A big data driven intelligent supply chain management method and system

Abstract

This invention discloses a big data-driven intelligent supply chain management method and system, relating to the field of transaction order data processing technology, including the following steps: acquiring transaction order data and transaction address data; obtaining transaction demand starting point data based on the transaction order data and transaction address data; generating fulfillment migration chain data based on the transaction demand starting point data; this invention obtains transaction demand starting point data through transaction order data and transaction address data, and associates the transaction order, initial service area, and first target fulfillment node, which can retain the basis for transaction demand attribution during the transaction order generation stage, and avoid the transaction demand attribution node being covered by the actual transaction fulfillment node after continuous warehousing changes of the transaction order; generating fulfillment migration chain data through the transaction demand starting point data, and recording the fulfillment node before warehousing change, the fulfillment node after warehousing change, the warehousing change trigger time, and the actual transaction fulfillment node, can completely retain the continuous warehousing change process of the transaction order.

Description

Technical Field

[0001] This invention relates to the field of transaction order data processing technology, and in particular to a big data-driven intelligent supply chain management method and system. Background Technology

[0002] In the existing omnichannel order processing, e-commerce platforms, store-to-home platforms, and instant retail platforms typically assign corresponding fulfillment nodes to orders based on transaction address data, inventory status, service area, and delivery timeliness. Upon order completion, the sales volume corresponding to that order is recorded as part of the actual fulfillment node. This method is more effective when orders are completed by fixed fulfillment nodes, directly reflecting the sales volume within the service area corresponding to fulfillment nodes such as stores, forward warehouses, and regional warehouses.

[0003] However, in actual transaction fulfillment, a transaction order is not necessarily completed by the initially assigned fulfillment node. Especially in scenarios such as store-to-home delivery, pre-positioned warehouse delivery, same-city instant retail, and regional warehouse collaborative fulfillment, transaction orders may be reassigned due to stock shortages, inventory freezes, insufficient order processing capacity, or failure to meet delivery time requirements at the initial target fulfillment node. For example, if a user is located within the transaction service area of ​​store A, the system originally assigned the transaction order to store A for fulfillment. However, due to stock shortages at store A's fulfillment node, the order was reassigned to pre-positioned warehouse B for fulfillment. After pre-positioned warehouse B also experienced stock shortages, the order was reassigned to regional warehouse C for fulfillment, and ultimately, the transaction was delivered by regional warehouse C.

[0004] In the aforementioned scenario of continuous warehouse relocation, the actual transaction fulfillment node is not equivalent to the transaction demand attribution node. Although warehouse C completes the transaction delivery, the transaction demand attribution node for this order may still fall within the transaction service scope of store A. If the system continues to calculate sales transaction volume based on the actual transaction fulfillment node, it will misjudge the temporary transaction demand received by warehouse C as an increase in its local sales demand, thus causing the transaction supply allocation result to continuously tilt towards warehouse C. At the same time, store A, due to stockouts at the fulfillment node, may not have an actual transaction completion record, but may instead be misjudged by the system as having a decrease in local sales demand, resulting in a lower subsequent transaction supply allocation result for store A.

[0005] Therefore, existing technologies have at least the following shortcomings: Existing transaction order processing systems typically only record the actual transaction fulfillment node of an order, without fully recording the initial service area, the first target fulfillment node, and the continuous warehousing relocation process; existing sales demand statistics methods typically determine demand attribution based on the actual transaction fulfillment node, without determining whether the sales volume belongs to temporary transaction demand; existing systems have difficulty distinguishing between the transaction demand attribution node and the actual transaction fulfillment node, causing transaction demand to be masked by the continuous warehousing relocation process; existing systems do not include the reasons for continuous warehousing relocation in the correction of transaction demand attribution, which easily leads to the misjudgment of passive warehousing relocation caused by stockouts, inventory freezes, insufficient order processing capacity, or failure to meet delivery timeliness as changes in the transaction demand attribution node, thereby causing the transaction supply allocation result to deviate and affecting the accuracy and fulfillment efficiency of omnichannel transaction order processing. Summary of the Invention

[0006] The purpose of this invention is to solve the problem that after continuous warehousing changes of omnichannel transaction orders, the sales transaction volume at the actual transaction fulfillment node is mistakenly regarded as the local sales demand, resulting in distorted transaction demand attribution and affecting the transaction supply allocation result. Therefore, a big data-driven intelligent supply chain management method and system is proposed.

[0007] To achieve the above objectives, the present invention employs the following technology: a big data-driven intelligent supply chain management method and system, comprising the following steps:

[0008] Obtain transaction order data and transaction address data, and based on the transaction order data and transaction address data, obtain the starting point data of transaction demand;

[0009] Based on the transaction demand starting point data, generate fulfillment migration chain data; based on the fulfillment migration chain data, obtain position modification reason data; based on the position modification reason data, determine whether the position modification in the fulfillment migration chain data belongs to passive position modification. If it does, generate a passive position modification identifier in the fulfillment migration chain data; based on the passive position modification identifier and the transaction demand starting point data, obtain demand attribution data.

[0010] Based on the fulfillment migration chain data, the actual transaction fulfillment node data is determined; based on the demand attribution data, the actual transaction fulfillment node data, and the fulfillment migration chain data, the temporary transaction acceptance demand in the actual transaction fulfillment node is identified, and the temporary transaction acceptance demand data is obtained; based on the demand attribution data and the temporary transaction acceptance demand data, the sales transaction volume of each fulfillment node is corrected, and the corrected sales demand volume is obtained.

[0011] Based on the revised sales demand, a transaction supply configuration result is generated.

[0012] Furthermore, based on demand attribution data, actual transaction fulfillment node data, and fulfillment migration chain data, methods for identifying temporary transaction fulfillment demands within actual transaction fulfillment nodes and obtaining temporary transaction fulfillment demand data include:

[0013] Based on the demand attribution data, determine the transaction demand attribution node corresponding to the transaction order;

[0014] Based on actual transaction fulfillment node data, determine the actual transaction fulfillment node at which the transaction order is finally delivered;

[0015] Compare the node to which the transaction demand is assigned with the actual transaction fulfillment node;

[0016] If the node to which the transaction demand belongs is inconsistent with the actual transaction fulfillment node, the fulfillment migration path of the transaction order from the node to which the transaction demand belongs to the actual transaction fulfillment node is determined based on the fulfillment migration chain data.

[0017] Based on the passive position modification identifier in the fulfillment migration path, determine whether the sales transaction volume completed at the actual transaction fulfillment node was formed by passive position modification;

[0018] If the sales volume completed at the actual transaction fulfillment node is formed by passive warehouse repositioning, then the temporary sales volume completed at the actual transaction fulfillment node will be identified as temporary transaction demand, and the transaction order, the transaction demand attribution node, the actual transaction fulfillment node, the fulfillment migration path, and the temporary sales volume will be associated to generate temporary transaction demand data.

