Fastener in-bin sorting and transferring integrated method based on multi-source receipt driving

Abstract

The invention discloses a multi-source receipt driven in-bin sorting and goods transfer integrated method for fasteners, which comprises the following steps: S1, performing unified priority calculation on order lines of multi-source receipts from a sales order, a purchase return order and an outsourcing material receiving order, and writing the order lines and scores into a Redis queue; s2, carrying out batch clustering on the sorted order lines according to SKU similarity, storage location distance and delivery route proximity to obtain a compliant batch set; step S3, on the basis of the compliance batch set, taking a comprehensive behavior cost function as a target, considering walking time, load balancing, time window penalty rules, equipment lane occupation, fatigue, speed attenuation, storage location staying, error reworking and consumables, and performing human-area-batch optimal allocation; and S4, performing dynamic adjustment and strategic triggering on the net income by taking'gap driving 'as a triggering mechanism. According to the invention, the efficiency and accuracy of sorting and cargo transfer are improved.

Description

Technical Field

[0001] This invention relates to an integrated method for sorting and transferring fasteners within a warehouse based on multi-source document-driven processes. Background Technology

[0002] Current warehouse sorting systems are mostly driven by single-source documents, making it difficult to balance cross-source priority, regional congestion, and restocking benefits. Furthermore, they lack sufficient support for special rules regarding outsourced material requisition, such as completeness, fragility, perishability, and excessive weight. The lack of unified, calculable metrics and algorithmic frameworks makes it difficult to achieve both efficiency and accuracy.

[0003] Therefore, a method for integrating fastener warehouse sorting and relocation based on multi-source document-driven technology is provided. Summary of the Invention

[0004] To address the aforementioned problems in the existing technology, this invention provides an integrated method for sorting and transferring fasteners within a warehouse based on multi-source document-driven processes, which improves the efficiency and accuracy of sorting and transferring.

[0005] The technical solution to achieve the above objectives is: A method for integrated sorting and relocation of fasteners in a warehouse based on multi-source document-driven processes includes: Step S1: Perform a unified priority calculation on the order lines from multiple source documents, including sales orders, purchase return orders, and outsourced material requisition orders, and write the order lines and scores to a Redis queue (a queue used to store messages, which is an asynchronous communication mechanism for transmitting messages between different application services). Step S2: Cluster the sorted order lines by SKU (inventory management coding method) similarity, warehouse location distance, and delivery route proximity to obtain a set of compliant batches; Step S3: Based on the set of compliant batches, with the comprehensive behavioral cost function as the objective, and taking into account factors such as walking time, load balancing, time window penalty rules, equipment occupancy, fatigue, speed decay, warehouse dwell time, error rework and consumables, perform optimal allocation of people-area-batch. Step S4: Using "gap-driven" as the triggering mechanism, the net return is dynamically adjusted and strategically triggered. Step S5: Receive batch / sorting completion events and atomic states, calculate and organize priority and benefit categories, clustering and route categories, allocation and path categories, timeliness and quality categories, resource and concurrency categories, and economic and customer categories, and output detailed and summary tables for each type of indicator.

[0006] Preferably, in step S1, priority scoring... The formula is as follows: ; In the formula, Corresponding to urgency scores Distance score Out-of-stock risk Order value / gross profit normalization Consolidated revenue sharing Constraint Costs Manually pinned logo Replenishment trigger priority distribution Customer weight Warehouse location-picker fit Equipment occupancy and channel congestion penalties Customer revenue weighting Weighting based on proximity to the shipping location or the order of the route. , Rating factors The corresponding weights; in, Urgency score The calculation formula is as follows: ; In the formula, The current system time. For the document creation time, For order or task deadlines; Distance score The calculation formula is as follows: ; In the formula, The coordinates of the current order line's location in the warehouse. The coordinates of the picking start point or batch center. This represents the maximum reference distance within the warehouse; the closer the distance is to the picking center, the higher the distance score. The higher; Out-of-stock risk The calculation formula is as follows: ; In the formula, This represents the available inventory quantity. For the required quantity; Consolidated Profit Sharing The calculation formula is as follows: ; In the formula, Represents the Jakarta similarity coefficient. For order line SKU set, This is a batch KSU set.

[0007] Preferably, step S1 includes: Step S11: Retrieve order line data from multiple source documents and standardize it into structured objects; Step S12: Obtain inventory snapshot, storage location coordinates, picker location, and area parameters from the database or cache; Step S13: Calculate each scoring component, including the urgency score. Distance score Out-of-stock risk Order value / gross profit normalization Consolidated revenue sharing Constraint Costs Manually pinned logo Replenishment trigger priority distribution Customer weight Warehouse location-picker fit Equipment occupancy and channel congestion penalties Customer revenue weighting Weighting based on proximity to the shipping location or the order of the route. ; Step S14: Perform min-max normalization on each rating component and apply the rating factor. Corresponding weights Calculate the overall priority ; Step S15, based on overall priority The order line and score are written to the Redis queue as the sorting criteria.

[0008] Preferably, in step S2, an order line is set. , The attributes are respectively and The center coordinates of the storage location are Then the similarity Defined as: ; ; ; ; In the formula, All of them are configurable weights. This represents the maximum distance within the reservoir area. It is a semi-sine distance function used for geographic calculations. The maximum geographical distance of the delivery route. This is the overlap length between the two delivery routes. This is a geometric distance function for the warehouse location / batch center, defaulting to Euclidean distance. When channel constraints need to be considered, it can be replaced with the shortest reachable path distance within the warehouse. For geographical proximity, For route similarity, For orders , The destination's latitude and longitude coordinates This is a broken-line sequence of delivery routes from the warehouse to their respective destinations. To take the smaller of the two, This represents the total length of the broken-line sequence of delivery routes from the warehouse to their respective destinations; Clustering algorithms use similarity thresholds Guidelines: like Order line , Merged into the same batch; In special circumstances: like Then they will be forcibly merged into a "same route wave" batch, among which, Force a merging threshold based on geographical / route similarity; If the capacity constraint is violated after a merge, the batch will be reassigned to an adjacent similar batch. Capacity constraints are defined as: ; In the formula, For order line Effective picking volume, This refers to the maximum load capacity of the equipment or picking cart.

