Intelligent sorting equipment bag type intelligent arrangement method and system based on intelligent algorithm

Abstract

The invention discloses a sorting equipment package type intelligent arrangement method and system based on an intelligent algorithm, and relates to the technical field of express sorting, and the intelligent arrangement steps are as follows: S1, calculating an optimal package type, integrating multi-dimensional data to construct a two-dimensional traffic prediction model, and generating a package type traffic prediction result; establishing a digital twin model of the sorting equipment, and adaptively analyzing physical attributes and running states of the equipment; establishing a packet type priority system based on a service rule and equipment capability, and screening an optimal packet type set; and S2, distribution of the number of the bag type grids: sequentially completing grid distribution of a necessary bag type and a replaceable bag type according to the principles of necessary building priority, dynamic adaptation and closed-loop verification. The problem of traditional manual pain point arrangement is solved, and standardized, automatic and efficient operation is achieved; through dual-dimensional traffic prediction and equipment digital twin adaptation and in combination with four-level priority screening, packet type prediction is enabled to be precisely fit with a service and highly matched with equipment capability, and occupation of invalid resources is avoided.

Description

Technical Field

[0001] This invention relates to the field of express sorting technology, specifically to a sorting equipment package type intelligent arrangement method and system based on intelligent algorithms. Background Technology

[0002] In today's rapidly developing express delivery industry, traditional bag arrangement methods in sorting equipment have gradually revealed many drawbacks, failing to meet the ever-increasing customer demands and the industry's efficiency requirements. Specifically, express delivery involves a variety of bag types, including transit bags, intra-city bags, direct bags, connecting bags, and city service bags. How to rationally arrange bag types to appropriate slots on sorting equipment, and how to accurately and quickly adjust the bag arrangement when bag types change, all place high demands on sorting equipment managers.

[0003] Although equipment managers have attempted to solve these problems by manually arranging package types based on personal experience, current practices have many shortcomings. The allocation of package types to each set of equipment does not take into account the actual sorting capacity and number of slots on site, resulting in unreasonable package type allocations. There is also no unified standard for how many slots should be allocated to each package type. Packages with high traffic volume are not allocated enough slots, leading to insufficient processing time for slot changers, while packages with low traffic volume occupy too many slots, resulting in wasted capacity.

[0004] Furthermore, the arrangement of package types is rather arbitrary. Grouping packages with large geographical distances together can lead to missorting and severely impact delivery time. The package arrangement fails to consider the balance of package types and flow across different supply areas, increasing the return rate and reducing sorting efficiency. If packages are arranged at the corners or bends of the sorting equipment, they are prone to drifting and missorting.

[0005] Finally, the packaging rules are frequently adjusted. Manually notifying on-site staff via WeChat to modify the plan requires on-site managers to manually re-import the package layout plan from Excel, resulting in low efficiency and significantly reducing user experience. Failure to adjust the package layout in a timely manner can also lead to mismatched or missing compartments, causing packages to be unsorted or mis-sorted. To address this, we propose a smart algorithm-based intelligent package layout method and system for sorting equipment. Summary of the Invention

[0006] To address the aforementioned technical problems, this paper provides a method and system for intelligent bag arrangement in sorting equipment based on intelligent algorithms. This technical solution resolves the problems mentioned above.

[0007] To achieve the above objectives, the technical solution adopted by this invention is: a method for intelligent arrangement of package types in sorting equipment based on intelligent algorithms, wherein the intelligent arrangement steps are as follows: S1. Optimal package type calculation: Integrate multi-dimensional data to construct a two-dimensional traffic prediction model and generate package type traffic prediction results; Establish a digital twin model of sorting equipment to adapt and analyze the physical attributes and operating status of the equipment; Establish a package type priority system based on business rules and equipment capabilities to select the optimal package type set; S2. Packet type grid allocation: Based on the principles of priority for mandatory construction, dynamic adaptation and closed-loop verification, the grid allocation of mandatory packet types and replaceable packet types is completed in sequence. The allocation is accurately carried out through differential routing rules, dynamic coefficient correction mechanism and single-area grid upper limit verification, and the grid resource closed-loop control is triggered by the allocation stop mechanism. S3. The arrangement and distribution of package-type compartments are based on the goals of prioritizing geographical association, equipment adaptation, and optimizing flow balance. The basic layout rules are executed to complete the initial allocation of compartment positions. Fine-tuning is achieved through flow balance adjustments in the supply area and visualization and dynamic linkage updates of easily floating compartments.

