand points into the supply layer according to the order information, and obtain a supply and demand feature map; wherein, the supply and demand feature map and the order information correspond one-to-one in time; Step S100 obtains the product supply information, while step S200 obtains the product demand information, acquires the product order information, analyzes the order information, and inserts the analysis results into the supply layer. The resulting comprehensive layer containing both demand and supply information is called the supply and demand feature map. At this point, all product information is represented by an image. In the context of current technology, images are a very common data structure, and there are numerous data processing solutions applicable to images. Therefore, they have strong potential for secondary development.

[0022] It is worth mentioning that the above content has default permissions, which are used to obtain information on product supply locations and order information. Obtaining permissions requires sending a permission request in advance. The request recipients include not only the entity corresponding to the supply location but also the user who placed the order. Since the order information obtained by the present invention does not involve identity information but only product and time information, the permission acquisition process is not difficult. From a technical point of view, the permission acquisition process is simply a data transmission and reception process. Therefore, the present invention has already obtained the relevant permissions by default, and will not be described in detail here.

[0023] In addition, passenger flow information and order information are information within a certain period of time, with obvious time characteristics; in other words, they have labels indicating time.

[0024] Step S300: Generate buffer nodes based on supply and demand characteristic maps of different time periods, and generate supply chain control instructions pointing to the buffer nodes; The supply and demand characteristic map for different time periods represents the supply and demand information for the current time period. Since the supply information is actually based on a map layer, it naturally contains location information. In summary, the supply and demand characteristic map represents the supply and demand situation at different locations in the current time period. By analyzing the supply and demand situation, some buffer nodes can be selected. The buffer nodes can be understood as a type of supply node. They can be some logistics transfer points or some temporary sales points. When order information is received later, the buffer nodes are used as new supply points to generate supply chain control instructions. The supply chain control instructions are actually the process of determining the delivery route. In the traditional solution, the path of delivering goods from the supply point to the demand point is the supply chain instruction. When a new "supply point" appears, the supply chain instruction will also be updated accordingly, hence the name supply chain control instructions.

[0025] Step S400: Obtain environmental parameters and modify the supply chain control instructions based on the environmental parameters; In practical applications, step S300 can be the final step, but the delivery process also needs to take environmental factors into account. The above content obtains environmental parameters and modifies the supply chain control instructions based on the environmental parameters. Its correction target is actually the buffer node. When the buffer node is modified, the supply chain control instructions based on the buffer node will also be updated accordingly.

[0026] Regarding step S100, the step of obtaining the product's supply locations and customer flow information, and creating a supply layer based on the supply locations and customer flow information, includes: Obtain the online supply point location and its supply vector for the product, and determine the online radiation layer based on the online supply point location and its supply vector; wherein, the direction of the supply vector represents the supply direction, and the magnitude of the supply vector represents the supply speed; Obtain the location of the product's offline supply points and their customer traffic, and determine the offline radiation layer based on the location of the offline supply points and their customer traffic; By overlaying the offline radiation layer and the online radiation layer, a supply layer is obtained.

[0027] In one example of the technical solution of this invention, the process of creating the supply layer is specifically described. It is divided into two types: online points and offline points. For online points, the online supply point location and its supply vector of the product are obtained. Based on the online supply point location and its supply vector, the online radiation layer is determined. Here, the online supply point location refers to the warehouse location, and the supply vector indicates the direction and speed of the shipment from the supply point. Since this process only involves the supply situation, the supply volume in each direction (which is actually the demand) is not considered.

[0028] For offline locations, the location of the product's offline supply points and their customer traffic are obtained. Based on the location of the offline supply points and their customer traffic, an offline radiation layer is determined. Location is easy to understand. As for customer traffic, it can be obtained from the access control terminals installed at the offline locations, recording how many users enter each day. In addition, the population of the area can also be used. Only parameters that can reflect the characteristics of traffic flow are needed. Of course, the process of obtaining customer traffic also requires prior authorization. If authorization is not available, on-site measurement in public places is also feasible.

[0029] The online radiation layer can be determined based on the location of online supply points and their supply vectors, while the offline radiation layer can be determined based on the location of offline supply points and their customer flow. The online radiation layer has a clear direction of change, while the offline radiation layer has no set direction and is set to an isotropic distribution by default.