[0009] Preferably, step S2 includes: Step S21, according to comprehensive priority Read the first N order lines from the Redis priority queue in descending order of value; Step S22: Calculate the SKU similarity, warehouse location distance, and delivery route proximity of any two order lines to obtain the comprehensive similarity. ; Step S23, according to the similarity threshold Clustering is performed based on criteria; Step S24: Check batch capacity constraints; if the constraints are exceeded, perform a split. Step S25: Output the clustering results, the resulting set of compliant batches, and write them to the Redis priority queue.

[0010] Preferably, in step S3, the cost function Defined as: ; In the formula, For picking staff, For batches, Walking time corrected for behavior / channel effects. This is a load balancing component used to suppress individual loads. For time window penalty rules, batch If the promised timeframe is approached or exceeded, additional costs will be incurred. Due to equipment requirements and the cost of occupying passageways, As a form of punishment for fatigue, Penalty for speed Cost of storage space dwell time, For the cost of rework due to errors, For consumable costs, With non-negative weights, the system can be configured hierarchically by class / region / person or for online learning; in, ; In the formula, For batch Job path In the middle, two adjacent nodes in the execution order and The resulting set of directed arcs For nodes and The walking distance between them is calculated by default based on the shortest reachable path length on the internal tunnel topology. If there is no tunnel topology, it can degenerate into Euclidean distance. For sorting workers Movement speed; set up Indicates "batch" Should it be assigned to a picker? Then the objective function minimizes the total cost as follows: .

[0011] Preferably, step S3 includes: Step S31: Read the "compliant batch set" and the "candidate picker set", and retrieve the pickers. Qualifications, regional accessibility, equipment availability, shift schedules; Step S32: Calculate the costs of travel time, load balancing, time window penalty rules, equipment road occupancy, fatigue, speed decay, warehouse dwell time, error rework, and consumable costs. Step S33: Normalize the components and assign non-negative weights Weighting yields the total cost matrix. ; Step S34: If the scale is moderate, execute the Hungarian algorithm to quickly obtain the optimal or near-optimal solution, produce the assignment results and store them in the database / cached, and perform optimal allocation of people-area-batch. If there is equipment occupancy or time window conflict, recalculate the optimal or near-optimal solution.

[0012] Preferably, in step S4, the gap-driven triggering mechanism is as follows: when the available quantity in this warehouse is insufficient to meet the demand, candidate source warehouses are enumerated, and the net revenue from the transfer is calculated by comprehensively considering the transfer time, transfer cost and risk cost. The optimal source warehouse and the recommended transfer quantity are selected. At the same time, historical shortage intensity, no transfer time and predicted shortage risk enhancement factors are introduced to dynamically adjust and strategically trigger the net revenue. The gap is defined as: ; In the formula, The number of gaps, For the required quantity, This is the local available amount; Construct a set of candidate source repositories ,source transit time Transportation costs Single row price / gross profit Then the net income from baseline inventory adjustments : ; In the formula, For source Local availability The time cost coefficient, This refers to the risk cost coefficient. The optimal source is: ; like If so, it is recommended to adjust the quantity of goods. ; in, Net revenue from the optimal source baseline for goods restocking. Optimal source Local availability; Shortage Intensity Index : ; In the formula, Weighted moving average, The upper bound of the reference strength used for normalization is taken as the historical quantile or the business upper limit, ensuring... ; No-transfer-time index : ; In the formula, The current system time. for, The preset maximum allowable transfer interval threshold, The maximum observation period for normalization is used to limit , For identifying areas / warehouses / warehouse zones; Predicting the risk of stockouts : ; In the formula, Based on the recent daily average shortage / daily demand EWMA estimate, For the prospective period; Adjusted net income : ; In the formula, , and All are weighting coefficients. For holding costs, Costs incurred due to stockouts; When the shortage intensity index A high index increases revenue from inventory transfers and encourages replenishment; a low index indicates no inventory transfers. High inventory levels, combined with high holding costs, can thus curb excessive inventory transfers. like Then calculate the safety stock replenishment: ; And the recommended adjustment quantity Adjusted to: ; In the formula, For the service level factor, The standard deviation of demand; like And demand intensity Then restrictions and increase the trigger threshold. ; in, The threshold for stockout intensity. For items without a transfer time threshold, Demand intensity threshold This represents the minimum executable revenue threshold when NT is high.

