A multi-platform order aggregation processing system
By using a multi-platform order aggregation and processing system, the problems of inconsistent order data and improper resource utilization have been solved, thereby improving the stability and efficiency of order fulfillment.
Patent Information
- Application Number
- CN202611402444.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-10
- Publication Date
- 2026-11-17
AI Technical Summary
Existing multi-platform order processing methods cannot simultaneously consider the current operational load of warehouse nodes, vehicle arrival time, vehicle remaining load, remaining loading space, and platform delivery time windows. This results in inconsistent fulfillment requirements for the same physical inventory on different order platforms, which can easily lead to problems such as duplicate pre-occupancy, over-selling, and duplicate resource scheduling.
Design a multi-platform order aggregation and processing system, including an order access module, an order unification module, an inventory processing module, a warehousing and distribution matching module, and a fulfillment correction module. By adding source identifiers, field mappings, inventory pre-occupancy vouchers, and resource replacements, a unified order object and a corrected fulfillment link are generated to ensure the effective utilization of inventory and resources.
It enables unified processing of product identification and status across different order platforms, avoiding duplicate inventory releases and resource scheduling, and improving the stability and efficiency of order fulfillment.
Smart Images

Figure CN123117987A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of order data processing and warehousing and distribution collaboration technology, and more specifically, to a multi-platform order aggregation processing system. Background Technology
[0002] With the development of traditional e-commerce and instant retail businesses, a single business entity typically needs to receive orders from multiple order platforms simultaneously and utilize supply chain systems, warehouse management systems, and delivery management systems to complete product outbound and last-mile delivery. Different order platforms use inconsistent product codes, order fields, status values, and data update time fields, and the delivery time windows for regular e-commerce orders, pre-orders, and hourly delivery orders differ significantly. Therefore, order data conversion and inventory status synchronization are necessary between multiple order platforms and the warehousing and delivery system.
[0003] Existing multi-platform order processing methods typically integrate orders from each platform separately, then publish available inventory to the platforms based on warehouse inventory levels, or simply aggregate inventory from multiple warehousing nodes to form a shared inventory. This approach struggles to simultaneously consider factors such as the current workload of warehousing nodes, vehicle arrival time, remaining vehicle load, remaining loading space, and platform delivery time windows. This means the same physical inventory might be fulfillable on a regular e-commerce platform but not on a real-time retail platform. Furthermore, when two platforms simultaneously request the last item, deducting based solely on the read inventory quantity can easily lead to duplicate pre-allocation or overselling. Additionally, order execution may encounter warehouse picking delays, vehicle delays, or changes in delivery time. Existing methods often require reselecting warehouses, inventory, and delivery vehicles, making it difficult to distinguish between completed or still-valid fulfillment nodes and those truly affected by delays. This can result in duplicate inventory releases and pre-allocations, redundant resource scheduling, and unnecessary restructuring of the order fulfillment chain. Therefore, inventing a multi-platform order aggregation processing system to address these issues has become a pressing problem for those skilled in the art. Summary of the Invention
[0004] To overcome the above shortcomings, this invention provides a multi-platform order aggregation processing system, which aims to solve the problems of inconsistent order fields across multiple platforms, concurrent conflicts in shared inventory, disconnect between heavy goods inventory and delivery capacity, and repeated construction of the entire chain due to local delays.
[0005] This invention is implemented as follows: This invention provides a multi-platform order aggregation processing system, the system comprising: The module includes an order access module, an order unification module, an inventory processing module, a warehouse and distribution matching module, and a fulfillment correction module. The order access module receives order data output from multiple order platforms and adds source identifiers to generate a cloud platform order dataset; The order unification module performs field mapping on the cloud platform order dataset to generate a unified order object, and generates order fulfillment features based on product attributes, delivery time windows, and last-mile delivery conditions. The inventory processing module filters out a set of candidate inventory based on the warehousing node operation time, delivery resources, and order fulfillment characteristics. The warehousing and distribution matching module determines the target inventory unit, target warehousing node, and target delivery resource from the candidate warehousing and distribution combinations and generates an order fulfillment chain. The fulfillment correction module calculates the remaining fulfillment time based on the actual processing status of the order fulfillment link. When the remaining fulfillment time does not meet the delivery time window, it determines the affected fulfillment node range and performs resource replacement to generate a corrected order fulfillment link.
[0006] Preferably, the steps for generating the cloud platform order dataset are as follows: Establish data access channels corresponding to traditional e-commerce platforms, instant retail platforms, supply chain management systems, warehousing and distribution management systems, and delivery management systems; Receive order data through various data access channels, add platform source identifiers, and extract platform order identifiers and data reception times; The order data is arranged according to the data reception time to generate a cloud platform order dataset.