[0030] Finally, the offline radiation layer and the online radiation layer are overlaid to obtain the supply layer. In the technical solution of this invention, the layer generation process actually involves determining the color values ​​at each location. For any location in the online radiation layer, the supply vectors of each online supply point are obtained. Simultaneously, the position vectors between the online supply points and the locations to be calculated are determined. The projection of the supply vector onto the position vector is calculated, and the ratio of the projection to the magnitude of the original supply vector is calculated. The resulting value represents the distribution ratio of the supply vector on the position vector. When calculating the radiation value, the components can be calculated first, for example, calculating the online supply points. The corresponding radiation value (related to scale, a pre-set value) is used to calculate the product of the radiation value and the allocation ratio, thus obtaining the simulated radiation value of the online supply point at that location. A radiation function inversely proportional to the distance (inverse proportional function of distance or negative exponential function of distance) is created from the simulated radiation value. Combined with the distance between this location and the online supply points, the radiation value of a certain supply vector of the online supply point at that location is obtained. The radiation values ​​of all supply vectors of all online supply points at that location are accumulated to obtain the final total radiation value at that location. The total radiation value is mapped to between 0 and 255 to obtain the color value.

[0031] Regarding the mapping and overlay process, mapping to the range of 0 to 255 is only for ease of display (after overlay, another mapping is required). In fact, it is also feasible to directly use the original values ​​as the final output. In this case, it is more appropriate to call the layer a two-dimensional matrix. The image is a two-dimensional matrix with values ​​within the preset range. The above two-dimensional matrix can also be regarded as an image with an unrestricted range.

[0032] For the offline radiation layer, the calculation process for each location is similar. The difference is that it does not have a supply vector parameter. Therefore, the radiation value at the offline supply point is directly read, a radiation function that is inversely proportional to the distance is created, and the radiation value is obtained by combining the distance and accumulating the values. Then, it is converted into color values.

[0033] Regarding step S200, the step of obtaining product order information and inserting demand points into the supply layer based on the order information to obtain a supply and demand feature map includes: Based on preset permissions, order information is obtained in real time, and time information, order location, and order quantity are extracted from the order information; the time information includes the order placement time and the receipt time. The time axis is binned according to a preset time step to obtain time intervals; For any time interval, the matching time information is included in the order information within that time interval. The order location is determined in the supply layer based on the order location, and the order feature value of the order location is determined based on the order quantity. The order feature value contains an adjustment coefficient, and the adjustment coefficient matched to the delivery time is greater than the adjustment coefficient matched to the order placement time. When the matching is complete, the demand layer corresponding to the time interval is obtained. The demand layer and the supply layer are treated as two independent channels and merged to obtain the supply and demand feature map.

[0034] In one example of the technical solution of this invention, the process of generating a supply and demand feature map is described. Order information is acquired in real time based on preset permissions. Time information, order location, and order quantity are extracted from the order information. The time information, order location, and order quantity do not involve identity information, and the order location only needs to be a general range, such as restricted to a street or town. The permission acquisition process is not difficult. The time information includes the order placement time and the delivery time. The time axis is binned according to a preset time step to obtain time intervals. The binning process divides the time axis into multiple time periods, each time period corresponding to a time interval. For any time interval, order information whose time information is contained within that time interval is matched. The order location is determined in the supply layer based on the order location, and the order feature value of the order location is determined based on the order quantity. The order feature value is directly proportional to the order quantity. The order feature value contains an adjustment coefficient. The adjustment coefficient matched to the delivery time is greater than the adjustment coefficient matched to the order placement time. The practical significance is that the order feature value represents the demand situation of the order. When the time interval includes the order placement time, it indicates that there is a certain order demand. At this time, the user may still return the goods. When the time interval includes the delivery time, it means that the user has clearly demanded the product. At this time, the existing order demand is more real. This is reflected in the order feature value parameter, which is that the order feature value corresponding to the delivery time is larger.

[0035] For each time interval, when the matching is complete, the corresponding demand layer is obtained. The demand layer and the supply layer are treated as two independent channels and merged to obtain the supply and demand feature map.

[0036] Regarding step S300, the step of generating buffer nodes based on supply and demand characteristic maps of different time periods and generating supply chain control instructions pointing to the buffer nodes includes: Read all supply and demand feature graphs within a preset time period, sort them according to time order, and obtain a graph sequence; For any supply and demand feature map, cluster the locations in the demand layer of the supply and demand feature map to obtain demand regions labeled with time intervals; where the number of clusters is a preset value, and the number of clusters is the same for all supply and demand feature maps; Based on the supply layer of the supply and demand feature map, the satisfaction rate of each demand region is determined, and a satisfaction rate matrix is ​​constructed according to the location relationship of the demand regions. Based on the sequential statistical satisfaction rate matrix of the graph sequence, buffer nodes are determined based on the satisfaction rate matrix, and supply chain control instructions pointing to the buffer nodes are generated.