[0013] Preferably, step S4 includes: Step S41: Calculate the gap for each order line and trigger the restocking process; Step S42, construct the candidate source set The net revenue from cargo relocation is calculated by considering transit time, transit costs, and risk costs. ; Step S43, introduce the shortage intensity index No stock transfer time index With the prediction of shortage risk Enhancement factors were used to calculate the adjusted net income. ; Step S44, with adjusted net income Choose the optimal source based on the maximum value. ; Step S45, calculate the recommended transfer quantity ; Step S46: Check shift / time window, vehicle / wave capacity, cross-regional permissions, and SLA (Service Level Agreement) impact. If not met, revert to the suboptimal source. Or cut ; Step S47, finally output the optimal net profit from inventory transfer. Adjusted net income Optimal source and suggested inventory transfer quantity This enables dynamic adjustment and strategic triggering of net income; In step S4, for the material completeness requirements of the outsourced work order, the completeness rate of each material and the overall completeness of the work order are calculated first. When the overall completeness is not complete, partial picking is allowed above the set threshold. The insufficient part enters the gap pool to trigger the inventory transfer revenue optimization process of steps S41-S47. Specifically, it includes: Calculate the kitting rate for each material. : ; In the formula, For materials in the work order The available quantity for distribution in this warehouse / designated storage area, after deducting the occupied quantity and calculating the net available quantity based on packaging unpacking / equivalent conversion. For materials in the work order The required quantity is calculated and measured using the smallest unit of measurement. Calculate overall homogeneity : ; In the formula, For the material index set of this work order Shangge Take the minimum value; Determine the overall fit ,like =1, mark "Complete Set Issued", proceed to outbound / removal from shelves; like If so, a "partial picking" task is generated, and the items are unpacked and reassembled according to the packaging rules; Calculate the gap: ; Will( , Write the data into the gap pool, triggering the inventory transfer revenue optimization process in steps S41-S47; like They go directly into the gap pool without undergoing partial picking; in, Sensitivity.

[0014] Preferably, in step S5, priority and benefit indicators are obtained from steps S1 and S4, clustering and route indicators are obtained from step S2, and allocation and path indicators are obtained from step S3. The allocation and path indicators also include unit walking distance. and unit work time ; Timeliness and quality metrics include: On-time achievement rate SLA Default Rate Picking accuracy Personnel error rate , number of reworked pieces and rework costs , in, unit walking distance The calculation formula is as follows: ; In the formula, The total walking distance for this window. The cumulative number of items picked at this window; Unit operation time The calculation formula is as follows: ; In the formula, The effective working time is defined as "picking + handling + necessary stops". The cumulative number of items picked at this window; On-time achievement rate The calculation formula is as follows: ; In the formula, The percentage of orders that were shipped or handed over within the promised timeframe; Picking accuracy The calculation formula is as follows: ; In the formula, To ensure the correct number of order lines is selected during the picking operation, This represents the total number of order rows picked within the same time period; Resource and concurrency metrics include: resource utilization rate Queue waiting delay Number of lock retries And idempotent number of times , in, resource utilization rate The calculation formula is as follows: ; In the formula, This refers to the cumulative time that personnel / areas are in a "work activity" state within a shift. The total available time for each shift of the corresponding statistical object; Economic and customer-related metrics include: gross profit per unit time. Customer weighting regional / shift fairness , in, Gross profit per unit time The calculation formula is as follows: ; In the formula, This represents the sum of the actual gross profit for all orders / order lines that have been shipped and confirmed within the window. This represents the effective work duration of the corresponding statistical object within the window.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention improves sorting and repositioning efficiency and accuracy by reading multi-source documents → standardizing them into "order lines", calculating priorities, clustering them into batches according to SKU / location similarity, allocating pickers and planning routes based on regional load and walking costs, optimizing repositioning revenue triggered by gaps, maintaining queues, locks and atomic states with Redis throughout the process, and outputting verifiable indicators including unit walking cost, net repositioning revenue, and resource utilization. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of an integrated method for sorting and transferring fasteners in a warehouse based on multi-source document-driven technology, according to the present invention. Figure 2 This is a flowchart illustrating the process of performing unified priority calculation on order lines from multiple sources such as sales orders, purchase return orders, and outsourced material requisition orders, and writing the order lines and scores into a Redis queue. Figure 3 This is a flowchart illustrating the process of clustering sorted order lines by SKU similarity, warehouse location distance, and delivery route proximity to obtain a set of compliant batches in this invention. Figure 4 This is a flowchart illustrating the optimal allocation of people, areas, and batches based on a set of compliant batches in this invention. Figure 5 This is a flowchart illustrating the specific process of dynamically adjusting and strategically triggering net income using a "gap-driven" mechanism in this invention. Detailed Implementation

[0017] 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, and 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.

[0018] like Figure 1 As shown, a method for integrated sorting and relocation of fasteners in a warehouse based on multi-source document-driven processes includes: Step S1: Perform a unified priority calculation on the order lines from multiple source documents, including sales orders, purchase return orders, and outsourced material requisition orders, and write the order lines and scores to the Redis queue.

[0019] In the embodiment, priority scoring The formula is as follows: ; In the formula, Corresponding to urgency scores Distance score Out-of-stock risk Order value / gross profit normalization Consolidated revenue sharing Constraint Costs Manually pinned logo Replenishment trigger priority distribution Customer weight Warehouse location-picker fit Equipment occupancy and channel congestion penalties Customer revenue weighting Weighting based on proximity to the shipping location or the order of the route. , Rating factors The corresponding weights; in, Urgency score The calculation formula is as follows: ; In the formula, The current system time. For the document creation time, For order or task deadlines; Distance score The calculation formula is as follows: ; In the formula, The coordinates of the current order line's location in the warehouse. The coordinates of the picking start point or batch center. This represents the maximum reference distance within the warehouse; the closer the distance is to the picking center, the higher the distance score. The higher; Out-of-stock risk The calculation formula is as follows: ; In the formula, This represents the available inventory quantity. For the required quantity; Consolidated Profit Sharing The calculation formula is as follows: ; In the formula, Represents the Jakarta similarity coefficient. For order line SKU set, This is a batch KSU set.