[0019] Furthermore, based on demand attribution data and temporary transaction demand data, the sales transaction volume of each fulfillment node is adjusted to obtain the adjusted sales demand volume. Methods include:

[0020] Based on the temporary acceptance of transaction demand data, determine the temporary acceptance of sales transaction volume formed by temporary acceptance of transaction demand at the actual transaction fulfillment node;

[0021] Based on the demand attribution data, determine the transaction demand attribution node corresponding to the temporarily accepted sales transaction volume.

[0022] Based on transaction order data and actual transaction fulfillment node data, determine the actual transaction fulfillment node at which the transaction order is finally completed and delivered, and determine the node sales transaction volume corresponding to the actual transaction fulfillment node;

[0023] Based on the temporary sales transaction volume, the local sales demand of the actual transaction fulfillment node is obtained by deducting the temporary sales transaction volume from the node sales transaction volume corresponding to the actual transaction fulfillment node.

[0024] Based on the transaction demand attribution node and the temporary sales transaction volume, the temporary sales transaction volume is returned to the sales transaction volume of the transaction demand attribution node to obtain the replenished sales demand volume of the transaction demand attribution node.

[0025] Based on local sales demand and replenished sales demand, the revised sales demand for each fulfillment node is generated.

[0026] Furthermore, methods for obtaining the starting point data of transaction demand based on transaction order data and transaction address data include:

[0027] Based on the transaction address data, the transaction address corresponding to the transaction address data is matched with the transaction service range pre-configured by each fulfillment node to determine the initial service area corresponding to the transaction order;

[0028] Based on the initial service area, candidate fulfillment nodes capable of responding to the transaction order are identified;

[0029] Based on transaction order data and candidate fulfillment nodes, determine the first target fulfillment node corresponding to the transaction order when the transaction is generated;

[0030] By associating transaction orders, initial service areas, and initial target fulfillment nodes, we can obtain the starting point data for transaction demand.

[0031] Furthermore, methods for generating fulfillment migration chain data based on transaction demand starting point data include:

[0032] Based on the initial data of transaction demand, determine the first target fulfillment node corresponding to the transaction order;

[0033] When a transaction order is transferred from the initial target fulfillment node to other fulfillment nodes for processing, the fulfillment node before the position change, the fulfillment node after the position change, and the position change trigger time are recorded.

[0034] When a transaction order is transferred from the current fulfillment node to another fulfillment node for processing, the corresponding fulfillment node before the position change, the fulfillment node after the position change, and the position change trigger time are continued to be recorded.

[0035] After a transaction order is completed and delivered, the actual transaction fulfillment node is determined based on the final fulfillment node of the transaction.

[0036] Based on the order of the position change trigger time, the initial target fulfillment node, each fulfillment node before the position change, each fulfillment node after the position change, and the actual transaction fulfillment node are associated to obtain the fulfillment migration chain data.

[0037] Furthermore, methods for obtaining warehouse relocation reason data based on fulfillment migration chain data include:

[0038] Based on the fulfillment migration chain data, extract the fulfillment node before the position change, the fulfillment node after the position change, and the position change trigger time for each position change in a transaction order;

[0039] Based on the warehouse change trigger time, collect the inventory status, order processing capacity status and delivery timeliness status of the fulfillment node before the warehouse change;

[0040] If the inventory status indicates that the fulfillment node could not provide the goods corresponding to the transaction order before the warehouse change, then the reason for the warehouse change will be determined as the fulfillment node being out of stock.

[0041] If the inventory status indicates that the goods in the fulfillment node before the warehouse change are in a state that cannot be used to complete the transaction delivery, then the reason for the warehouse change is determined to be inventory freeze.

[0042] If the order processing capacity status indicates that the fulfillment node before the repositioning cannot complete the order processing within the processing time required by the transaction order, then the reason for the repositioning is determined to be insufficient order processing capacity.

[0043] If the delivery timeliness status indicates that the fulfillment node before the warehouse relocation cannot complete the transaction delivery within the delivery time required by the transaction order, then the reason for the warehouse relocation is determined as delivery timeliness not being met.

[0044] By associating the reason for each position change with the fulfillment migration chain data, we can obtain the position change reason data.

[0045] Furthermore, based on the reason data for the change of warehouse position, methods for determining whether the change of warehouse position in the fulfillment migration chain data belongs to passive change of warehouse position include:

[0046] Based on the fulfillment migration chain data, determine each warehouse modification corresponding to a transaction order;

[0047] Based on the data on reasons for position changes, determine the reasons for each position change.

[0048] If the reason for changing the warehouse is out of stock at the fulfillment node, frozen inventory, insufficient order processing capacity, or failure to meet delivery time, then the corresponding warehouse change is determined to be a passive warehouse change.

[0049] Furthermore, the methods for generating passive warehouse relocation identifiers in the fulfillment migration chain data include:

[0050] Based on the fulfillment migration chain data, identify the warehouses that were judged to be passively modified.

[0051] Extract the transaction order, fulfillment node before the position change, fulfillment node after the position change, position change trigger time, and position change reason corresponding to the passive position change;

[0052] Associate the transaction order, the fulfillment node before the position change, the fulfillment node after the position change, the position change trigger time, and the position change reason to generate a passive position change identifier corresponding to the passive position change;

[0053] Write the passive warehouse modification identifier into the corresponding passive warehouse modification data in the fulfillment migration chain data.

[0054] Furthermore, methods for obtaining demand attribution data based on passive position modification identifiers and transaction demand origination data include:

[0055] Based on the passive position modification identifier, identify the existence of transaction orders that have been passively modified;

[0056] Based on the initial data of transaction demand, determine the initial service area and the first target fulfillment node corresponding to the transaction order;

[0057] If a transaction order involves passive position modification, the first target fulfillment node will be determined as the node to which the transaction demand belongs for the corresponding transaction order.

[0058] By associating transaction orders, initial service areas, first target fulfillment nodes, and transaction demand attribution nodes, demand attribution data can be obtained.

[0059] In summary, due to the adoption of the above-mentioned technologies in a big data-driven intelligent supply chain management method and system, the beneficial effects of this invention are:

[0060] This invention obtains the starting point data of transaction demand through transaction order data and transaction address data, and associates it with the transaction order, initial service area, and first target fulfillment node. It can retain the basis for the attribution of transaction demand during the transaction order generation stage, and avoid the transaction demand attribution node being covered by the actual transaction fulfillment node after continuous position changes of the transaction order. It generates fulfillment migration chain data through the starting point data of transaction demand, and records the fulfillment node before position change, the fulfillment node after position change, the position change trigger time, and the actual transaction fulfillment node. It can completely preserve the continuous position change process of the transaction order, and solve the problem that the existing technology only records the actual transaction fulfillment node, which makes it difficult to restore the fulfillment migration process of the transaction order.