[0037] In one example of the technical solution of this invention, the process of generating supply chain control instructions is described. The core of the supply chain control instruction generation process is the process of determining the buffer node. After the buffer node is determined, the existing supply chain generation process can be applied. It is compared with the original supply chain. The process of converting the original supply chain instruction into the new supply chain instruction is the supply chain control process, corresponding to the supply chain control instruction.

[0038] Read all supply and demand feature maps within a preset time period, sort them according to time order to obtain a sequence of supply and demand feature maps, called a graph sequence. Then, analyze each supply and demand feature map in the graph sequence in turn. The analysis process is as follows: for any supply and demand feature map, cluster the positions in the demand layer of the supply and demand feature map to obtain demand regions labeled with time intervals. The clustering process is essentially to cluster the pixels in the image, which can be completed by applying existing region segmentation algorithms to obtain demand regions labeled with time intervals. Specifically, the region segmentation algorithm can use the K-means clustering algorithm, but this requires setting the K value in advance, that is, the total number of regions obtained by clustering. Once the K value is determined, the same clustering operation needs to be performed on all supply and demand feature maps.

[0039] Based on this, the satisfaction rate of each demand region is determined by the supply layer of the supply and demand feature map. A satisfaction rate matrix is ​​constructed according to the positional relationship of the demand regions. Since the number of demand regions is the same, even if the demand regions change slightly, the row and column positions corresponding to the demand regions in the satisfaction rate matrix will not change. This actually eliminates the process of judging whether demand regions correspond. This is one of the unique features of the technical solution of this invention. That is, the demand regions corresponding to the same row and column positions in the supply and demand feature map may have differences within a certain range, but their corresponding row and column positions are the same. In other words, demand will definitely change over time. As long as the change is small or not too large, the demand regions always correspond. In the prior art, when analyzing graph sequences, extremely complex region shape comparison algorithms are often used. However, in the technical solution of this invention, only the number of clusters needs to be limited to eliminate the complex region shape comparison process.

[0040] Finally, based on the sequential statistical satisfaction rate matrix of the graph sequence, buffer nodes are determined based on the satisfaction rate matrix, and supply chain control instructions pointing to the buffer nodes are generated. The steps of sequential statistical statistical satisfaction rate matrix based on the graph sequence, determining buffer nodes based on the satisfaction rate matrix, and generating supply chain control instructions pointing to the buffer nodes include: Based on the sequential statistical satisfaction rate matrix of the graph sequence, a matrix sequence is obtained; Extract the satisfaction rate at each time point corresponding to each row and column position to obtain the satisfaction rate sequence; The satisfaction rate sequence is analyzed, and the row and column positions are marked; the analysis process includes mean analysis and deviation analysis. The buffer node is determined based on the row and column position of the marker, and supply chain control instructions pointing to the buffer node are generated.

[0041] The above content explains the analysis process of the satisfaction rate matrix. Based on the sequential statistical analysis of the graph sequence, a matrix sequence is obtained. All matrices have the same format. For any row and column position, the satisfaction rate at each time corresponding to each row and column position is extracted to obtain the satisfaction rate sequence. At this time, the satisfaction rate sequence is a one-dimensional array. By analyzing the satisfaction rate sequence, it can be determined whether there are large fluctuations at that row and column position. The row and column positions are marked according to the judgment results. The analysis process includes mean analysis and deviation analysis. One feasible approach is to calculate the mean and standard deviation of the satisfaction rate sequence. If the mean is less than the preset mean threshold and the standard deviation is less than the preset standard deviation threshold, it indicates that the satisfaction rate is consistently low (the standard deviation is also small, not fluctuating). In this case, it means that demand is greater than supply, and the row and column positions are marked.

[0042] Based on the above architecture, the marked row and column positions are actually the row and column positions where demand is greater than supply. By querying the demand regions corresponding to the row and column positions in different supply and demand feature maps, points can be selected as buffer points within the demand regions.