[0020] like Figure 2 As shown, step S1 includes: Step S11: Retrieve order line data from multiple source documents and standardize it into structured objects; Step S12: Obtain inventory snapshot, storage location coordinates, picker location, and area parameters from the database or cache; Step S13: Calculate each scoring component, including the urgency score. Distance score Out-of-stock risk Order value / gross profit normalization Consolidated revenue sharing Constraint Costs Manually pinned logo Replenishment trigger priority distribution Customer weight Warehouse location-picker fit Equipment occupancy and channel congestion penalties Customer revenue weighting Weighting based on proximity to the shipping location or the order of the route. ; Step S14: Perform min-max normalization on each rating component and apply the rating factor. Corresponding weights Calculate the overall priority ; Step S15, based on overall priority The order line and score are written to the Redis queue as the sorting criteria.

[0021] Step S2 involves clustering the sorted order lines into batches based on SKU similarity, warehouse location distance, and delivery route proximity. The resulting compliant batch set reduces unit travel costs and improves order consolidation efficiency by maximizing the overlap rate of picking paths and the co-occurrence of SKUs.

[0022] In the embodiment, an order line is set. , The attributes are respectively and The center coordinates of the storage location are Then the similarity Defined as: ; ; ; ; In the formula, All of them are configurable weights. This represents the maximum distance within the reservoir area. It is a semi-sine distance function used for geographic calculations. The maximum geographical distance of the delivery route. This is the overlap length between the two delivery routes. This is a geometric distance function for the warehouse location / batch center, defaulting to Euclidean distance. When channel constraints need to be considered, it can be replaced with the shortest reachable path distance within the warehouse. For geographical proximity, For route similarity, For orders , The destination's latitude and longitude coordinates This is a broken-line sequence of delivery routes from the warehouse to their respective destinations. To take the smaller of the two, This represents the total length of the broken-line sequence of delivery routes from the warehouse to their respective destinations; Clustering algorithms use similarity thresholds Guidelines: like Order line , Merged into the same batch; In special circumstances: like Then they will be forcibly merged into a "same route wave" batch, among which, Force a merging threshold based on geographical / route similarity; If the capacity constraint is violated after a merge, the batch will be reassigned to an adjacent similar batch. Capacity constraints are defined as: ; In the formula, For order line Effective picking quantity (calculated based on the system's smallest unit of measurement; optional calculation based on "equivalent load," i.e., converting the number of items into weight / volume before including them in the constraints). This refers to the maximum load capacity of the equipment or picking cart.

[0023] like Figure 3 As shown, step S2 includes: Step S21, according to comprehensive priority Read the first N order lines from the Redis priority queue in descending order of value; Step S22: Calculate the SKU similarity, warehouse location distance, and delivery route proximity of any two order lines to obtain the comprehensive similarity. ; Step S23, according to the similarity threshold Clustering is performed based on criteria; Step S24: Check batch capacity constraints; if the constraints are exceeded, perform a split. Step S25: Output the clustering results, the resulting set of compliant batches, and write them to the Redis priority queue.

[0024] Step S3: Based on the compliant batch set, with the comprehensive behavioral cost function as the objective, and taking into account walking time, load balancing, time window penalty rules, equipment occupation, fatigue, speed decay, warehouse dwell time, error rework and consumables, perform optimal allocation of people-area-batch.

[0025] In the embodiment, the cost function Defined as: ; In the formula, For picking staff, For batches, Walking time corrected for behavior / channel effects. This is a load balancing component used to suppress individual loads. For time window penalty rules, batch If the promised timeframe is approached or exceeded, additional costs will be incurred. Due to equipment requirements and the cost of occupying passageways, As a form of punishment for fatigue, Penalty for speed Cost of storage space dwell time, For the cost of rework due to errors, For consumable costs, With non-negative weights, the system can be configured hierarchically by class / region / person or for online learning; in, ; In the formula, For batch Job path In the middle, two adjacent nodes in the execution order and The resulting set of directed arcs For nodes and The walking distance between them is calculated by default based on the shortest reachable path length on the internal tunnel topology. If there is no tunnel topology, it can degenerate into Euclidean distance. For sorting workers Movement speed; set up Indicates "batch" Should it be assigned to a picker? Then the objective function minimizes the total cost as follows: .

[0026] like Figure 4 As shown, step S3 includes: Step S31: Read the "compliant batch set" and the "candidate picker set", and retrieve the pickers. Qualifications, regional accessibility, equipment availability, shift schedules; Step S32: Calculate the costs of travel time, load balancing, time window penalty rules, equipment road occupancy, fatigue, speed decay, warehouse dwell time, error rework, and consumable costs. Step S33: Normalize the components and assign non-negative weights Weighting yields the total cost matrix. ; Step S34: If the scale is moderate, execute the Hungarian algorithm to quickly obtain the optimal or near-optimal solution, produce the assignment results and store them in the database / cached, and perform optimal allocation of people-area-batch. If there is equipment occupancy or time window conflict, recalculate the optimal or near-optimal solution.

[0027] Step S4 uses the "gap-driven" mechanism to dynamically adjust and strategically trigger net returns.