[0007] Preferably, the steps for generating a unified order object are as follows: Identify product identifiers, product quantities, delivery locations, order status, required delivery time, and service requirements from the cloud platform order dataset; Based on the field mapping relationship corresponding to each order platform, the identified fields are name-converted to generate a standard order field set; A unified order object is generated by associating the platform order identifier with the standard order field.
[0008] Preferably, the steps for generating order fulfillment characteristics are as follows: Extract the product weight, product volume, and required delivery time from the unified order object; Determine vehicle access conditions and loading / unloading assistance requirements based on the receiving location and service requirements, and generate final delivery conditions; Combining product weight, product volume, required delivery time, and last-mile delivery conditions into order fulfillment characteristics, and matching them with the operational capacity of warehousing nodes and the transportation capacity of distribution resources, respectively.
[0009] Preferably, the steps for generating the candidate inventory set are as follows: Determine the available inventory level based on physical inventory, locked inventory, and isolated inventory; The estimated outbound time is obtained based on the operational status of the warehousing node where each inventory unit is located, and the estimated delivery time is obtained based on the delivery relationship between the warehousing node and the receiving location. The estimated outbound time and estimated delivery time are combined to obtain the fulfillable completion time. Inventory units with fulfillable completion times within the delivery time window and available inventory greater than zero are grouped into a candidate inventory set.
[0010] Preferably, the steps for determining the platform's published inventory quantity are as follows: Obtain the location of candidate inventory units, available inventory, warehouse node workload, and remaining capacity of available delivery resources to generate warehouse and distribution status data; Based on the warehouse and distribution status data and the delivery time windows corresponding to each order platform, inventory quantity constraints are applied to determine the platform-released inventory quantity for each order platform. When the status of warehousing or distribution resources changes, the inventory published on the platform will be recalculated and updated.
[0011] Preferably, the step of performing inventory pre-positioning for the target inventory unit is as follows: Extract the current available inventory quantity and current inventory version number of the target inventory unit, and generate an inventory reservation voucher based on the order identifier, target inventory unit identifier, reserved quantity, current inventory version number, generation time and validity period; The inventory version number in the inventory reservation document is compared with the current inventory version number of the target inventory unit to generate a version comparison result. Inventory pre-allocation is performed when the version number is consistent and the current inventory quantity is not less than the pre-allocated quantity, and the inventory version number is updated.
[0012] Preferably, the steps for determining the target inventory unit, target warehousing node, and target distribution resource are as follows: Based on the order fulfillment characteristics, a set of fulfillment constraints is formed, including the weight and volume of the goods, vehicle type and traffic conditions, and loading and unloading assistance requirements. A delivery capacity set is formed based on the rated load, remaining load, remaining loading space, existing delivery tasks, vehicle model information, and the status of loading and unloading auxiliary equipment of the delivery vehicles; The set of performance constraints and the set of delivery capabilities are matched one by one to generate a set of candidate delivery resources, and candidate warehousing and distribution combinations are generated by combining the set of candidate inventory and the delivery time window. From the candidate warehousing and distribution combinations, identify the target inventory units, target warehousing nodes, and target delivery resources that meet the delivery time window.
[0013] Preferably, the steps for calculating the remaining performance time are as follows: Identify inventory pre-positioning nodes, warehousing operation nodes, loading nodes, delivery nodes, and home delivery nodes in the order fulfillment chain; Obtain the planned completion time and actual completion time of each fulfillment node, identify unfulfilled fulfillment nodes based on their execution status, and obtain the corresponding estimated processing time. Each incomplete fulfillment node and its estimated processing time are combined to form an incomplete node duration set. The estimated processing times are then merged to obtain the remaining fulfillment time for the order fulfillment chain.
[0014] Preferably, the steps for generating a corrected order fulfillment chain are as follows: Compare the remaining performance time with the delivery time window to generate a performance timeout determination result, and locate the delayed performance node whose processing time has changed when the determination result is a timeout. Based on the time impact of delayed performance nodes on subsequent performance nodes, a delay propagation determination is performed to identify the range of affected performance nodes; Retain fulfillment nodes outside the affected fulfillment node range, and replace warehousing nodes, delivery resources, or loading and unloading auxiliary equipment within the range to generate a corrected order fulfillment chain.
[0015] The beneficial effects of this invention are: 1. This invention adds source identifiers to order data from different order platforms through an order access module and performs field mapping through an order unification module, which can convert product identifiers, product quantities, delivery locations, order status, delivery time windows, and service requirements into a unified order object. Based on this, order fulfillment characteristics are formed by combining product weight, product volume, vehicle type access conditions, and loading and unloading assistance requirements, enabling traditional e-commerce orders, instant retail orders, and supply chain-side order statuses to be processed for subsequent inventory and warehousing under the same data structure.