[0043] It's worth noting that the process of selecting locations as buffer points within a demand area can be random, or existing location selection methods based on ease of access can be applied to select a suitable buffer point. One feasible approach is to query the demand areas corresponding to row and column positions in different supply and demand characteristic maps. The queried demand areas are constantly changing (demand itself changes over time). The intersection of the demand areas is calculated, the intersection center is selected, and the warehouse point closest to the intersection center is chosen as the buffer node. However, the process of calculating the intersection of demand areas still has some issues. For example, if the intersection of all demand areas is empty, the number of demand areas participating in the intersection calculation should be reduced. Assuming there are N demand areas and the intersection of N demand areas is empty, N-1 demand areas should be selected from the N demand areas for intersection calculation. If it is still empty, N-2 demand areas should be selected, and so on, until an intersection occurs.

[0044] Regarding step S400, the step of acquiring environmental parameters and modifying the supply chain control instructions based on the environmental parameters includes: The process of determining the buffer nodes is constructed as a stochastic process; The random process is executed cyclically to obtain a preset number of buffer node schemes; Obtain environmental parameters and determine the supply speed of each buffer node in each buffer node scheme based on the environmental parameters; Update the online radiation layer based on each buffer node and its supply speed; The final buffer node scheme is determined based on the updated online radiation layer.

[0045] In one example of the technical solution of this invention, the correction process is described. The above provides a conventional buffer node selection scheme, which is essentially a site selection process within a region. A simple approach is to select the feasible location closest to the region's center. This is a very deterministic and restrictive scheme, making it difficult to find a superior solution. In the above, the buffer node determination process is constructed as a random process. That is, candidate locations (locations that can serve as buffer nodes, such as vacant warehouses) are first determined within the region. Then, a location is randomly selected from these candidate locations as the buffer node. The probability of selecting each location is the same. This is equivalent to constructing the buffer node determination process as a random process. The number of buffer nodes to be set is not unique; one is selected from each region. The process involves establishing buffer points and then repeatedly executing a random process to obtain a preset number of buffer node schemes. The preset number is essentially the preset number of iterations. Environmental parameters are acquired, and the supply speed of each buffer node in each buffer node scheme is determined based on these parameters. This process is somewhat similar to the process of constructing a supply layer based on supply points, since buffer nodes themselves are equivalent to temporary supply points. However, the difference is that the supply vector is no longer limited; instead, an isotropic expansion process is used (this step is somewhat similar to offline points). Environmental parameters are acquired, and the supply speed of each buffer node in each buffer node scheme is determined based on these parameters. The online radiation layer is updated based on each buffer node and its supply speed, and the final buffer node scheme is determined based on the updated online radiation layer.

[0046] The process of determining the final buffer node scheme based on the updated online radiation layer is an optimization process. This optimization process requires a parameter, such as an evaluation score. By comparing the evaluation scores, the optimal scheme can be selected. In the technical solution of this invention, the process of determining the evaluation score is as follows: Update the supply layer based on the updated online radiation layer, update the supply and demand feature map based on the updated supply layer, determine the satisfaction rate matrix based on the updated supply and demand feature map, select the minimum value in the satisfaction rate matrix, and determine the evaluation score based on the direct proportion of the minimum value (the larger the minimum value, the larger the evaluation score). At this time, even the least satisfied demand area in the selected optimal solution has a high satisfaction rate.

[0047] Figure 2 A structural diagram of a digital cloud warehouse supply chain spatial intelligent control system is shown. The system 10 comprises: The supply layer determination module 11 is used to obtain the supply locations of products and their customer flow information, and to create a supply layer based on the supply locations and their customer flow information; wherein, the supply locations include online locations and offline locations; The supply and demand feature analysis module 12 is used to obtain product order information and insert demand points into the supply layer based on the order information to obtain a supply and demand feature map; wherein, the supply and demand feature map and the order information correspond one-to-one in time; The regulation instruction generation module 13 is used to generate buffer nodes based on the supply and demand characteristic maps of different time periods, and generate supply chain regulation instructions pointing to the buffer nodes. The control command correction module 14 is used to acquire environmental parameters and correct the supply chain control commands based on the environmental parameters.

[0048] Furthermore, the supply layer determination module 11 includes: The online analysis unit is used to obtain the online supply point location and its supply vector of the product, and to determine the online radiation layer based on the online supply point location and its supply vector; wherein, the direction of the supply vector represents the supply direction, and the magnitude of the supply vector represents the supply speed; The offline analysis unit is used to obtain the location of the product's offline supply points and their customer traffic, and to determine the offline radiation layer based on the location of the offline supply points and their customer traffic. The layer overlay unit is used to overlay the offline radial layer and the online radial layer to obtain the supply layer.