[0008] Preferably, the dual-dimensional flow prediction model in step S1 is built based on the LSTM time series algorithm, which integrates the historical flow data of sorting equipment over the past 12 months, seasonal fluctuation coefficients, holiday delivery patterns and regional e-commerce promotion cycles, and dynamically corrects the prediction results by connecting to the express delivery business system in real time, and outputs the peak, average and fluctuation range of each type of package flow in the next 30 days.

[0009] Preferably, in step S1, the digital twin model of the sorting equipment is used to map the physical attributes and operating status of the equipment, specifically including the total number of cross belt slots, available slot threshold, number of single-ring supply areas, supply table carrying capacity, slot operation efficiency, and distribution parameters of easily drifting slots; based on the digital twin model of the sorting equipment, sorting scenarios with different package combinations are simulated, and the sorting efficiency, return rate, and slot utilization rate of the equipment under different flow loads are calculated, excluding package combinations that exceed the equipment's carrying capacity; Package type priority system is ordered according to the following hierarchy: The first priority is the mandatory package type that has been marked and confirmed by the package creation system; The second priority is high-traffic direct-to-the-port packets with a predicted traffic share of ≥5% and matching processing capacity at the end-point centers; The third priority is package types with potential growth that have seen a month-on-month increase of ≥15% in traffic over the past 3 months and meet the equipment sorting process requirements; The fourth priority is to select alternative package types that are not mandatory and have low traffic fluctuations and strong compatibility with other package types.

[0010] Preferably, in step S2, the differential routing rule triggers different allocation logic based on the difference between the number of packet types and the number of available slots. When the difference is greater than 5, the packet types are allocated in descending order according to the total number of available slots until the slots are exhausted. When the difference is less than 1, the number of slots is allocated by combining the packet type flow range with the comparison results of the average packet type flow in the single-ring supply area and the baseline value.

[0011] Preferably, the dynamic coefficient correction mechanism in step S2 introduces a coefficient for the average number of votes per package in the package organization, and its calculation formula is as follows: Average votes per package = Average votes per package in the past 30 days ÷ 30; The calculation results are rounded to two decimal places; when the coefficient is greater than 1, the correction benchmark 1 = 1000 votes × coefficient, and the correction benchmark 2 = 2500 votes × coefficient; when the coefficient is less than or equal to 1, the benchmark 1 and benchmark 2 remain at their default values. The upper limit of single-zone grid slots is accurately allocated. The upper limit threshold for single-package, single-supply zone grid slots is set to 4. Real-time verification is performed during the allocation process. Specifically: If the number of slots in a single supply area calculated according to the rules is ≤4, allocate directly according to the calculation results; If the number of slots in a single supply area calculated according to the rules is greater than 4, it will be forced to be allocated to 4 slots, synchronized to the package creation system, and a slot allocation over-limit reminder will be given.

[0012] Preferably, the basic layout rules in step S3 include: Grouping is based on geographic priority, and the package type is grouped according to the hierarchy of unpacking organization → terminal center → city where the terminal center is located → province where the terminal center is located → region. Based on the geographical grouping results and combined with the physical attributes of the sorting equipment, the location of the sorting slots is allocated in descending order of the number of slots in the three-level sorting system: province-city-unpacking organization; multiple slots for the same package type are distributed in different supply areas.

[0013] Preferably, in step S3, the flow balance fine-tuning of the supply area aims to minimize the flow difference between each supply area. When the flow difference between the supply area with the largest flow and the supply area with the smallest flow exceeds the configurable threshold, within the package type set of the same region and city, the grid positions of the package type with the largest flow in the supply area with the smallest flow in the supply area with the smallest flow are swapped, and the polling continues until the flow difference is ≤ the threshold or the number of iterations reaches 10.

[0014] Preferably, in step S3, the optimization and fine-tuning of the easy-to-float grid is aimed at the grids at curved corners with a historical misclassification rate of ≥3%. Priority is given to replacing the package type with the lowest flow rate under the same end center in the same supply area. This process is then expanded upwards to the same city, the same province, and the same region. If no package type meets the criteria, the original solution is retained and a prompt is displayed.