[0016] 2. This invention combines available inventory, estimated operation time of warehousing nodes, remaining carrying capacity of delivery resources, and delivery time windows of each platform to screen candidate inventory, so that the same physical inventory can form corresponding platform-released inventory according to the fulfillment requirements of different platforms; at the same time, consistency verification is performed through inventory version number and inventory pre-occupancy voucher to avoid duplicate deduction of the same inventory for multiple platform orders, and to prevent heavy goods inventory lacking suitable vehicles or loading and unloading auxiliary equipment from entering the immediately fulfillable inventory.
[0017] 3. This invention binds target inventory units, target warehousing nodes, and target delivery resources into an order fulfillment chain through a warehousing and distribution matching module. The fulfillment correction module calculates the remaining fulfillment time based on the estimated processing time of the unfulfilled fulfillment nodes. When the expected delivery time window cannot be met, the invention locates the delayed fulfillment nodes whose processing time has changed and their affected range. Nodes that have been completed or are still valid outside the affected fulfillment node range are retained. Only the warehousing nodes, delivery resources, or loading and unloading auxiliary equipment within the range are replaced. This reduces the repeated release and pre-occupancy of inventory and the ineffective switching of warehousing and distribution resources, thereby improving the correction efficiency of abnormal orders and the stability of the fulfillment chain. Attached Figure Description
[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.
[0019] Figure 1 This is a flowchart of a multi-platform order aggregation processing system provided by an embodiment of the present invention. Detailed Implementation
[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0021] This invention provides a multi-platform order aggregation and processing system. The system includes an order access module, an order unification module, an inventory processing module, a warehouse-distribution matching module, and a fulfillment correction module. Each module is deployed on a cloud platform server and communicates with a traditional e-commerce platform, an instant retail platform, a supply chain management system, a warehouse management system, and a delivery management system, respectively. The order access module outputs a cloud platform order dataset to the order unification module; the order unification module outputs a unified order object and order fulfillment characteristics to the inventory processing module; the inventory processing module outputs a candidate inventory set and inventory pre-positioning results to the warehouse-distribution matching module; and the warehouse-distribution matching module outputs the order fulfillment chain to the fulfillment correction module. The following describes the process in order to... Figure 1 Steps S11 to S15 in the text describe the continuous processing procedure.
[0022] Step S11: Receive order data output from multiple order platforms and add source identifiers to generate a cloud platform order dataset.
[0023] The order access module pre-establishes a platform access configuration table, which records at least the platform type, interface address, authentication method, data transmission method, order status code, and data update time field. Traditional e-commerce platforms, instant retail platforms, supply chain management systems, and warehousing and distribution systems each correspond to a data access channel, which can adopt application programming interfaces, message queues, or data file interfaces.
[0024] After receiving an order data, the order access module first writes the platform source identifier P according to the data access channel. i And extract the platform order identifier I i plat Platform order status S i Field update time τ i upd Based on the data reception time, order data received within the processing period t is arranged according to the data reception time to form a cloud platform order dataset: ; In the formula, This represents the i-th order data with the platform source identifier, and n represents the total number of order data within the processing cycle. The order data includes at least the product identifier, product quantity, delivery location, order status, required delivery time, and service requirements.
[0025] To prevent duplicate orders from being generated on the same platform due to retrying push notifications, the order access module generates an idempotent verification key for each order data entry: ; In the formula, For order data An idempotent check key is used when new order data is received. and Same as existing orders, and If the data is updated no later than the field update time of the existing data, it is determined to be duplicate data and further writing is stopped; when When the data is later than the existing data, the existing data is updated with the new order status and new field values; thus, duplicate pushes, status updates and cancellation notifications can all be classified into the same platform order.
[0026] After the cloud platform order dataset is generated, the order access module sends the dataset identifier, processing cycle, and storage location of each order data to the order unification module, enabling the order unification module to call the corresponding field mapping relationship according to the platform source.
[0027] Step S12: Perform field mapping on the cloud platform order dataset to generate a unified order object and generate order fulfillment features.
[0028] The order unification module reads product identifiers, product quantities, delivery locations, order status, required delivery time, and service requirements from order data. It then establishes a field mapping table for each order platform, recording the platform's original field names, standard field names, data types, and status value conversion relationships. For example, it uniformly converts the product code field from different platforms into a standard product identifier. Convert the quantity field to the number of goods. Convert the scheduled delivery time or hourly delivery deadline into a delivery time window. , ].
[0029] The order unification module performs field name conversion, data type conversion, and status value conversion according to the field mapping table, and then generates a unified order object by associating the converted standard fields with the platform order identifier. ; In the formula, For internal order identification within the system; This serves as a platform source and platform order identifier; For standard product labeling; For the quantity of goods; To standardize the receiving location; The order status is standard. and These are the start and end times of the delivery time window, respectively; For service requirements.
[0030] For regular e-commerce orders that do not provide a clear start time, the time when the order enters the system will be the start time of the delivery time window, and the latest delivery time promised by the platform will be the end time. For orders with hourly delivery, the time when the order is placed will be the start time, and the end time will be determined according to the 120-minute or 180-minute fulfillment time limit promised by the platform. The standardized delivery location should at least include the delivery area, road traffic restrictions, floor, and whether there is a freight elevator.