[0049] Specifically, the supply and demand characteristic analysis module 12 includes: The information extraction unit is used to obtain order information in real time based on preset permissions, and extract time information, order location and order quantity from the order information; the time information includes the order placement time and the receipt time; The time axis binning unit is used to divide the time axis into bins according to a preset time step to obtain time intervals; The order analysis unit is used to match order information whose time information is contained within any given time interval, determine the order location in the supply layer based on the order location, and determine the order feature value of the order location based on the order quantity; wherein, the order feature value contains an adjustment coefficient, and the adjustment coefficient matched to the delivery time is greater than the adjustment coefficient matched to the order placement time; The merged execution unit is used to obtain the demand layer corresponding to the time interval when the matching is completed. The demand layer and the supply layer are treated as two independent channels and merged to obtain the supply and demand feature map.

[0050] Furthermore, the control command generation module 13 includes: The graph sequence generation unit is used to read all supply and demand feature graphs within a preset time period, sort them according to time order, and obtain a graph sequence. The location clustering unit is used to cluster locations in the demand layer of any supply and demand feature map to obtain demand regions labeled with time intervals; the number of clusters is a preset value, and the number of clusters is the same for all supply and demand feature maps; The satisfaction rate determination unit is used to determine the satisfaction rate of each demand region based on the supply layer of the supply and demand feature map, and to construct a satisfaction rate matrix according to the location relationship of the demand regions. The satisfaction rate analysis unit is used to sequentially calculate the satisfaction rate matrix based on the graph sequence, determine the buffer node based on the satisfaction rate matrix, and generate supply chain control instructions pointing to the buffer node.

[0051] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for intelligent spatial control of the supply chain in a digital cloud warehouse, characterized in that, The method includes: Obtain product supply locations and their customer flow information, and create a supply layer based on the supply locations and their customer flow information; the supply locations include online locations and offline locations; Obtain product order information, insert demand points into the supply layer based on the order information, and obtain a supply and demand feature map; wherein, the supply and demand feature map corresponds one-to-one with the order information in time; Buffer nodes are generated based on supply and demand characteristic maps of different time periods, and supply chain control instructions pointing to the buffer nodes are generated. Obtain environmental parameters and modify supply chain control instructions based on the environmental parameters.

2. The intelligent supply chain spatial control method for digital cloud warehouses according to claim 1, characterized in that, The steps of obtaining product supply locations and customer flow information, and creating a supply layer based on the supply locations and customer flow information, include: Obtain the online supply point location and its supply vector for the product, and determine the online radiation layer based on the online supply point location and its supply vector; wherein, the direction of the supply vector represents the supply direction, and the magnitude of the supply vector represents the supply speed; Obtain the location of the product's offline supply points and their customer traffic, and determine the offline radiation layer based on the location of the offline supply points and their customer traffic; By overlaying the offline radiation layer and the online radiation layer, a supply layer is obtained.

3. The intelligent supply chain spatial control method for digital cloud warehouses according to claim 1, characterized in that, The steps of obtaining product order information and inserting demand points into the supply layer based on the order information to obtain a supply and demand feature map include: Based on preset permissions, order information is obtained in real time, and time information, order location, and order quantity are extracted from the order information; the time information includes the order placement time and the receipt time. The time axis is binned according to a preset time step to obtain time intervals; For any time interval, the matching time information is included in the order information within that time interval. The order location is determined in the supply layer based on the order location, and the order feature value of the order location is determined based on the order quantity. The order feature value contains an adjustment coefficient, and the adjustment coefficient matched to the delivery time is greater than the adjustment coefficient matched to the order placement time. When the matching is complete, the demand layer corresponding to the time interval is obtained. The demand layer and the supply layer are treated as two independent channels and merged to obtain the supply and demand feature map.

4. The intelligent supply chain spatial control method for digital cloud warehouses according to claim 1, characterized in that, The step of generating buffer nodes based on supply and demand characteristic maps of different time periods, and generating supply chain control instructions pointing to the buffer nodes, includes: Read all supply and demand feature graphs within a preset time period, sort them according to time order, and obtain a graph sequence; For any supply and demand feature map, cluster the locations in the demand layer of the supply and demand feature map to obtain demand regions labeled with time intervals; where the number of clusters is a preset value, and the number of clusters is the same for all supply and demand feature maps; Based on the supply layer of the supply and demand feature map, the satisfaction rate of each demand region is determined, and a satisfaction rate matrix is ​​constructed according to the location relationship of the demand regions. Based on the sequential statistical satisfaction rate matrix of the graph sequence, buffer nodes are determined based on the satisfaction rate matrix, and supply chain control instructions pointing to the buffer nodes are generated.