[0015] Preferably, the intelligent layout process also includes special scenario handling steps. Specifically, when initializing the layout of a new device, the unoccupied slots in the last outbound slot plan before going online are used as the initial available slots. After completing the allocation of mandatory package types according to the principle of geographical priority grouping and average distribution of supply areas, flow balancing fine-tuning is performed. When inserting temporary package types, the idle slots in the same region and city are selected first. If there are no idle slots, one slot of low flow package type is temporarily replaced according to the principle of minimum flow priority. After insertion, the fine-tuning process is triggered again.

[0016] A sorting equipment package intelligent layout system based on intelligent algorithms, the layout system includes: The optimal package type calculation module accurately outputs traffic data through a two-dimensional traffic prediction model, and combines it with the device digital twin model adaptation analysis to filter the optimal package type set according to four levels of priority. The package type slot allocation module completes the allocation of slots for mandatory and replaceable package types through differential routing and dynamic coefficient correction logic; The package-type grid arrangement module completes the basic layout based on geographical association and equipment characteristics. After two levels of fine-tuning and optimization of flow balancing and easy-to-float grids, the results are displayed and dynamically updated through a visual interface.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention addresses the pain points of traditional manual sorting, achieving standardized, automated, and efficient operations. Through dual-dimensional traffic prediction and digital twin adaptation of equipment, combined with four-level priority screening, it ensures accurate package type prediction that aligns with business needs and is highly compatible with equipment capabilities, avoiding unnecessary resource consumption. Standardized quantitative rules replace subjective human judgment, enabling flexible and efficient allocation of sorting resources through dynamic correction, upper limit verification, and closed-loop control, avoiding overload and waste, and reducing human intervention and errors. Based on geographically correlated sorting, traffic balance fine-tuning, and optimization of easily drifting sorting areas, it significantly reduces missorting rates and improves sorting smoothness. Simultaneously, a visual display and dynamic linkage update mechanism allow for rapid response to business changes when adjusting the solution, ultimately achieving the core objectives of controllable sorting efficiency fluctuations, meeting missorting rate targets, and improving resource utilization. Attached Figure Description

[0018] Figure 1 This is a flowchart of the intelligent layout steps of the present invention. Detailed Implementation

[0019] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0020] Reference Figure 1As shown, a method for intelligent bag arrangement in sorting equipment based on intelligent algorithms is described. The intelligent arrangement steps are as follows: S1. Optimal package type calculation: Integrate multi-dimensional data to construct a two-dimensional traffic prediction model and generate package type traffic prediction results; Establish a digital twin model of sorting equipment to adapt and analyze the physical attributes and operating status of the equipment; Establish a package type priority system based on business rules and equipment capabilities to select the optimal package type set; S2. Packet type grid allocation: Based on the principles of priority for mandatory construction, dynamic adaptation and closed-loop verification, the grid allocation of mandatory packet types and replaceable packet types is completed in sequence. The allocation is accurately carried out through differential routing rules, dynamic coefficient correction mechanism and single-area grid upper limit verification, and the grid resource closed-loop control is triggered by the allocation stop mechanism. S3. The arrangement and distribution of package-type compartments are based on the goals of prioritizing geographical association, equipment adaptation, and optimizing flow balance. The basic layout rules are executed to complete the initial allocation of compartment positions. Fine-tuning is achieved through flow balance adjustments in the supply area and visualization and dynamic linkage updates of easily floating compartments.

[0021] This application achieves precise matching between package type and actual needs and equipment capabilities through dual-dimensional traffic prediction, equipment digital twin models, and a standardized priority system. This avoids ineffective consumption from the source and ensures screening consistency. S2 uses differential routing, dynamic correction, and single-zone upper limit verification to quantitatively allocate grid slots, coupled with a closed-loop stopping mechanism to avoid resource waste and congestion. Grid slot utilization is increased to over 85%, eliminating human error. S3 relies on geographic association priority, traffic balance fine-tuning, and visualization functions to reduce the missorting rate to ≤1%, improving sorting smoothness and management convenience. Overall, it achieves standardized and intelligent operation, controlling sorting efficiency fluctuations within 5%, reducing adjustment time to within 30 minutes, adapting to multiple devices and multiple business scenarios, and significantly reducing management costs and operational risks.

[0022] In step S1, the dual-dimensional flow prediction model is built based on the LSTM time series algorithm. It integrates the historical flow data of sorting equipment over the past 12 months, seasonal fluctuation coefficients, holiday delivery patterns, and regional e-commerce promotion cycles. It also connects to the express delivery business system in real time to dynamically correct the prediction results and outputs the peak, average, and fluctuation range of each package type flow in the next 30 days.