[0031] The order unification module is based on standard product identifiers. Retrieve the unit weight and unit volume from the product master data, and then determine the delivery location. and service requirements Determine vehicle type access conditions and loading / unloading support requirements, and generate order fulfillment characteristics: ; In the formula, This refers to the total weight of the items in the order. This refers to the total volume of the ordered items. This represents the current remaining performance period; Conditions for vehicle type to pass; For loading and unloading assistance requirements, the total weight and total volume of goods are respectively composed of the weight of each item, the volume of each item, and the quantity of goods. Multiply them to get the result.
[0032] At the current time t, the remaining fulfillment time of the order is determined according to the following formula: ; For example, product Z has a unit weight of 450 kg and a volume of 1.8 m³. 3 If an order for item Z requires delivery within 180 minutes, the delivery location must be accessible to medium-sized trucks, and a hydraulic tailgate is required, then... =450 kg =1.8 m 3 The corresponding 180-minute remaining fulfillment time, vehicle type access conditions, and hydraulic tailgate requirements are written into the order fulfillment characteristics. The order fulfillment characteristics are then matched with the warehouse node operation capacity and the delivery resource transportation capacity, respectively.
[0033] Step S13: Select a candidate inventory set based on the warehouse node operation time, delivery resources and order fulfillment characteristics, and execute platform inventory release and inventory pre-positioning.
[0034] The inventory processing module uses a combination of "standard product identifier, warehouse node, and inventory batch" as the inventory unit s. It retrieves the physical inventory quantity, locked inventory quantity, isolated inventory quantity, warehouse node, and inventory version number of the inventory unit from the warehouse management system. The available inventory quantity is determined using the following formula: ; In the formula, (s,t) represents the physical inventory level; (s,t) represents the inventory that has been locked or officially occupied by other orders; (s,t) represents the isolated inventory that cannot be released temporarily due to order status pending confirmation, quality inspection, or inventory discrepancies. (s,t) represents the inventory available for new orders, where the available inventory is less than the number of items in the order. Inventory units will not proceed to the next screening stage.
[0035] The inventory processing module reads the number of pending picking tasks, the number of available workers or equipment, and the current loading station status from the warehouse management system, and calculates the estimated picking and outbound time accordingly. and loading time The location, existing tasks, remaining load, remaining loading space, and estimated arrival time of available vehicles are read from the delivery management system to calculate the vehicle arrival time. and estimated delivery time Then, the delivery time is determined based on the floor number of the order and the home delivery service requirements. And set a performance buffer period. The expected duration of performance is uniformly expressed as: ; This formula uniformly includes the time taken for six parts: vehicle arrival at the warehouse, picking and dispatching, loading, delivery, home delivery, and buffer. For idle vehicles already located at the warehouse node, Set to 0; for vehicles that need to enter the storage node from other locations, Use the estimated delivery time to the warehouse to avoid using different estimated fulfillment time standards for ordinary goods and heavy goods.
[0036] The inventory processing module aggregates inventory units with available inventory not less than the order quantity and estimated fulfillment time not exceeding the remaining fulfillment time of the order into a candidate inventory set for order i:
[0037] ;
[0038] For each inventory unit in the candidate inventory set, the inventory processing module further determines the fulfillment time limit corresponding to the order platform p. Generate fulfillment adaptation identifiers: ; When =1, it means that inventory unit s can meet the fulfillment time limit of platform p under the current warehousing and distribution status; when When =0, even if the inventory unit has physical inventory, it will not be published as an immediately fulfillable inventory on the platform.
[0039] The inventory processing module obtains the set of candidate delivery resources corresponding to inventory unit s. After deducting the load and loading space occupied by the assigned tasks, it summarizes the remaining load and remaining loading space of the candidate vehicles that can reach the inventory unit and complete the delivery within the fulfillment time limit of platform p, and obtains the corresponding total remaining load Λ. p (s,t) and total remaining loading space Γ p (s,t), and based on this, determine the inventory quantity to be released by the platform: ; In the formula, floor(·) represents rounding down. The formula uses the available inventory, the number of items that the vehicle can carry, and the number of items that the vehicle space can accommodate as upper limits, respectively. Therefore, the inventory released by the platform will not exceed the number of items that can be delivered in the current actual situation.
[0040] Taking the last item X in warehouse node A as an example, Q a =1、Q l =0、Q g =0, therefore Q b =1; The idle vehicle is already located at the warehouse node. =0; the picking, loading, delivery, home delivery, and buffering times are 35 min, 15 min, 65 min, 15 min, and 10 min respectively, then: ; The standard e-commerce platform P1 requires delivery to the customer within 24 hours, and the calculated total fulfillment time is 140 minutes, so its fulfillment compatibility indicator is 1. The instant retail platform P2 requires delivery to the customer within 120 minutes. Since 140 minutes > 120 minutes, its fulfillment compatibility indicator is 0. Therefore, the same product X has an inventory quantity of 1 on platform P1 and an inventory quantity of 0 on platform P2.