[0061] This invention obtains warehouse relocation reason data through fulfillment migration chain data, and generates a passive warehouse relocation identifier based on the warehouse relocation reason data. It can identify passive warehouse relocation caused by stockouts, frozen inventory, insufficient order processing capacity, or failure to meet delivery timeliness at fulfillment nodes, avoiding misjudging passive warehouse relocation as a change in the transaction demand attribution node. It obtains demand attribution data through passive warehouse relocation identifier and transaction demand starting point data, and obtains temporary transaction demand data based on demand attribution data, actual transaction fulfillment node data, and fulfillment migration chain data. It can identify the temporary sales transaction volume in the actual transaction fulfillment node, avoiding misjudging the sales transaction volume as the local sales demand volume of the actual transaction fulfillment node.

[0062] This invention corrects the sales transaction volume of each fulfillment node by using demand attribution data and temporary transaction demand data. The temporary sales transaction volume is deducted from the actual transaction fulfillment node's corresponding node's sales transaction volume and returned to the sales transaction volume of the transaction demand attribution node, resulting in a more accurate corrected sales demand volume. Based on the corrected sales demand volume, the transaction supply configuration result is generated, which reduces the problem of erroneous bias in the transaction supply configuration result towards the actual transaction fulfillment node, reduces the risk of the transaction supply configuration result being too low for the transaction demand attribution node and too high for the actual transaction fulfillment node, and improves the accuracy and fulfillment efficiency of omnichannel transaction order processing. Attached Figure Description

[0063] Figure 1 A flowchart of the present invention is shown;

[0064] Figure 2 A system block diagram of the present invention is shown. Detailed Implementation

[0065] The following will describe, with reference to the accompanying drawings of the embodiments of the present invention, a big data-driven intelligent supply chain management method and system. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0066] To more clearly and intuitively demonstrate the practical application effects and advantages of the big data-driven intelligent supply chain management method and system of this invention, and to verify its feasibility and effectiveness, the invention is further described below with reference to embodiments. Through specific scenario simulations and data calculations, the method is explained in detail how it plays a role in actual supply chain management, helping readers better understand the technical details and practical value of the invention. The invention is further described below with reference to embodiments;

[0067] Example 1:

[0068] See Figure 1 A big data-driven intelligent supply chain management method includes the following steps:

[0069] The process involves acquiring transaction order data and transaction address data, and then using this data to derive the starting point data for transaction demands. This process preserves the correspondence between transaction orders and their corresponding transaction addresses during the order generation stage, ensuring that the attribution of subsequent transaction demands no longer solely depends on which fulfillment node ultimately completes the transaction order. This solves the problem in existing technologies where sales volume is calculated solely based on actual fulfillment nodes, leading to the transaction demand attribution node being overwritten by the final fulfillment result. It provides foundational data for subsequently determining the initial service area, the first target fulfillment node, and the transaction demand attribution node.

[0070] It should be noted that the methods for obtaining the starting point data of transaction demand based on transaction order data and transaction address data include:

[0071] Based on transaction address data, the transaction address corresponding to the transaction address data of the transaction order is matched with the pre-configured transaction service range of each fulfillment node to determine the initial service area corresponding to the transaction order. By matching the transaction address data with the pre-configured transaction service range of each fulfillment node, the initial service area corresponding to the transaction order at the time of its generation can be determined, thus limiting the node to which the transaction demand belongs to to the first part to the actual transaction service range corresponding to the transaction address. This solves the problem in the prior art where, after a transaction order has undergone continuous warehouse modifications, only the fulfillment node that finally completed the transaction delivery can be seen, and it is impossible to determine which service area the transaction order originally belonged to. This avoids mistaking the temporary acceptance results of other fulfillment nodes as the sales transaction volume corresponding to the node to which the transaction demand belongs.

[0072] Based on the initial service area, candidate fulfillment nodes capable of responding to the transaction order are identified. By determining candidate fulfillment nodes based on the initial service area, the range of fulfillment options for the transaction order can be limited to fulfillment nodes related to the initial service area, avoiding the direct use of the fulfillment node that ultimately completes the transaction delivery as the transaction demand attribution node. This solves the problem in existing technologies that ignore the initial service area of ​​the transaction order and only determine the transaction demand attribution based on subsequent fulfillment results, leading to the deviation in the determination of the transaction demand attribution node.

[0073] Based on transaction order data and candidate fulfillment nodes, the initial target fulfillment node corresponding to a transaction order at the time of transaction generation is determined. By determining the initial target fulfillment node corresponding to a transaction order at the time of transaction generation, its initial target fulfillment node can be fixed before the transaction order is repositioned, so that even if subsequent repositioning occurs, it is still possible to identify which fulfillment node should have handled the transaction order initially. This solves the problem in existing technologies where the initial target fulfillment node is not sufficiently recorded, resulting in the inability to restore the transaction demand's attribution node after a transaction order is repositioned due to stockouts, frozen inventory, insufficient order processing capacity, or failure to meet delivery timeliness.

[0074] By associating transaction orders, initial service areas, and initial target fulfillment nodes, we obtain transaction demand origination data. This association creates transaction demand origination data that is not overwritten during continuous position rebalancing, ensuring that the transaction demand attribution node and initial target fulfillment node of each transaction order are simultaneously saved. This solves the problem in existing technologies where, after a transaction order is finally completed and delivered, the system typically only retains the actual transaction fulfillment node, leading to the loss of the initial service area and initial target fulfillment node. It provides a stable basis for subsequently determining demand attribution data based on passive position rebalancing identifiers.

[0075] Based on the starting point data of transaction demand, fulfillment migration chain data is generated. This allows for the continuous recording of the fulfillment migration process of a transaction order from its initial target fulfillment node, preventing situations where only the final transaction delivery result remains after multiple position adjustments. This solves the problem in existing technologies where insufficient recording of continuous position adjustments leads to unclear information about which fulfillment node the order was transferred from, into, and when. It provides a foundation of fulfillment migration chain data for subsequent identification of position adjustment reasons and temporary acceptance of transaction demand data.

[0076] It should be noted that the methods for generating fulfillment migration chain data based on the starting point data of transaction demand include:

[0077] Based on the initial data of transaction demand, determine the first target fulfillment node corresponding to the transaction order;

[0078] When a trading order is transferred from its initial target fulfillment node to other fulfillment nodes for processing, the system records the fulfillment node before the position change, the fulfillment node after the position change, and the trigger time of the position change. By recording the fulfillment node before the position change, the fulfillment node after the position change, and the trigger time of the position change when the trading order is first changed, the specific node relationship and time position of the trading order's transfer from the initial target fulfillment node to other fulfillment nodes can be clearly identified. This solves the problem in existing technologies where, after a trading order undergoes its first position change, the system has difficulty determining from which fulfillment node the position change originated, to which fulfillment node it was transferred to, and when it occurred, thus preventing the initial position change process from being overwritten by the final actual trading fulfillment node.