[0023] This application leverages the strong fitting capability of the LSTM algorithm for time series data, integrating multi-dimensional data such as 12 months of historical traffic, seasonal fluctuations, holidays, and e-commerce promotions. It avoids the one-sidedness of traditional single historical data prediction, accurately captures traffic change patterns, and makes the prediction results closer to real business scenarios, reducing grid resource mismatch caused by high traffic omissions and low traffic misjudgments.

[0024] In step S1, the digital twin model of the sorting equipment is used to map the physical attributes and operating status of the equipment. Specifically, it includes the total number of cross belt slots, the available slot threshold, the number of single-ring supply areas, the load-bearing capacity of the supply table, the slot operation efficiency, and the distribution parameters of easily drifting slots. Based on the digital twin model of the sorting equipment, sorting scenarios with different package combinations are simulated to calculate the sorting efficiency, return rate, and slot utilization rate of the equipment under different flow loads, and package combinations that exceed the equipment's load-bearing capacity are excluded. Package type priority system is ordered according to the following hierarchy: The first priority is the mandatory package type that has been marked and confirmed by the package creation system; The second priority is high-traffic direct-to-the-port packets with a predicted traffic share of ≥5% and matching processing capacity at the end-point centers; The third priority is package types with potential growth that have seen a month-on-month increase of ≥15% in traffic over the past 3 months and meet the equipment sorting process requirements; The fourth priority is to select alternative package types that are not mandatory and have low traffic fluctuations and strong compatibility with other package types.

[0025] This application comprehensively maps key parameters such as the total number of sorting slots, the capacity of the feeding station, and the location of easily overflowing slots. Through scenario simulation, it accurately calculates the sorting efficiency and return rate of different package combinations, and preemptively eliminates package combinations that exceed the equipment's capacity. This avoids sorting congestion and equipment failure caused by traditional experience-based package selection, ensuring that the equipment always operates within its optimal load range. Without actually running the equipment, the application can predict the operational effect of package combinations through digital twin simulation, avoiding passive adjustments that only become apparent after the solution has been implemented, and reducing sorting interruptions and rework costs.

[0026] In step S2, the differential routing rule triggers different allocation logic based on the difference between the number of packet types and the number of available slots. When the difference is greater than 5, the packet types are allocated in descending order according to the total number of available slots until the slots are exhausted. When the difference is less than 1, the number of slots is allocated by combining the packet type flow range with the comparison results of the average packet type flow in the single-ring supply area and the baseline value.

[0027] This application addresses different supply and demand imbalances by dividing scenarios based on the difference between "number of package types - number of available slots": when the difference is greater than 5 (more package types, fewer slots), allocation is based on descending flow rate to ensure that high-flow, high-value package types receive priority access to slots, avoiding "core package types being squeezed out by low-flow package types"; when the difference is less than 1 (few package types, sufficient slots), allocation is based on the flow range and the average flow rate of the supply area, satisfying both package type sorting needs and avoiding waste of surplus slots, achieving "supply and demand matching without redundancy".

[0028] The dynamic coefficient correction mechanism described in step S2 introduces a coefficient for the average number of votes per package within the package organization. The calculation formula for this coefficient is as follows: Average votes per package = Average votes per package in the past 30 days ÷ 30; The calculation results are rounded to two decimal places; when the coefficient is greater than 1, the correction benchmark 1 = 1000 votes × coefficient, and the correction benchmark 2 = 2500 votes × coefficient; when the coefficient is less than or equal to 1, the benchmark 1 and benchmark 2 remain at their default values. The upper limit of single-zone grid slots is accurately allocated. The upper limit threshold for single-package, single-supply zone grid slots is set to 4. Real-time verification is performed during the allocation process. Specifically: If the number of slots in a single supply area calculated according to the rules is ≤4, allocate directly according to the calculation results; If the number of slots in a single supply area calculated according to the rules is greater than 4, it will be forced to be allocated to 4 slots, synchronized to the package creation system, and a slot allocation over-limit reminder will be given.

[0029] The sorting efficiency and operational proficiency of different packaging organizations vary (e.g., some organizations have a higher average number of tickets per package per day). Traditional fixed benchmark values ​​would lead to "insufficient storage space for organizations with strong capabilities and idle storage space for organizations with weak capabilities". By dynamically adjusting the benchmark value through the average number of tickets per package, the allocation standard is linked to the actual capabilities of the organization, achieving a precise "one standard per organization" fit.