[0041] When the workload of warehouse node A decreases, reducing the picking and outbound time from 35 minutes to 25 minutes, and new idle vehicles reduce the delivery time from 65 minutes to 40 minutes, the updated estimated fulfillment time is: ; Since 105 min ≤ 120 min, the inventory processing module updates the fulfillment adaptation flag of platform P2 to 1 and updates its platform published inventory quantity to 1; when an order from platform P1 or platform P2 successfully pre-books product X, the available inventory quantity becomes 0, and the inventory processing module updates the published inventory quantity of both platforms to 0 simultaneously.
[0042] After an order selects a target inventory unit, the inventory processing module reads the current available inventory quantity and current inventory version number of the target inventory unit, and generates an inventory reservation voucher based on the order identifier, target inventory unit identifier, reserved quantity, current inventory version number, voucher generation time, and voucher validity period. ; In the formula, I i For internal order identification; S i inv Identify the target inventory unit; q i For the quantity reserved; v i The inventory version number read when generating the voucher; τ i For voucher generation time; Δτ i The voucher validity period is used as a reference. Before writing the pre-allocation result, the inventory processing module reads the current version number and available inventory quantity of the target inventory unit again and generates the pre-allocation determination result. ; When Zi=1, perform inventory pre-positioning and synchronously update the available inventory quantity and inventory version number: ; When two platform orders simultaneously reserve the last item of product Y's inventory and both read the same version number v, the order that completes the atomic comparison and update first will successfully verify its inventory, changing the available inventory of product Y from 1 to 0 and updating the inventory version number from v to v+1. The order that completes the verification later will still carry the version number v, which is inconsistent with the current version number v+1, so the verification will fail and no duplicate deduction will be performed. The inventory version number is a monotonically increasing integer, incrementing by 1 after each reserve, release, or isolation status change.
[0043] For orders that fail to be reserved, the inventory processing module re-queries the candidate inventory set. If other inventory units that meet the delivery time window exist, the current version number of the inventory unit is read and a new inventory reservation voucher is generated. If no such unit exists, the corresponding order platform is not found to have sufficient inventory. When a successfully reserved order is cancelled, the reserved quantity is released and the inventory version number is incremented again. When a reservation voucher expires but the platform payment status cannot be determined, the corresponding quantity is converted to isolated inventory. Once the status is clear, it is converted to officially occupied inventory or available inventory.
[0044] For heavy goods orders, the inventory processing module also performs item-by-item matching based on order fulfillment characteristics and delivery vehicle capabilities. The matching result between order i and vehicle j is represented as follows:
[0045] In the formula, K W,ij K V,ij K L,ij and K E,ij These represent the adaptation values for load, loading space, loading and unloading auxiliary equipment, and vehicle type passage conditions, respectively. A value of 1 is assigned if the condition is met, and 0 if it is not. Only when all four values are 1 is K... ij Only 1.
[0046] Warehouse node C contains three items Z, each weighing 450 kg and having a volume of 1.8 m³. 3 It also requires delivery vehicles to be equipped with hydraulic tailgates. The remaining load is 800 kg, and the remaining loading space is 3.5 m. 3 However, it is not equipped with a hydraulic tailgate, therefore the auxiliary equipment is compatible with K. L,ij =0; Vehicle It is equipped with a hydraulic tailgate, but the remaining load is only 300 kg, therefore the load compatibility designation is K. W,ij =0, the overall compatibility rating of both vehicles is 0, therefore the three items Z will not be included in the immediately fulfillable inventory.
[0047] When the vehicle The vehicle is within the service area and is equipped with a hydraulic tailgate, has a remaining load capacity of 1200 kg, and a remaining loading space of 4.5 m². 3 And when the end-point passage conditions are met, its integrated adaptation signage =1. The times for vehicle arrival at the warehouse, picking and dispatching, loading, delivery, home delivery, and buffering are 15 min, 30 min, 20 min, 70 min, 25 min, and 10 min respectively. Therefore: ; 170 min less The stipulated 180-minute fulfillment timeframe, combined with three available inventory items, vehicle load capacity, and loading space, The corresponding release inventory is: ; Therefore, the inventory processing module is adapted to the instant retail platform. Two items Z are released, when the vehicle When a user leaves the service area, the remaining load is occupied by other tasks, the hydraulic tailgate is in an abnormal state, or the operating load of the storage node changes, the above adaptation and quantity calculation will be re-executed, and the inventory published on the platform will be updated.