[0079] When a transaction order is transferred from the current fulfillment node to another fulfillment node for processing, the corresponding fulfillment node before the position change, the fulfillment node after the position change, and the position change trigger time are continuously recorded. By continuing to record the fulfillment node before the position change, the fulfillment node after the position change, and the position change trigger time corresponding to the subsequent position change, the multi-level migration relationship of the transaction order during continuous position changes can be completely preserved. This solves the problem in existing technologies that can only record one position change or the final fulfillment result, and cannot reflect the continuous transfer of transaction orders between multiple fulfillment nodes, thereby avoiding the occurrence of multiple overwritings of the transaction demand's attribution node due to continuous position changes.

[0080] After a transaction order is completed and delivered, the actual transaction fulfillment node is determined based on the final fulfillment node of the transaction.

[0081] By linking the initial target fulfillment node, each pre-fulfillment node, each post-fulfillment node, and the actual transaction fulfillment node in chronological order of the position modification trigger time, fulfillment migration chain data can be obtained. This data reflects the complete fulfillment migration process of a transaction order. This solves the problems of fragmented position modification records and lack of sequential relationships between nodes in existing technologies, making it difficult to determine the path of a transaction order migration from the transaction demand node to the actual transaction fulfillment node. It provides a basis for subsequently determining the fulfillment migration path and judging whether sales volume is formed by passive position modification.

[0082] Based on fulfillment migration chain data, the reason data for warehouse relocation is obtained. By obtaining the reason data for warehouse relocation based on fulfillment migration chain data, each warehouse relocation can be mapped to its cause, rather than simply recording that the transaction order node has been transferred. This solves the problem in existing technologies that do not include the reasons for continuous warehouse relocation in the transaction demand attribution correction, which easily leads to the misjudgment of passive warehouse relocation caused by stockouts, inventory freezes, insufficient order processing capacity, or failure to meet delivery timeliness as changes in the transaction demand attribution node.

[0083] It should be noted that the methods for obtaining the reason data for warehouse modification based on fulfillment migration chain data include:

[0084] Based on the fulfillment migration chain data, extract the fulfillment node before the position change, the fulfillment node after the position change, and the position change trigger time for each position change in a transaction order;

[0085] Based on the warehouse relocation trigger time, the system collects the inventory status, order processing capacity status, and delivery timeliness status of the fulfillment node before the relocation. By collecting these data at the warehouse relocation trigger time, the actual fulfillment conditions of the fulfillment node before the relocation can be obtained. This solves the problem in existing technologies where the attribution of transaction demand is determined only after the final transaction delivery result, making it impossible to determine why the transaction order was transferred from the fulfillment node before the relocation. The system ensures that the relocation reason data reflects the fulfillment conditions represented by the inventory status, order processing capacity status, and delivery timeliness status at the time the relocation is triggered.

[0086] Specifically, when collecting the inventory status, order processing capacity status, and delivery timeliness status corresponding to the fulfillment node before the warehouse modification, these statuses are synchronously saved along with the fulfillment node before the warehouse modification, the fulfillment node after the warehouse modification, and the warehouse modification trigger time, ensuring that each warehouse modification corresponds to the fulfillment status at the time of its trigger. When generating subsequent warehouse modification reason data, the inventory status, order processing capacity status, or delivery timeliness status after the transaction order is completed and delivered will not replace the fulfillment status corresponding to the warehouse modification trigger time.

[0087] If the inventory status indicates that the fulfillment node could not provide the goods corresponding to the transaction order before the warehouse change, then the reason for the warehouse change will be determined as the fulfillment node being out of stock.

[0088] If the inventory status indicates that the goods in the fulfillment node before the warehouse change are in a state that cannot be used to complete the transaction delivery, then the reason for the warehouse change is determined to be inventory freeze.

[0089] If the order processing capacity status indicates that the fulfillment node before the repositioning cannot complete the order processing within the processing time required by the transaction order, then the reason for the repositioning is determined to be insufficient order processing capacity.

[0090] If the delivery timeliness status indicates that the fulfillment node before the warehouse relocation cannot complete the transaction delivery within the delivery time required by the transaction order, then the reason for the warehouse relocation is determined as delivery timeliness not being met.

[0091] When a single warehouse relocation corresponds to multiple reasons for relocation, including stockouts at fulfillment nodes, frozen inventory, insufficient order processing capacity, or unmet delivery timeliness, these multiple reasons are collectively linked to the relocation to obtain the corresponding relocation reason data. If any of these reasons exist, and the initial service area corresponding to the transaction address data has not changed, then the relocation is considered a passive relocation.

[0092] By associating the reason for each warehouse relocation with the fulfillment migration chain data, we obtain warehouse relocation reason data. This association ensures that each node in the fulfillment migration chain has corresponding reason information. This solves the problem of existing technologies where the fulfillment migration process and the reason for warehouse relocation are disconnected, preventing the system from judging transaction demand transfers solely based on node changes. It enables accurate differentiation of passive warehouse relocations caused by fulfillment node stockouts, inventory freezes, insufficient order processing capacity, or unmet delivery times.

[0093] Based on the reason data for warehouse relocation, it determines whether a warehouse relocation in the fulfillment migration chain data is a passive relocation. This method distinguishes relocations caused by fulfillment node stockouts, inventory freezes, insufficient order processing capacity, or unmet delivery times from other relocations. It solves the problem in existing technologies where any transaction order completed by another fulfillment node is easily considered a change in the transaction demand attribution node, enabling the system to identify whether a transaction order was forcibly transferred because the fulfillment node before the relocation could not complete the transaction delivery. If so, a passive relocation identifier is generated in the fulfillment migration chain data; based on the passive relocation identifier and the transaction demand starting point data, demand attribution data is obtained; by obtaining demand attribution data based on the passive relocation identifier and the transaction demand starting point data, when a transaction order has a passive relocation, the transaction demand attribution can be redirected to the initial service area and the first target fulfillment node corresponding to the transaction order generation. This solves the problem in existing technologies where it is difficult to distinguish between the transaction demand attribution node and the actual transaction fulfillment node, avoiding the direct identification of an actual transaction fulfillment node as the transaction demand attribution node simply because the actual transaction fulfillment node has completed the transaction delivery.

[0094] It should be noted that, based on the reason data for the change of position, the methods for determining whether the change of position in the fulfillment migration chain data belongs to passive position change include:

[0095] Based on the fulfillment migration chain data, determine each warehouse modification corresponding to a transaction order;

[0096] Based on the data on reasons for position changes, determine the reasons for each position change.