[0028] In the embodiment, the gap-driven triggering mechanism is as follows: when the available quantity in the warehouse is insufficient to meet the demand, candidate source warehouses are enumerated, and the net revenue from the transfer is calculated by comprehensively considering the transfer time, transfer cost and risk cost. The optimal source warehouse and the recommended transfer quantity are selected. At the same time, historical shortage intensity, no transfer time and predicted shortage risk enhancement factors are introduced to dynamically adjust and strategically trigger the net revenue. The gap is defined as: ; In the formula, The number of gaps, For the required quantity, This is the local available amount; Construct a set of candidate source repositories ,source transit time Transportation costs Single row price / gross profit Then the net income from baseline inventory adjustments : ; In the formula, For source Local availability The time cost coefficient, This refers to the risk cost coefficient. The optimal source is: ; like If so, it is recommended to adjust the quantity of goods. ; in, Net revenue from the optimal source baseline for goods restocking. Optimal source Local availability; Shortage Intensity Index : ; In the formula, Weighted moving average, The upper bound of the reference intensity (events / day) used for normalization can be taken as the historical quantile (e.g., 95th percentile) or the business upper limit, to ensure... The more frequent the stockouts within a unit of time, the higher the stockout intensity index. The higher; No-transfer-time index : ; In the formula, The current system time. The most recent stock transfer completion time (timestamp) for the specified SKU in the specified region. The preset maximum allowable transfer interval threshold, The maximum observation period for normalization (e.g., 30 days / 720 hours) is used to limit , For regions / warehouses / areas (consistent with the granularity of transfer decisions, such as "East China Central Warehouse Area G"), if there has been no transfer for a long time, there is no transfer duration index. The higher; Predicting the risk of stockouts : ; In the formula, It is estimated using the EWMA (Exponentially weighted moving average of daily shortage / daily demand over a recent period, calculated from actual observations (or measurements) to estimate the current value). For the prospective period; Adjusted net income : ; In the formula, , and All are weighting coefficients (non-negative, learnable / configurable), measuring CS, NT, and The weight of the item's impact on net income For holding costs, Costs incurred due to stockouts; When the shortage intensity index A high index increases revenue from inventory transfers and encourages replenishment; a low index indicates no inventory transfers. High inventory levels, combined with high holding costs, can thus curb excessive inventory transfers. like Then calculate the safety stock replenishment: ; And the recommended adjustment quantity Adjusted to: ; In the formula, For the service level factor, The standard deviation of demand; like And demand intensity Then restrictions and increase the trigger threshold. ; in, The stockout severity threshold is set; once reached, safety stock replenishment is initiated. There is no threshold for the time required for stock transfer; once this threshold is reached, the SKU is determined to be likely to be "stable / slow-moving" in that region. This is the demand intensity threshold; when Demand is considered weak. EWMA (Early Daily Demand) or its standardized value can be used. This is the minimum executable revenue threshold (in RMB) when NT is high, used to raise the execution threshold and prevent slow-moving SKUs from being mis-adjusted.

[0029] like Figure 5 As shown, step S4 includes: Step S41: Calculate the gap for each order line and trigger the restocking process; Step S42, construct the candidate source set The net revenue from cargo relocation is calculated by considering transit time, transit costs, and risk costs. ; Step S43, introduce the shortage intensity index No stock transfer time index With the prediction of shortage risk Enhancement factors were used to calculate the adjusted net income. ; Step S44, with adjusted net income Choose the optimal source based on the maximum value. ; Step S45, calculate the recommended transfer quantity ; Step S46: Check shift / time window, vehicle / wave capacity, cross-regional permissions, and SLA impact. If not met, revert to the suboptimal source. Or cut ; Step S47, finally output the optimal net profit from inventory transfer. Adjusted net income Optimal source and suggested inventory transfer quantity This enables dynamic adjustment and strategic triggering of net income.

[0030] In this embodiment, for the material completeness requirements of outsourced work orders, the completeness rate of each material and the overall completeness of the work order are first calculated. When the overall completeness is not complete, partial picking is allowed above a set threshold. The insufficient part enters the gap pool to trigger the inventory transfer revenue optimization process of steps S41-S47. Specifically, it includes: Calculate the kitting rate for each material. : ; In the formula, For materials in the work order The available quantity (converted to the system's smallest unit of measurement) for distribution within this warehouse / designated storage area, after deducting occupied quantities (locked, pending shipment, frozen in transit, occupied by quality inspection, etc.) and net available quantity after packaging unpacking / equivalent conversion. For materials in the work order The required quantity (material issuance requirement) is calculated and measured using the smallest unit of measurement. Calculate overall homogeneity : ; In the formula, For the material index set of this work order Shangge Take the minimum value; Determine the overall fit ,like =1, mark "Complete Set Issued", proceed to outbound / removal from shelves; like If so, a "partial picking" task is generated, and the items are unpacked and reassembled according to the packaging rules; Calculate the gap: ; Will( , Write the data into the gap pool, triggering the inventory transfer revenue optimization process in steps S41-S47; like They go directly into the gap pool without undergoing partial picking; in, Sensitivity.

[0031] Step S5: Receive batch / sorting completion events and atomic states, calculate and organize priority and benefit categories, clustering and route categories, allocation and path categories, timeliness and quality categories, resource and concurrency categories, and economic and customer categories, and output detailed and summary tables for each type of indicator.

[0032] In this embodiment, priority and benefit metrics are obtained from steps S1 and S4, clustering and route metrics are obtained from step S2, and allocation and path metrics are obtained from step S3. The allocation and path metrics also include the unit walking distance. and unit work time ; Timeliness and quality metrics include: On-time achievement rate SLA Default Rate Picking accuracy Personnel error rate , number of reworked pieces and rework costs , in, unit walking distance The calculation formula is as follows: ; In the formula, The total walking distance for this window. The cumulative number of items picked at this window; Unit operation time The calculation formula is as follows: ; In the formula, The effective working time is defined as "picking + handling + necessary stops". The cumulative number of items picked at this window; On-time achievement rate The calculation formula is as follows: ; In the formula, The percentage of orders that were shipped or handed over within the promised timeframe; Picking accuracy The calculation formula is as follows: ; In the formula, To ensure the correct number of order lines is selected during the picking operation, This represents the total number of order rows picked within the same time period; Resource and concurrency metrics include: resource utilization rate Queue waiting delay Number of lock retries And idempotent number of times , in, resource utilization rate The calculation formula is as follows: ; In the formula, This refers to the cumulative time that personnel / areas are in a "work activity" state within a shift (effective operations such as picking, handling, necessary stops / checking and waiting; excluding standby, rest, training, and downtime). The total available time for the corresponding statistical object's shift (e.g., 8 hours; when counting across shifts, it is the sum of the available time for each shift). Economic and customer-related metrics include: gross profit per unit time. Customer weighting regional / shift fairness , in, Gross profit per unit time The calculation formula is as follows: ; In the formula, This is the sum of the actual gross profit corresponding to orders / order lines that have been completed and confirmed within the window (= sales revenue − purchasing / transfer costs − direct delivery / processing and other attributable expenses; returned / voided goods are excluded; the scope is consistent with the company's financial statistics). The effective work time of the corresponding statistical object within the window (and) The definition of "active" is consistent: picking + handling + necessary stop, excluding standby / rest).