[0048] Step S14: Determine the target inventory unit, target warehousing node, and target delivery resource from the candidate inventory set, and generate the order fulfillment link.
[0049] The warehousing and distribution matching module extracts commodity weight, commodity volume, vehicle type access conditions, and loading / unloading auxiliary requirements from order fulfillment characteristics to form a fulfillment constraint set; it extracts vehicle rated load, remaining load, remaining loading space, existing delivery tasks, vehicle type information, and loading / unloading auxiliary equipment status from the delivery management system to form a delivery capacity set. The two sets are compared item by item, and the appropriate data is retained. Vehicles with a value of 1 form a candidate delivery resource set.
[0050] The warehouse matching module combines candidate inventory units s with suitable vehicles j, and continues to verify the vehicle's estimated arrival time, the estimated outbound time from the warehouse node, and the estimated delivery time. A candidate warehouse matching set is formed by combining inventory quantity, vehicle suitability, and delivery time window. ; In the formula, B isj The candidate warehouse-distribution combination consists of order i, inventory unit s, and vehicle j. If no candidate warehouse-distribution combination exists, the system does not perform inventory pre-holding, but instead returns the current unfulfillable status to the order platform or enters the delayed recalculation queue.
[0051] When there is more than one candidate warehousing and distribution combination, the warehousing and distribution matching module will determine the fulfillment time based on the estimated delivery time. The distance D(s) from the storage node to the receiving location. ) and the warehouse node operation load rate ρs(t) are used to calculate the combination parameters, and the target warehouse distribution combination with the smallest parameter value is selected: ; In the formula, λ1, λ2 and λ3 are preset weights, and include proportional coefficients that convert fulfillment time, driving distance and load utilization into dimensionless normalized values; after the system selects the target combination, it determines the inventory unit, storage node and delivery resource in it as the target inventory unit, target storage node and target delivery resource respectively; then it performs version verification and inventory pre-occupancy according to step S13; if the pre-occupancy fails, the corresponding inventory unit is deleted from the candidate warehousing and distribution combination set, and the remaining combinations are reordered.
[0052] After successful inventory pre-positioning, the warehousing and distribution matching module generates the order fulfillment chain according to the resource execution order: ; In the formula, Pre-positioning nodes for inventory For warehousing operation nodes, For loading nodes, For delivery nodes, This is the home delivery milestone.
[0053] Each fulfillment node records the node identifier, preceding nodes, node status, bound resources, planned completion time, actual completion time, and estimated processing time. Inventory pre-positioning nodes are bound to the target inventory unit and pre-positioning voucher; warehousing operation nodes are bound to the target warehousing node; loading nodes are bound to the loading station and loading / unloading auxiliary equipment; delivery nodes are bound to the target delivery vehicle; and home delivery nodes are bound to the receiving location and service requirements. Once the previous node is completed, the system writes the completion event and actual completion time into the order fulfillment chain and triggers the next node to enter the executable state.
[0054] Step S15: Calculate the remaining fulfillment time based on the actual processing status of the order fulfillment link, and generate a corrected order fulfillment link when the expected timeout occurs.
[0055] The fulfillment correction module identifies completed nodes, in-process nodes, and unstarted nodes according to their status. For completed nodes, the actual completion time is read and their processing time is no longer included in the remaining fulfillment time. For in-process nodes and unstarted nodes, the estimated processing time is updated based on the warehouse operation queue, vehicle arrival status, loading station status, road status, and home delivery service conditions.
[0056] Let U iLet (t) be the set of unfinished nodes for order i at time t, where t̂ j (t) represents the incomplete node N. j Given the estimated processing time at time t, the remaining fulfillment time is: ; This formula only sums the estimated processing time for incomplete nodes. Assuming the inventory pre-holding node is complete and the remaining order delivery time is 120 minutes; the warehouse picking, loading, delivery, home delivery, and buffer times are 25 minutes, 15 minutes, 55 minutes, 15 minutes, and 5 minutes respectively. Then: ; Since 115 min ≤ 120 min, the fulfillment correction module determines that the original order fulfillment chain can meet the delivery time window, and each node continues to execute according to the original bound resources.
[0057] If, during order fulfillment, the workload at the warehouse node increases, causing the estimated processing time for warehouse picking to increase from 25 minutes to 50 minutes, then the updated remaining fulfillment time will be: ; The fulfillment correction module compares the updated remaining fulfillment time with the current remaining fulfillment time of the order, Tirem(t) = 120 min, to obtain the estimated timeout duration: ; The fulfillment correction module identifies warehousing operation nodes whose processing time changes from 25 min to 50 min as delayed fulfillment nodes and performs delay propagation judgment along the order fulfillment chain. If the original delivery vehicle cannot wait for the delayed goods, the loading node and delivery node will be affected. Since the inventory pre-holding has been completed and the target inventory unit is still valid, the inventory pre-holding node is located outside the affected fulfillment node range.