[0097] If the reason for warehouse relocation is stockouts at fulfillment nodes, frozen inventory, insufficient order processing capacity, or failure to meet delivery deadlines, then the corresponding warehouse relocation is determined to be a passive relocation. By classifying warehouse relocations caused by stockouts at fulfillment nodes, frozen inventory, insufficient order processing capacity, or failure to meet delivery deadlines as passive relocations, it can be clarified that such relocations are not due to a natural transfer of transaction demand, but rather are caused by the inability of the fulfillment node to complete transaction delivery before the relocation. This solves the problem in existing technologies of misjudging passive warehouse relocations as an increase in local sales demand at actual transaction fulfillment nodes, and provides a basis for identifying temporary transaction demand from actual transaction fulfillment nodes in the future.

[0098] If the reason for the warehouse relocation is not due to stockouts at the fulfillment node, frozen inventory, insufficient order processing capacity, or failure to meet delivery time, a passive warehouse relocation flag will not be generated, and temporary acceptance of transaction demand will not be determined based on this warehouse relocation.

[0099] It should be noted that the methods for generating passive warehouse modification identifiers in the fulfillment migration chain data include:

[0100] Based on the fulfillment migration chain data, identify the warehouses that were judged to be passively modified.

[0101] Extract the transaction order, fulfillment node before the position change, fulfillment node after the position change, position change trigger time, and position change reason corresponding to the passive position change;

[0102] Associate the transaction order, the fulfillment node before the position change, the fulfillment node after the position change, the position change trigger time, and the position change reason to generate a passive position change identifier corresponding to the passive position change;

[0103] By writing the passive position modification identifier into the corresponding passive position modification in the fulfillment migration chain data, the fulfillment migration chain data can not only record node changes but also whether the node change belongs to passive position modification. This solves the problem in the existing technology that the continuous position modification process only reflects the fulfillment node change and cannot reflect the nature of the position modification. It enables subsequent identification of temporary transaction needs to directly determine whether the sales transaction volume is formed by passive position modification based on the passive position modification identifier in the fulfillment migration chain data.

[0104] It should be noted that the methods for obtaining demand attribution data based on passive position modification identifiers and the starting point data of trading demands include:

[0105] Based on the passive position modification identifier, identify the existence of transaction orders that have been passively modified;

[0106] Based on the initial data of transaction demand, determine the initial service area and the first target fulfillment node corresponding to the transaction order;

[0107] If a transaction order is subject to passive warehousing changes, the initial target fulfillment node is determined as the transaction demand attribution node corresponding to the transaction order. By determining the initial target fulfillment node as the transaction demand attribution node when a transaction order is subject to passive warehousing changes, transaction demands masked by stockouts, frozen inventory, insufficient order processing capacity, or unmet delivery timeliness can be returned to the initial target fulfillment node. This solves the problem in existing technologies where the initial target fulfillment node is misjudged as a decrease in local sales demand due to the failure to complete actual transaction delivery, and avoids the continuous underestimation of this fulfillment node in subsequent transaction supply allocation results.

[0108] By associating transaction orders, initial service areas, first target fulfillment nodes, and transaction demand attribution nodes, demand attribution data can be obtained.

[0109] Based on the fulfillment migration chain data, the actual transaction fulfillment node data is determined. By identifying the actual transaction fulfillment node data based on the fulfillment migration chain data, it is possible to clearly identify which fulfillment node ultimately completes the transaction delivery and to correlate that actual transaction fulfillment node with the fulfillment migration chain data. This solves the problem in existing technologies where, although the actual transaction fulfillment node is recorded, it is not linked to the continuous position modification process of the transaction order, passive position modification identifiers, and demand attribution data, providing a foundation for subsequent comparison of transaction demand attribution nodes and actual transaction fulfillment nodes.

[0110] It should be noted that the methods for determining the actual transaction fulfillment node data based on the fulfillment migration chain data include:

[0111] Based on the fulfillment migration chain data, determine the initial target fulfillment node, each fulfillment node before the position change, each fulfillment node after the position change, and the position change trigger time for each transaction order;

[0112] Based on the order of the position modification trigger time, the fulfillment nodes after each position modification are sorted to determine the last fulfillment node after the position modification in the fulfillment migration process of the transaction order.

[0113] After a transaction order is completed and delivered, the final delivery point is defined as the actual transaction delivery point.

[0114] By associating the transaction order, the last fulfillment node after the position change, and the actual transaction fulfillment node, the actual transaction fulfillment node data is obtained.

[0115] Based on demand attribution data, actual transaction fulfillment node data, and fulfillment migration chain data, temporary transaction acceptance demand in actual transaction fulfillment nodes is identified, and temporary transaction acceptance demand data is obtained.

[0116] It should be noted that, based on demand attribution data, actual transaction fulfillment node data, and fulfillment migration chain data, the methods for identifying temporary transaction fulfillment demands within actual transaction fulfillment nodes and obtaining temporary transaction fulfillment demand data include:

[0117] Based on the demand attribution data, determine the transaction demand attribution node corresponding to the transaction order;

[0118] Based on actual transaction fulfillment node data, determine the actual transaction fulfillment node at which the transaction order is finally delivered;

[0119] By comparing the transaction demand attribution node with the actual transaction fulfillment node, it can be determined whether the transaction demand attribution node of a transaction order is consistent with the final fulfillment node of the completed transaction. This solves the problem in existing technologies where the actual transaction fulfillment node is assumed to be equivalent to the transaction demand attribution node, enabling the system to identify deviations in sales transaction volume attribution caused by continuous position changes.

[0120] If the node attributing the transaction demand is inconsistent with the actual transaction fulfillment node, the fulfillment migration path of the transaction order from the transaction demand node to the actual transaction fulfillment node is determined based on the fulfillment migration chain data. By determining the fulfillment migration path when the transaction demand node and the actual transaction fulfillment node are inconsistent, the specific path of the transaction order from the transaction demand node to the actual transaction fulfillment node can be clearly identified. This solves the problem in the prior art that only it is known that the transaction is ultimately delivered by the actual transaction fulfillment node, but it is impossible to determine whether the transaction order was transferred from another transaction demand node. It provides a path basis for subsequent determination of whether the sales transaction volume completed by the actual transaction fulfillment node belongs to the temporary acceptance of transaction demand.