5. The intelligent supply chain spatial control method for digital cloud warehouses according to claim 4, characterized in that, The steps of generating supply chain control instructions pointing to the buffer nodes based on the sequential statistical satisfaction rate matrix of the graph sequence include: Based on the sequential statistical satisfaction rate matrix of the graph sequence, a matrix sequence is obtained; Extract the satisfaction rate at each time point corresponding to each row and column position to obtain the satisfaction rate sequence; The satisfaction rate sequence is analyzed, and the row and column positions are marked; the analysis process includes mean analysis and deviation analysis. The buffer node is determined based on the row and column position of the marker, and supply chain control instructions pointing to the buffer node are generated.

6. The intelligent supply chain spatial control method for digital cloud warehouses according to claim 1, characterized in that, The step of acquiring environmental parameters and modifying supply chain control instructions based on the environmental parameters includes: The process of determining the buffer nodes is constructed as a stochastic process; The random process is executed cyclically to obtain a preset number of buffer node schemes; Obtain environmental parameters and determine the supply speed of each buffer node in each buffer node scheme based on the environmental parameters; Update the online radiation layer based on each buffer node and its supply speed; The final buffer node scheme is determined based on the updated online radiation layer.

7. A supply chain spatial intelligent control system for a digital cloud warehouse, characterized in that, The system includes: The supply layer determination module is used to obtain the product supply locations and their customer flow information, and create a supply layer based on the supply locations and their customer flow information; the supply locations include online locations and offline locations; The supply and demand characteristic analysis module is used to obtain product order information and insert demand points into the supply layer based on the order information to obtain a supply and demand characteristic map; wherein, the supply and demand characteristic map and the order information correspond one-to-one in time; The regulation instruction generation module is used to generate buffer nodes based on the supply and demand characteristic maps of different time periods, and generate supply chain regulation instructions pointing to the buffer nodes. The control command correction module is used to acquire environmental parameters and correct the supply chain control commands based on the environmental parameters.

8. The intelligent supply chain spatial control system for digital cloud warehouses according to claim 7, characterized in that, The supply layer determination module includes: The online analysis unit is used to obtain the online supply point location and its supply vector of the product, and to determine the online radiation layer based on the online supply point location and its supply vector; wherein, the direction of the supply vector represents the supply direction, and the magnitude of the supply vector represents the supply speed; The offline analysis unit is used to obtain the location of the product's offline supply points and their customer traffic, and to determine the offline radiation layer based on the location of the offline supply points and their customer traffic. The layer overlay unit is used to overlay the offline radial layer and the online radial layer to obtain the supply layer.

9. The intelligent supply chain spatial control system for digital cloud warehouses according to claim 7, characterized in that, The supply and demand characteristic analysis module includes: The information extraction unit is used to obtain order information in real time based on preset permissions, and extract time information, order location and order quantity from the order information; the time information includes the order placement time and the receipt time; The time axis binning unit is used to divide the time axis into bins according to a preset time step to obtain time intervals; The order analysis unit is used to match order information whose time information is contained within any given time interval, determine the order location in the supply layer based on the order location, and determine the order feature value of the order location based on the order quantity; wherein, the order feature value contains an adjustment coefficient, and the adjustment coefficient matched to the delivery time is greater than the adjustment coefficient matched to the order placement time; The merged execution unit is used to obtain the demand layer corresponding to the time interval when the matching is completed. The demand layer and the supply layer are treated as two independent channels and merged to obtain the supply and demand feature map.

10. The intelligent supply chain spatial control system for digital cloud warehouses according to claim 7, characterized in that, The control command generation module includes: The graph sequence generation unit is used to read all supply and demand feature graphs within a preset time period, sort them according to time order, and obtain a graph sequence. The location clustering unit is used to cluster locations in the demand layer of any supply and demand feature map to obtain demand regions labeled with time intervals; the number of clusters is a preset value, and the number of clusters is the same for all supply and demand feature maps; The satisfaction rate determination unit is used to determine the satisfaction rate of each demand region based on the supply layer of the supply and demand feature map, and to construct a satisfaction rate matrix according to the location relationship of the demand regions. The satisfaction rate analysis unit is used to sequentially calculate the satisfaction rate matrix based on the graph sequence, determine the buffer node based on the satisfaction rate matrix, and generate supply chain control instructions pointing to the buffer node.

Images