[0033] Example 1: Implementation of Priority Scoring 1) Scene Three order lines (A, B, C) within the same region are initially weighted equally. 2) Components A: =0.9, =0.8, =0.1, =0.6, =0.5, =0.1, =0, =0, =0.7, =0.8, =0.2, =0.6, =0.7 B: =0.6, =0.9, =0.4, =0.3, =0.7, =0.0, =0, =0, =0.3, =0.7, =0.1, =0.3, =0.5 C: =0.7, =0.5, =0.6, =0.8, =0.1, =0.2, =1 (manually pinned), the rest are in the middle.

[0034] 3) Results Initial sort: C (because) =1) > A > B; Post-training (pairwise sorting + business constraints): Urgency score Weights of rating factors Weighting of the scoring factor based on proximity to the shipping location or route sequence Rising, consolidated profit sharing Weights of rating factors Slightly increased, A exceeds C (when the pinned post expires or decays), actual on-time achievement rate Increased by 2.3% per unit of working time It decreased by 1.4%.

[0035] Example 2: Clustering Parameter Optimization Baseline: =0.4, =0.4, =0.5, =0.6, =120kg; After optimization: =0.45, =0.35, =0.6, =0.58; Results: Path adhesion increased by 6.1%, SKU overlap increased by 4.7%, and unit walking distance increased. SLA does not increase.

[0036] Example 3: Assignment Cost Targeting Two pickers (New) and (Skilled), same batch B.

[0037] Observation: Sorting worker movement speed Sorting staff movement speed Sorting staff Storage space dwell cost Lower cost of rework due to errors Lower.

[0038] After learning: speed penalty Storage space dwell costs Error rework costs Non-negative weights Elevate; Solution: Cost of rework after deployment Sorting error rate -2.8%, +0.6.

[0039] Example 4: Revenue from Stock Transfers and Safety Stock parameter: =5 yuan / piece; Source : =20 yuan =60 points =50, =4.

[0040] Baseline: ≈5·4−(20+ ·60)− • Risk; among which, taking =0.05 yuan / minute, Risk = 2 yuan, that is ≈28; Enhancement: =0.7、 =0.1、 (1d)=0.25、 =3、 =1、 =1、 =0.3、 =2, thus obtaining ≈30+; Safety stock: =1.28、 =2、 =1 day, received = =3; suggestion: = =7; After launch, out-of-stock complaints decreased, and on-time delivery rate increased. rise.

[0041] Sensitivity and robustness analysis Sensitivity: From 0.8 to 0.9, the initial operating rate of the production line decreased by 1.2 percentage points, but the on-time achievement rate... +0.9pp; Recommended regional configuration within the 0.85–0.9 range; threshold :improve This can reduce slow-moving replenishment; if demand is strong At the same time, if it is too low, then... The impact is significant and positive.

[0042] Robustness: Stress replay is performed for scenarios with abnormally high congestion / device shortage to ensure... and Automatically revert to baseline when out of bounds occurs.

[0043] Finally, it should be noted that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for integrated sorting and transfer of fasteners in a warehouse based on multi-source document-driven processes, characterized in that, include: Step S1: Perform a unified priority calculation on the order lines from multiple source documents, including sales orders, purchase return orders, and outsourced material requisition orders, and write the order lines and scores to a Redis queue. Step S2: Cluster the sorted order lines by SKU similarity, warehouse location distance, and delivery route proximity to obtain a set of compliant batches; Step S3: Based on the set of compliant batches, with the comprehensive behavioral cost function as the objective, and taking into account factors such as walking time, load balancing, time window penalty rules, equipment occupancy, fatigue, speed decay, warehouse dwell time, error rework and consumables, perform optimal allocation of people-area-batch. Step S4: Using "gap-driven" as the triggering mechanism, the net return is dynamically adjusted and strategically triggered. Step S5: Receive batch / sorting completion events and atomic states, calculate and organize priority and benefit categories, clustering and route categories, allocation and path categories, timeliness and quality categories, resource and concurrency categories, and economic and customer categories, and output detailed and summary tables for each type of indicator.

2. The integrated method for fastener warehouse sorting and transfer based on multi-source document driving as described in claim 1, characterized in that, In step S1, priority scoring The formula is as follows: ； In the formula, Corresponding to urgency scores Distance score Out-of-stock risk Order value / gross profit normalization Consolidated revenue sharing Constraint Costs Manually pinned logo Replenishment trigger priority distribution Customer weight Warehouse location-picker fit Equipment occupancy and channel congestion penalties Customer revenue weighting Weighting based on proximity to the shipping location or the order of the route. , Rating factors The corresponding weights; in, Urgency score The calculation formula is as follows: ； In the formula, The current system time. For the document creation time, For order or task deadlines; Distance score The calculation formula is as follows: ； In the formula, The coordinates of the current order line's location in the warehouse. The coordinates of the picking start point or batch center. This represents the maximum reference distance within the warehouse; the closer the distance is to the picking center, the higher the distance score. The higher; Out-of-stock risk The calculation formula is as follows: ； In the formula, This represents the available inventory quantity. For the required quantity; Consolidated Profit Sharing The calculation formula is as follows: ； In the formula, Represents the Jakarta similarity coefficient. For order line SKU set, This is a batch KSU set.