[0121] Based on the passive position modification identifier in the fulfillment migration path, determine whether the sales transaction volume completed at the actual transaction fulfillment node was formed by passive position modification;

[0122] If the sales volume completed by the actual transaction fulfillment node is formed by passive warehouse repositioning, then the temporary sales volume undertaken by the actual transaction fulfillment node is identified as temporary transaction demand. By identifying the temporary sales volume undertaken by the actual transaction fulfillment node, which is formed by passive warehouse repositioning and completed by the actual transaction fulfillment node, it is possible to distinguish the local sales demand of the actual transaction fulfillment node from its temporary transaction demand undertaken by other fulfillment nodes. This solves the problem in the prior art of not determining whether the sales volume belongs to temporary transaction demand, which leads to the actual transaction fulfillment node being misjudged as having increased local sales demand. Furthermore, it generates temporary transaction demand data by associating transaction orders, transaction demand attribution nodes, actual transaction fulfillment nodes, fulfillment migration paths, and temporary sales volume. By associating transaction orders, transaction demand attribution nodes, actual transaction fulfillment nodes, fulfillment migration paths, and temporary sales volume to generate temporary transaction demand data, it is possible to completely represent the source, receiving node, migration path, and sales volume of temporary transaction demand. This solves the problem in existing technologies where temporary transaction demands lack structured records, making it impossible to accurately deduct and revert them in subsequent sales transaction volume adjustments;

[0123] The temporary sales volume is determined based on the valid transaction content of the actually completed transaction delivery of the order. When a transaction order is split during the fulfillment migration process, the corresponding node sales volume is determined based on the valid transaction content of each actual transaction fulfillment node. For node sales volume that is inconsistent with the node to which the transaction demand belongs and is formed by passive warehouse modification, it is determined as the temporary sales volume. When a transaction order has canceled, returned, or incomplete transaction content, the corresponding transaction content is not included in the temporary sales volume.

[0124] Based on demand attribution data and temporary transaction demand data, the sales transaction volume of each fulfillment node is corrected to obtain the corrected sales demand volume. By correcting the sales transaction volume of each fulfillment node based on demand attribution data and temporary transaction demand data, the sales transaction volume generated by temporary transaction demand in the actual transaction fulfillment node can be separated from its local sales demand volume, and this part of the sales transaction volume can be returned to the corresponding transaction demand attribution node. This solves the problems in the prior art where the sales transaction volume of the actual transaction fulfillment node is inflated due to temporary transaction demand, and the sales demand volume that is not included in the calculation when the transaction demand attribution node has not formed an actual transaction completion record.

[0125] It should be noted that, based on demand attribution data and temporary transaction demand data, the sales transaction volume of each fulfillment node is adjusted to obtain the adjusted sales demand volume. The methods include:

[0126] Based on the temporary acceptance of transaction demand data, determine the temporary acceptance of sales transaction volume formed by temporary acceptance of transaction demand at the actual transaction fulfillment node;

[0127] Based on the demand attribution data, determine the transaction demand attribution node corresponding to the temporarily accepted sales transaction volume.

[0128] Based on transaction order data and actual transaction fulfillment node data, the actual transaction fulfillment node for the final completion of the transaction order is determined, and the node sales transaction volume corresponding to the actual transaction fulfillment node is determined. When a transaction order is completed by multiple actual transaction fulfillment nodes, the node sales transaction volume corresponding to each actual transaction fulfillment node is determined based on the transaction order data, and each actual transaction fulfillment node is compared with the node to which the transaction demand belongs. For node sales transaction volumes that are inconsistent with the node to which the transaction demand belongs and are formed by passive position modification, they are determined as temporary accepted sales transaction volumes.

[0129] Based on the temporarily accepted sales transaction volume, the local sales demand of the actual transaction fulfillment node is obtained by subtracting the temporarily accepted sales transaction volume from the node sales transaction volume corresponding to the actual transaction fulfillment node. By subtracting the temporarily accepted sales transaction volume from the node sales transaction volume corresponding to the actual transaction fulfillment node, the true local sales demand of the actual transaction fulfillment node within its own transaction service scope can be obtained. This solves the problem in the existing technology that considers all sales transactions completed by the actual transaction fulfillment node as its local sales demand, leading to an overestimation of the local sales demand of the actual transaction fulfillment node.

[0130] Specifically, when deducting temporarily accepted sales transaction volume, only valid transaction content formed by passive warehouse relocation and delivered by the actual transaction fulfillment node is deducted; node sales transaction volume formed within the transaction service scope of the actual transaction fulfillment node itself is not deducted as temporary accepted sales transaction volume. When returning temporary accepted sales transaction volume to the transaction demand originating node, the transaction order, transaction demand originating node, actual transaction fulfillment node, and fulfillment migration path corresponding to the temporary accepted sales transaction volume are used as the basis to avoid duplicate return of sales transaction volume corresponding to the same transaction order;

[0131] Based on the transaction demand attribution node and the temporary sales transaction volume, the temporary sales transaction volume is returned to the sales transaction volume of the transaction demand attribution node to obtain the replenished sales demand volume of the transaction demand attribution node. By returning the temporary sales transaction volume to the sales transaction volume of the transaction demand attribution node, the sales demand of the transaction demand attribution node that did not form an actual transaction completion record due to passive position rebalancing can be recovered. This solves the problem in the prior art where the transaction demand attribution node is misjudged as a decrease in local sales demand due to incomplete transaction delivery, allowing the true transaction demand of the transaction demand attribution node that was masked to be included in the subsequent transaction supply allocation results.

[0132] Based on local sales demand and replenished sales demand, revised sales demand is generated for each fulfillment node. By generating revised sales demand for each fulfillment node based on local and replenished sales demand, the artificially inflated sales demand caused by temporarily accepting transaction demand in actual transaction fulfillment nodes can be reduced, while the sales demand masked by passive position adjustments in the transaction demand attribution node can be increased. This solves the problem in existing technologies where some fulfillment nodes are continuously overestimated due to temporarily accepting transaction demand, and some fulfillment nodes are continuously underestimated due to a lack of actual transaction completion records.

[0133] Based on the revised sales demand, a transaction supply configuration result is generated. This ensures that the transaction supply configuration result is based on the revised actual transaction demand, rather than on the sales transaction volume corresponding to the actual transaction fulfillment node. This solves the problem in existing technologies where the transaction supply configuration result continuously tilts towards the actual transaction fulfillment node, leading to an underestimation of the transaction supply configuration result for the transaction demand-attributed node and an overestimation of the transaction supply configuration result for the actual transaction fulfillment node. This improves the accuracy and fulfillment efficiency of omnichannel transaction order processing.

[0134] It should be noted that the methods for generating transaction supply configuration results based on corrected sales demand include:

[0135] Based on the revised sales demand, the sales demand attribution results corresponding to each fulfillment node are determined;

[0136] Among them, the sales demand attribution result is used to represent the sales demand attribution situation of each fulfillment node based on the corrected sales demand quantity;

[0137] Based on the sales demand attribution results, determine the configuration direction of each fulfillment node in the transaction supply configuration results;

[0138] Among them, the configuration direction is used to indicate the direction in which the transaction supply configuration result increases or decreases the configuration of the corresponding fulfillment node;

[0139] When the revised sales demand of a fulfillment node is higher than the original sales transaction volume, the allocation priority of that fulfillment node in the transaction supply allocation result is increased;

[0140] Among them, the configuration priority is used to indicate the priority order of each fulfilling node in the transaction supply configuration result; the original sales transaction volume is the node sales transaction volume before the correction.