[0058] After determining the affected fulfillment node range, the system first keeps the target inventory unit and target warehouse node unchanged, and only queries for alternative vehicles that meet the order weight, volume, loading and unloading assistance requirements, and remaining delivery time. If the alternative vehicle arrives at the warehouse node 5 minutes after picking, and loading, delivery, and home delivery take 10 minutes, 40 minutes, and 10 minutes respectively, then the remaining fulfillment time after replacing only the delivery resources is: ; Since 115 min ≤ 120 min, this solution meets the delivery time window. The system retains the inventory pre-occupancy node, target inventory unit and target warehouse node, and only replaces the original delivery vehicle in the affected fulfillment node range with the new vehicle, and updates the estimated processing time of the loading node and delivery node accordingly.
[0059] If no suitable vehicle can arrive in time at the original warehouse node, the fulfillment correction module will query other candidate warehouse nodes. Assuming that re-reservation, picking and outbound processing, loading, delivery, and home delivery at the new warehouse node take 5 minutes, 15 minutes, 10 minutes, 35 minutes, and 10 minutes respectively, the remaining fulfillment time after replacing the warehouse node is: ; When the scheme is selected, the system first generates a new inventory reservation voucher based on the current inventory version number of the new warehouse node; after the new inventory reservation is successful, the reservation of the original target inventory unit is released to avoid the failure of the new inventory reservation after the original inventory is released first. Then, the new warehouse node, loading station and delivery vehicle are written into the corrected order fulfillment link.
[0060] When both the vehicle replacement plan and the warehouse node replacement plan meet the delivery time window, the fulfillment correction module calculates the correction cost based on the corrected remaining fulfillment time, the number of fulfillment nodes that need to be replaced, and the number of resource switching times: ; In the formula, T m Let N be the remaining performance time after correction for scheme m. m C represents the number of fulfillment nodes that need to be replaced. m The resource switching count is denoted by α, β, and γ, which are the corresponding weights and include normalization coefficients to convert each evaluation metric to a uniform dimensionless scale. Taking α=0.5, β=20, and γ=10, the correction cost for replacing only the vehicle is: ; The calibration cost for replacing storage nodes, loading nodes, and delivery nodes is: ; The feasible solutions that can meet the delivery time window are combined into a set Ω, and the final solution is selected as follows: ; Since J1 < J2, the system selects the option of replacing only the vehicle. Therefore, when there are multiple feasible correction options, the system does not simply choose the option with the shortest expected delivery time, but takes into account the number of replacement nodes and the number of resource switching while meeting the delivery time window, so as to reduce changes to the original order fulfillment chain.
[0061] The fulfillment correction module writes the resource identifier, estimated processing time, and node connection relationship in the selected solution into the order fulfillment link, and retains the node identifier, completion status, actual completion time, and resource binding relationship of nodes outside the affected fulfillment node range. After the correction is completed, the order status is synchronized to the corresponding order platform, warehouse management system, and delivery management system. Thus, steps S11 to S15 form a continuous processing closed loop of order access, order unification, dynamic inventory release, concurrent inventory pre-positioning, warehouse and distribution resource matching, and partial fulfillment correction.
[0062] It should be noted that the above number of platforms, number of warehouse nodes, vehicle types and values are used to illustrate the data transmission, judgment and update relationship between each step. The interface form, processing cycle, weight and duration parameters can be set according to the actual business scale, but do not change the processing relationship between multi-platform order unification, dynamic fulfillable inventory, inventory version pre-occupancy, warehouse and distribution adaptation and local correction of affected fulfillment node intervals.
[0063] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the invention by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the invention should be included within the scope of protection of the invention.
Claims
1. A multi-platform order aggregation processing system, characterized in that, The system includes: The module includes an order access module, an order unification module, an inventory processing module, a warehouse and distribution matching module, and a fulfillment correction module. The order access module receives order data output from multiple order platforms and adds source identifiers to generate a cloud platform order dataset; The order unification module performs field mapping on the cloud platform order dataset to generate a unified order object, and generates order fulfillment features based on product attributes, delivery time windows, and last-mile delivery conditions. The inventory processing module filters out a set of candidate inventory based on the warehousing node operation time, delivery resources, and order fulfillment characteristics. The warehousing and distribution matching module determines the target inventory unit, target warehousing node, and target delivery resource from the candidate warehousing and distribution combinations and generates the order fulfillment link. The fulfillment correction module calculates the remaining fulfillment time based on the actual processing status of the order fulfillment link. When the remaining fulfillment time does not meet the delivery time window, it determines the affected fulfillment node range and performs resource replacement to generate a corrected order fulfillment link.
2. The multi-platform order aggregation processing system according to claim 1, characterized in that, The steps to generate a cloud platform order dataset are as follows: Establish data access channels corresponding to traditional e-commerce platforms, instant retail platforms, supply chain management systems, warehousing and distribution management systems, and delivery management systems; Receive order data through various data access channels, add platform source identifiers, and extract platform order identifiers and data reception times; The order data is arranged according to the data reception time to generate a cloud platform order dataset.