3. The method for integrated sorting and transfer of fasteners in a warehouse based on multi-source document driving according to claim 2, characterized in that, Step S1 includes: Step S11: Retrieve order line data from multiple source documents and standardize it into structured objects; Step S12: Obtain inventory snapshot, storage location coordinates, picker location, and area parameters from the database or cache; Step S13: Calculate each scoring component, including the urgency score. Distance score Out-of-stock risk Order value / gross profit normalization Consolidated revenue sharing Constraint Costs Manually pinned logo Replenishment trigger priority distribution Customer weight Warehouse location-picker fit Equipment occupancy and channel congestion penalties Customer revenue weighting Weighting based on proximity to the shipping location or the order of the route. ; Step S14: Perform min-max normalization on each rating component and apply the rating factor. Corresponding weights Calculate the overall priority ; Step S15, based on overall priority The order line and score are written to the Redis queue as the sorting criteria.

4. The integrated method for fastener warehouse sorting and transfer based on multi-source document driving as described in claim 3, characterized in that, In step S2, let the order line be set. , The attributes are respectively and The center coordinates of the storage location are Then the similarity Defined as: ； ； ； ； In the formula, All of them are configurable weights. This represents the maximum distance within the reservoir area. It is a semi-sine distance function used for geographic calculations. The maximum geographical distance of the delivery route. This is the overlap length between the two delivery routes. This is a geometric distance function for the warehouse location / batch center, defaulting to Euclidean distance. When channel constraints need to be considered, it can be replaced with the shortest reachable path distance within the warehouse. For geographical proximity, For route similarity, For orders , The destination's latitude and longitude coordinates This is a broken-line sequence of delivery routes from the warehouse to their respective destinations. To take the smaller of the two, This represents the total length of the broken-line sequence of delivery routes from the warehouse to their respective destinations; Clustering algorithms use similarity thresholds Guidelines: like Order line , Merged into the same batch; In special circumstances: like Then they will be forcibly merged into a "same route wave" batch, among which, Force a merging threshold based on geographical / route similarity; If the capacity constraint is violated after a merge, the batch will be reallocated to an adjacent similar batch. Capacity constraints are defined as: ； In the formula, For order line Effective picking volume, This refers to the maximum load capacity of the equipment or picking cart.

5. The integrated method for fastener warehouse sorting and transfer based on multi-source document driving according to claim 4, characterized in that, Step S2 includes: Step S21, according to comprehensive priority Read the first N order lines from the Redis priority queue in descending order of value; Step S22: Calculate the SKU similarity, warehouse location distance, and delivery route proximity of any two order lines to obtain the comprehensive similarity. ; Step S23, according to the similarity threshold Clustering is performed based on criteria; Step S24: Check batch capacity constraints; if the constraints are exceeded, perform a split. Step S25: Output the clustering results, the resulting set of compliant batches, and write them to the Redis priority queue.

6. The method for integrated sorting and transfer of fasteners in a warehouse based on multi-source document driving according to claim 5, characterized in that, In step S3, the cost function Defined as: ； In the formula, For picking staff, For batches, Walking time corrected for behavior / channel effects. This is a load balancing component used to suppress individual loads. For time window penalty rules, batch If the promised timeframe is approached or exceeded, additional costs will be incurred. Due to equipment requirements and the cost of occupying passageways, As a form of punishment for fatigue, Penalty for speed Cost of storage space dwell time, For the cost of rework due to errors, For consumable costs, With non-negative weights, the system can be configured hierarchically by class / region / person or for online learning; in, ； In the formula, For batch Job path In the middle, two adjacent nodes in the execution order and The resulting set of directed arcs For nodes and The walking distance between them is calculated by default based on the shortest reachable path length on the internal tunnel topology. If there is no tunnel topology, it can degenerate into Euclidean distance. For sorting workers Movement speed; set up Indicates "batch" Should it be assigned to a picker? Then the objective function minimizes the total cost as follows: 。 7. The integrated method for fastener warehouse sorting and transfer based on multi-source document driving according to claim 6, characterized in that, Step S3 includes: Step S31: Read the "compliant batch set" and the "candidate picker set", and retrieve the pickers. Qualifications, regional accessibility, equipment availability, shift schedules; Step S32: Calculate the costs of travel time, load balancing, time window penalty rules, equipment road occupancy, fatigue, speed decay, warehouse dwell time, error rework, and consumable costs. Step S33: Normalize the components and assign non-negative weights Weighting yields the total cost matrix. ; Step S34: If the scale is moderate, execute the Hungarian algorithm to quickly obtain the optimal or near-optimal solution, produce the assignment results and store them in the database / cached, and perform optimal allocation of people-area-batch. If there is equipment occupancy or time window conflict, recalculate the optimal or near-optimal solution.