[0141] When the revised sales demand of a fulfillment node is lower than its original sales transaction volume, the configuration priority of that fulfillment node in the transaction supply configuration result is reduced; the transaction supply configuration result is generated based on the configuration priority of each fulfillment node.

[0142] The aforementioned transaction supply allocation results represent the commodity supply allocation content formed by each fulfilling node based on the revised sales demand. For transaction demand nodes where the revised sales demand is higher than the original sales transaction volume, the allocation priority of the corresponding commodities for that transaction demand node is increased, and the transaction supply allocation results are added to that transaction demand node. For actual transaction fulfilling nodes where the revised sales demand is lower than the original sales transaction volume, the allocation priority formed by temporarily accepting sales transaction volume is reduced, and the transaction supply allocation results are less erroneously tilted towards that actual transaction fulfilling node. This ensures that the transaction supply allocation results simultaneously reflect both local sales demand and replenished sales demand.

[0143] In summary, this method retains the initial service area and first target fulfillment node corresponding to the transaction order's generation through transaction demand origination data. It records the continuous warehouse modification process of transaction orders through fulfillment migration chain data. It identifies passive warehouse modifications caused by fulfillment node stockouts, inventory freezes, insufficient order processing capacity, or unmet delivery times through warehouse modification reason data and passive warehouse modification identifiers. It determines the transaction demand attribution node corresponding to the transaction order through demand attribution data. It identifies temporary transaction demand formed by passive warehouse modifications in actual transaction fulfillment nodes through temporary acceptance transaction demand data. Finally, it deducts and reverts sales transaction volumes at each fulfillment node by correcting sales demand, ensuring that the transaction supply allocation result corresponds to the actual transaction demand. This solves the problems in existing technologies where actual transaction fulfillment nodes are misjudged as transaction demand attribution nodes, sales demand at transaction demand attribution nodes is masked, continuous warehouse modification reasons are not included in demand attribution correction, temporary acceptance transaction demand is not identified, and transaction supply allocation results are offset.

[0144] Example 2:

[0145] See Figure 2 A big data-driven intelligent supply chain management system includes:

[0146] The transaction start point generation module is used to obtain transaction order data and transaction address data, and based on the transaction order data and transaction address data, obtain transaction requirement start point data;

[0147] The data processing module is used to generate fulfillment migration chain data based on the transaction demand starting point data, obtain warehouse modification reason data based on the fulfillment migration chain data, determine whether the warehouse modification in the fulfillment migration chain data is a passive warehouse modification based on the warehouse modification reason data, and generate a passive warehouse modification identifier in the fulfillment migration chain data when the warehouse modification is a passive warehouse modification. The data processing module is also used to simultaneously save the inventory status, order processing capacity status, and delivery time status corresponding to the warehouse modification trigger time when recording the fulfillment node before warehouse modification, the fulfillment node after warehouse modification, and the warehouse modification trigger time, and generate warehouse modification reason data based on the above status. When the same warehouse modification corresponds to multiple warehouse modification reasons, the multiple warehouse modification reasons are jointly associated with the corresponding warehouse modification in the fulfillment migration chain data.

[0148] The temporary transaction generation module is used to obtain demand attribution data based on passive position modification identifiers and transaction demand origination data, determine actual transaction fulfillment node data based on fulfillment migration chain data, and identify temporary transaction acceptance demands in actual transaction fulfillment nodes based on demand attribution data, actual transaction fulfillment node data, and fulfillment migration chain data, thereby obtaining temporary transaction acceptance demand data. The temporary transaction generation module is also used to determine whether the transaction demand origination data remains consistent during the fulfillment migration process based on transaction address data and initial service area, and to determine the transaction demand attribution node when the transaction demand origination data remains consistent and a passive position modification identifier exists in the fulfillment migration chain data. Furthermore, it is used to generate temporary transaction acceptance demand data based on the valid transaction content of the actually completed transaction delivery when a transaction order is split for fulfillment, canceled, or returned.

[0149] The transaction configuration generation module is used to correct the sales transaction volume of each fulfillment node based on the demand attribution data and the temporary acceptance of transaction demand data, to obtain the corrected sales demand volume, and to generate the transaction supply configuration result based on the corrected sales demand volume. The transaction configuration generation module is also used to deduct the temporary acceptance of sales transaction volume from the node sales transaction volume corresponding to the actual transaction fulfillment node to obtain the local sales demand volume, and to return the temporary acceptance of sales transaction volume to the transaction demand attribution node to obtain the replenished sales demand volume. It is also used to generate the transaction supply configuration result based on the local sales demand volume and the replenished sales demand volume.

[0150] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the big data-driven intelligent supply chain management method and system and its inventive concept, should be covered within the scope of protection of the present invention.

Claims

1. A big data-driven intelligent supply chain management method, characterized in that, Includes the following steps: Obtain transaction order data and transaction address data, and based on the transaction order data and transaction address data, obtain the starting point data of transaction demand; Generate performance migration chain data based on transaction demand starting point data; Based on the fulfillment migration chain data, the reasons for the warehouse modification were obtained; Based on the reason data for warehouse modification, determine whether the warehouse modification in the fulfillment migration chain data belongs to passive warehouse modification. If it does, generate a passive warehouse modification identifier in the fulfillment migration chain data. Based on the passive position modification identifier and the starting point data of the transaction demand, the demand attribution data is obtained; Based on the performance migration chain data, the actual transaction performance node data is determined; Based on demand attribution data, actual transaction fulfillment node data, and fulfillment migration chain data, temporary transaction acceptance demand in actual transaction fulfillment nodes is identified, and temporary transaction acceptance demand data is obtained. Based on demand attribution data and temporary transaction demand data, the sales transaction volume of each fulfillment node is adjusted to obtain the adjusted sales demand volume. Based on the revised sales demand, a transaction supply configuration result is generated.