3. The multi-platform order aggregation processing system according to claim 1, characterized in that, The steps to generate a unified order object are as follows: Identify product identifiers, product quantities, delivery locations, order status, required delivery time, and service requirements from the cloud platform order dataset; Based on the field mapping relationship corresponding to each order platform, the identified fields are name-converted to generate a standard order field set; A unified order object is generated by associating the platform order identifier with the standard order field.
4. The multi-platform order aggregation processing system according to claim 1, characterized in that, The steps to generate order fulfillment characteristics are as follows: Extract the product weight, product volume, and required delivery time from the unified order object; Determine vehicle access conditions and loading / unloading assistance requirements based on the receiving location and service requirements, and generate final delivery conditions; Combining product weight, product volume, required delivery time, and last-mile delivery conditions into order fulfillment characteristics, and matching them with the operational capacity of warehousing nodes and the transportation capacity of distribution resources, respectively.
5. The multi-platform order aggregation processing system according to claim 1, characterized in that, The steps to generate a candidate inventory set are as follows: Determine the available inventory level based on physical inventory, locked inventory, and isolated inventory; The estimated outbound time is obtained based on the operational status of the warehousing node where each inventory unit is located, and the estimated delivery time is obtained based on the delivery relationship between the warehousing node and the receiving location. The estimated outbound time and estimated delivery time are combined to obtain the fulfillable completion time. Inventory units with fulfillable completion times within the delivery time window and available inventory greater than zero are grouped into a candidate inventory set.
6. The multi-platform order aggregation processing system according to claim 5, characterized in that, The steps to determine the platform's published inventory are as follows: Obtain the location of candidate inventory units, available inventory, warehouse node workload, and remaining capacity of available delivery resources to generate warehouse and distribution status data; Based on the warehouse and distribution status data and the delivery time windows corresponding to each order platform, inventory quantity constraints are applied to determine the platform-released inventory quantity for each order platform. When the status of warehousing or distribution resources changes, the inventory published on the platform will be recalculated and updated.
7. The multi-platform order aggregation processing system according to claim 1, characterized in that, The steps for performing inventory pre-positioning on the target inventory unit are as follows: Extract the current available inventory quantity and current inventory version number of the target inventory unit, and generate an inventory reservation voucher based on the order identifier, target inventory unit identifier, reserved quantity, current inventory version number, generation time and validity period; The inventory version number in the inventory reservation document is compared with the current inventory version number of the target inventory unit to generate a version comparison result. Inventory pre-allocation is performed when the version number is consistent and the current inventory quantity is not less than the pre-allocated quantity, and the inventory version number is updated.
8. The multi-platform order aggregation processing system according to claim 1, characterized in that, The steps to determine the target inventory unit, target warehousing node, and target distribution resources are as follows: Based on the order fulfillment characteristics, a set of fulfillment constraints is formed, including the weight and volume of the goods, vehicle type and traffic conditions, and loading and unloading assistance requirements. A delivery capacity set is formed based on the rated load, remaining load, remaining loading space, existing delivery tasks, vehicle model information, and the status of loading and unloading auxiliary equipment of the delivery vehicles; The set of performance constraints and the set of delivery capabilities are matched one by one to generate a set of candidate delivery resources, and candidate warehousing and distribution combinations are generated by combining the set of candidate inventory and the delivery time window. From the candidate warehousing and distribution combinations, identify the target inventory units, target warehousing nodes, and target delivery resources that meet the delivery time window.
9. The multi-platform order aggregation processing system according to claim 1, characterized in that, The steps for calculating the remaining performance time are as follows: Identify inventory pre-positioning nodes, warehousing operation nodes, loading nodes, delivery nodes, and home delivery nodes in the order fulfillment chain; Obtain the planned completion time and actual completion time of each fulfillment node, identify unfulfilled fulfillment nodes based on their execution status, and obtain the corresponding estimated processing time. Each incomplete fulfillment node and its estimated processing time are combined to form an incomplete node duration set. The estimated processing times are then merged to obtain the remaining fulfillment time for the order fulfillment chain.
10. A multi-platform order aggregation processing system according to claim 9, characterized in that, The steps to generate a corrected order fulfillment chain are as follows: Compare the remaining performance time with the delivery time window to generate a performance timeout determination result, and locate the delayed performance node whose processing time has changed when the determination result is a timeout. Based on the time impact of delayed performance nodes on subsequent performance nodes, a delay propagation determination is performed to identify the range of affected performance nodes; Retain fulfillment nodes outside the affected fulfillment node range, and replace warehousing nodes, delivery resources, or loading and unloading auxiliary equipment within the range to generate a corrected order fulfillment chain.