8. The method for integrated sorting and transfer of fasteners in a warehouse based on multi-source document driving according to claim 7, characterized in that, In step S4, the gap-driven triggering mechanism is as follows: when the available quantity in this warehouse is insufficient to meet the demand, candidate source warehouses are enumerated, and the net revenue from the transfer is calculated by comprehensively considering the transfer time, transfer cost and risk cost. The optimal source warehouse and the recommended transfer quantity are selected. At the same time, historical shortage intensity, no transfer time and predicted shortage risk enhancement factors are introduced to dynamically adjust and strategically trigger the net revenue. The gap is defined as: ； In the formula, The number of gaps, For the required quantity, This is the local available amount; Construct a set of candidate source repositories ,source transit time Transportation costs Single row price / gross profit Then the net income from baseline inventory adjustments : ； In the formula, For source Local availability The time cost coefficient, This refers to the risk cost coefficient. The optimal source is: ； like If so, it is recommended to adjust the quantity of goods. ; in, Net revenue from the optimal source baseline for goods restocking. Optimal source Local availability; Shortage Intensity Index : ； In the formula, Weighted moving average, The upper bound of the reference strength used for normalization is taken as the historical quantile or the business upper limit, ensuring... ; No-transfer-time index : ； In the formula, The current system time. This refers to the time when the most recent stock transfer was completed for a specified SKU in a specified region. The preset maximum allowable transfer interval threshold, The maximum observation period for normalization is used to limit , For identifying areas / warehouses / warehouse zones; Predicting the risk of stockouts : ； In the formula, Based on the recent daily average shortage / daily demand EWMA estimate, For the prospective period; Adjusted net income : ； In the formula, , and All are weighting coefficients. For holding costs, Costs incurred due to stockouts; When the shortage intensity index A high index increases revenue from inventory transfers and encourages replenishment; a low index indicates no inventory transfers. High inventory levels, combined with high holding costs, can thus curb excessive inventory transfers. like Then calculate the safety stock replenishment: ； And the recommended adjustment quantity Adjusted to: ； In the formula, For the service level factor, The standard deviation of demand; like And demand intensity Then restrictions and increase the trigger threshold. ; in, The threshold for stockout intensity. For items without a transfer time threshold, The threshold for demand intensity This represents the minimum executable revenue threshold when NT is high.

9. The integrated method for fastener warehouse sorting and transfer based on multi-source document driving according to claim 8, characterized in that, Step S4 includes: Step S41: Calculate the gap for each order line and trigger the restocking process; Step S42, construct the candidate source set The net revenue from cargo relocation is calculated by considering transit time, transit costs, and risk costs. ; Step S43, introduce the shortage intensity index No stock transfer time index With the prediction of shortage risk Enhancement factors were used to calculate the adjusted net income. ; Step S44, with adjusted net income Choose the best source based on the maximum value. ; Step S45, calculate the recommended transfer quantity ; Step S46: Check shift / time window, vehicle / wave capacity, cross-regional permissions, and SLA impact. If not met, revert to the suboptimal source. or cut ; Step S47, finally output the optimal net profit from inventory transfer. Adjusted net income Optimal source and suggested inventory transfer quantity This enables dynamic adjustment and strategic triggering of net income; In step S4, for the material matching requirements of the outsourced work order, the matching rate of each material and the overall matching degree of the work order are calculated first. When the overall matching is not complete, partial picking is allowed above the set threshold. The insufficient part enters the gap pool to trigger the goods transfer revenue optimization process of steps S41-S47. Specifically, it includes: Calculate the kitting rate for each material. : In the formula, For materials in the work order The available quantity for distribution in this warehouse / designated storage area, after deducting the occupied quantity and calculating the net available quantity based on packaging unpacking / equivalent conversion. For materials in the work order The required quantity is calculated and measured using the smallest unit of measurement. Calculate overall homogeneity : ； In the formula, For the material index set of this work order Shangge Take the minimum value; Determine the overall fit ,like =1, mark "Complete Set Issued", proceed to outbound / removal from shelves; like If so, a "partial picking" task is generated, and the items are unpacked and reassembled according to the packaging rules; Calculate the gap: ； Will( , Write the data into the gap pool, triggering the inventory transfer revenue optimization process in steps S41-S47; like They go directly into the gap pool without undergoing partial picking; in, Sensitivity.

10. The integrated method for fastener warehouse sorting and transfer based on multi-source document driving according to claim 9, characterized in that, In step S5, priority and benefit indicators are obtained from steps S1 and S4, clustering and route indicators are obtained from step S2, and allocation and path indicators are obtained from step S3. Allocation and path indicators also include unit walking distance. and unit work time ; Timeliness and quality metrics include: On-time achievement rate SLA Default Rate Picking accuracy Personnel error rate , number of reworked pieces and rework costs , in, unit walking distance The calculation formula is as follows: ； In the formula, The total walking distance for this window. This is the cumulative number of items picked at this window; Unit operation time The calculation formula is as follows: ； In the formula, The effective working time for "picking + handling + necessary stops" This is the cumulative number of items picked at this window; On-time achievement rate The calculation formula is as follows: ； In the formula, The percentage of orders that were shipped or handed over within the promised timeframe; Picking accuracy The calculation formula is as follows: ； In the formula, To ensure the exact number of order lines is correct during the picking operation, This represents the total number of order rows picked within the same time period; Resource and concurrency metrics include: resource utilization rate Queue waiting delay Number of lock retries And idempotent number of times , in, resource utilization rate The calculation formula is as follows: ； In the formula, This refers to the cumulative time that personnel / areas are in a "work activity" state within a shift. The total available time for each shift of the corresponding statistical object; Economic and customer-related metrics include: gross profit per unit time. Customer weighting regional / shift fairness , in, Gross profit per unit time The calculation formula is as follows: ； In the formula, This represents the sum of the actual gross profit for all orders / order lines that have been shipped and confirmed within the window. This represents the effective work duration of the corresponding statistical object within the window.

Images