2. The big data-driven intelligent supply chain management method according to claim 1, characterized in that, Based on demand attribution data, actual transaction fulfillment node data, and fulfillment migration chain data, methods for identifying temporary transaction fulfillment demands within actual transaction fulfillment nodes and obtaining temporary transaction fulfillment demand data include: Based on the demand attribution data, determine the transaction demand attribution node corresponding to the transaction order; Based on actual transaction fulfillment node data, determine the actual transaction fulfillment node at which the transaction order is finally delivered; Compare the node to which the transaction demand is assigned with the actual transaction fulfillment node; If the node to which the transaction demand belongs is inconsistent with the actual transaction fulfillment node, the fulfillment migration path of the transaction order from the node to which the transaction demand belongs to the actual transaction fulfillment node is determined based on the fulfillment migration chain data. Based on the passive position modification identifier in the fulfillment migration path, determine whether the sales transaction volume completed at the actual transaction fulfillment node was formed by passive position modification; If the sales volume completed at the actual transaction fulfillment node is formed by passive warehouse repositioning, then the temporary sales volume completed at the actual transaction fulfillment node will be identified as temporary transaction demand, and the transaction order, the transaction demand attribution node, the actual transaction fulfillment node, the fulfillment migration path, and the temporary sales volume will be associated to generate temporary transaction demand data.

3. The big data-driven intelligent supply chain management method according to claim 1, characterized in that, Based on demand attribution data and temporary transaction demand data, the sales transaction volume of each fulfillment node is adjusted to obtain the adjusted sales demand volume. Methods include: Based on the temporary acceptance of transaction demand data, determine the temporary acceptance of sales transaction volume formed by temporary acceptance of transaction demand at the actual transaction fulfillment node; Based on the demand attribution data, determine the transaction demand attribution node corresponding to the temporarily accepted sales transaction volume. Based on transaction order data and actual transaction fulfillment node data, determine the actual transaction fulfillment node at which the transaction order is finally completed and delivered, and determine the node sales transaction volume corresponding to the actual transaction fulfillment node; Based on the temporary sales transaction volume, the local sales demand of the actual transaction fulfillment node is obtained by deducting the temporary sales transaction volume from the node sales transaction volume corresponding to the actual transaction fulfillment node. Based on the transaction demand attribution node and the temporary sales transaction volume, the temporary sales transaction volume is returned to the sales transaction volume of the transaction demand attribution node to obtain the replenished sales demand volume of the transaction demand attribution node. Based on local sales demand and replenished sales demand, the revised sales demand for each fulfillment node is generated.

4. The big data-driven intelligent supply chain management method according to claim 1, characterized in that, Methods for obtaining the starting point data of transaction demand based on transaction order data and transaction address data include: Based on the transaction address data, the transaction address corresponding to the transaction address data is matched with the transaction service range pre-configured by each fulfillment node to determine the initial service area corresponding to the transaction order; Based on the initial service area, candidate fulfillment nodes capable of responding to the transaction order are identified; Based on transaction order data and candidate fulfillment nodes, determine the first target fulfillment node corresponding to the transaction order when the transaction is generated; By associating transaction orders, initial service areas, and initial target fulfillment nodes, we can obtain the starting point data for transaction demand.

5. The big data-driven intelligent supply chain management method according to claim 1, characterized in that, Methods for generating fulfillment migration chain data based on transaction demand starting point data include: Based on the initial data of transaction demand, determine the first target fulfillment node corresponding to the transaction order; When a transaction order is transferred from the initial target fulfillment node to other fulfillment nodes for processing, the fulfillment node before the position change, the fulfillment node after the position change, and the position change trigger time are recorded. When a transaction order is transferred from the current fulfillment node to another fulfillment node for processing, the corresponding fulfillment node before the position change, the fulfillment node after the position change, and the position change trigger time are continued to be recorded. After a transaction order is completed and delivered, the actual transaction fulfillment node is determined based on the final fulfillment node of the transaction. Based on the order of the position change trigger time, the initial target fulfillment node, each fulfillment node before the position change, each fulfillment node after the position change, and the actual transaction fulfillment node are associated to obtain the fulfillment migration chain data.

6. The big data-driven intelligent supply chain management method according to claim 1, characterized in that, Methods for obtaining warehouse relocation reason data based on fulfillment migration chain data include: Based on the fulfillment migration chain data, extract the fulfillment node before the position change, the fulfillment node after the position change, and the position change trigger time for each position change in a transaction order; Based on the warehouse change trigger time, collect the inventory status, order processing capacity status and delivery timeliness status of the fulfillment node before the warehouse change; If the inventory status indicates that the fulfillment node could not provide the goods corresponding to the transaction order before the warehouse change, then the reason for the warehouse change will be determined as the fulfillment node being out of stock. If the inventory status indicates that the goods in the fulfillment node before the warehouse change are in a state that cannot be used to complete the transaction delivery, then the reason for the warehouse change is determined to be inventory freeze. If the order processing capacity status indicates that the fulfillment node before the repositioning cannot complete the order processing within the processing time required by the transaction order, then the reason for the repositioning is determined to be insufficient order processing capacity. If the delivery timeliness status indicates that the fulfillment node before the warehouse relocation cannot complete the transaction delivery within the delivery time required by the transaction order, then the reason for the warehouse relocation is determined as delivery timeliness not being met. By associating the reason for each position change with the fulfillment migration chain data, we can obtain the position change reason data.

7. The big data-driven intelligent supply chain management method according to claim 1, characterized in that, Based on the reason data for the change of position, methods for determining whether a change of position in the fulfillment migration chain data is a passive change of position include: Based on the fulfillment migration chain data, determine each warehouse modification corresponding to a transaction order; Based on the data on reasons for position changes, determine the reasons for each position change. If the reason for changing the warehouse is out of stock at the fulfillment node, frozen inventory, insufficient order processing capacity, or failure to meet delivery time, then the corresponding warehouse change is determined to be a passive warehouse change.

8. The big data-driven intelligent supply chain management method according to claim 1, characterized in that, Methods for generating passive warehouse relocation identifiers in fulfillment migration chain data include: Based on the fulfillment migration chain data, identify the warehouses that were judged to be passively modified. Extract the transaction order, fulfillment node before the position change, fulfillment node after the position change, position change trigger time, and position change reason corresponding to the passive position change; Associate the transaction order, the fulfillment node before the position change, the fulfillment node after the position change, the position change trigger time, and the position change reason to generate a passive position change identifier corresponding to the passive position change; Write the passive warehouse modification identifier into the corresponding passive warehouse modification data in the fulfillment migration chain data.

9. The big data-driven intelligent supply chain management method according to claim 1, characterized in that, Methods for obtaining demand attribution data based on passive position rebalancing identifiers and the starting point data of trading demands include: Based on the passive position modification identifier, identify the existence of transaction orders that have been passively modified; Based on the initial data of transaction demand, determine the initial service area and the first target fulfillment node corresponding to the transaction order; If a transaction order involves passive position modification, the first target fulfillment node will be determined as the transaction demand attribution node corresponding to the transaction order; By associating transaction orders, initial service areas, first target fulfillment nodes, and transaction demand attribution nodes, demand attribution data can be obtained.

10. A big data-driven intelligent supply chain management system, characterized in that, Used to perform the method according to any one of claims 1-9.

Images