[0030] The basic layout rules in step S3 include: Grouping is based on geographic priority, and the package type is grouped according to the hierarchy of unpacking organization → terminal center → city where the terminal center is located → province where the terminal center is located → region. Based on the geographical grouping results and combined with the physical attributes of the sorting equipment, the location of the sorting slots is allocated in descending order of the number of slots in the three-level sorting system: province-city-unpacking organization; multiple slots for the same package type are distributed in different supply areas.

[0031] In step S3, the flow balance fine-tuning of the supply area aims to minimize the flow difference between each supply area. When the flow difference between the supply area with the largest flow and the supply area with the smallest flow exceeds the configurable threshold, the grid positions of the package type with the largest flow in the supply area with the smallest flow in the supply area with the smallest flow are swapped within the package type set of the same region and city. This process continues until the flow difference is ≤ the threshold or the number of iterations reaches 10.

[0032] In step S3, the optimization and fine-tuning of the easy-to-float grid is aimed at the grids at curved corners with a historical misclassification rate of ≥3%. Prioritize replacing the package type with the lowest flow rate under the same end center in the same supply area, and then expand upwards to the same city, the same province, and the same region. If no package type meets the conditions, retain the original solution and mark it as a prompt.

[0033] The intelligent deployment process also includes special scenario handling steps. Specifically, when initializing the deployment of new equipment, unoccupied slots from the last outbound slot plan before going online are used as the initial available slots. After completing the allocation of mandatory package types according to the principle of geographical priority grouping and average distribution of supply areas, flow balancing fine-tuning is performed. When inserting temporary package types, idle slots in the same region and city are prioritized. If there are no idle slots, one slot of low-flow package type is temporarily replaced according to the principle of minimum flow priority. After insertion, the fine-tuning process is retried.

[0034] A sorting equipment package intelligent layout system based on intelligent algorithms, the layout system includes: The optimal package type calculation module accurately outputs traffic data through a two-dimensional traffic prediction model, and combines it with the device digital twin model adaptation analysis to filter the optimal package type set according to four levels of priority. The package type slot allocation module completes the allocation of slots for mandatory and replaceable package types through differential routing and dynamic coefficient correction logic; The package-type grid arrangement module completes the basic layout based on geographical association and equipment characteristics. After two levels of fine-tuning and optimization of flow balancing and easy-to-float grids, the results are displayed and dynamically updated through a visual interface.

[0035] Example 1: Intelligent Packaging Arrangement of Cross-Belt Sorting Machine This embodiment is applied to a cross-belt sorting machine with a total of 200 sorting slots, 4 single-ring feeding areas (A, B, C, D), and 12 easily drifting slots (distributed at the curved corners).

[0036] Intelligent prediction of package type traffic integrates the sorting machine's historical traffic data for the past 12 months, seasonal fluctuation coefficients, holiday patterns, and regional e-commerce promotion cycles to build an LSTM prediction model. This model predicts the peak, average, and fluctuation range of traffic for each package type over the next 30 days. Simultaneously, it accesses new merchant cooperation data from the express delivery business system in real time to correct the prediction results. For example, if the predicted traffic share of direct-delivery packages from a newly partnered e-commerce platform reaches 6%, it is included in the category of high-traffic direct-delivery packages. A capacity adaptation analysis of the sorting equipment was conducted to establish a digital twin model of the cross-belt sorting machine. Parameters such as the total number of slots (200), the available slot threshold (180), the number of single-ring supply areas (4), the supply table capacity (500 packages / hour), and the location of easily floating slots were entered. The model was used to simulate sorting scenarios with different package combinations and to exclude package combinations that exceed the equipment's capacity. For example, a package type was predicted to require 50 slots for peak flow, which exceeded the available slot threshold and was therefore excluded. The intelligent filtering of package types prioritizes 15 mandatory package types, 8 high-traffic direct-access package types (predicted traffic share ≥5%), 5 potential growth package types (month-on-month growth ≥15% in the past 3 months), and 20 replaceable package types, forming an optimal set of 48 package types. The allocation of mandatory package types is calculated by "package type quantity (15) - available slot quantity (180) < 1", and then the allocation is entered into the subdivision rules. The flow rate of a certain mandatory package type is 8000 tickets, and the average package type flow rate of a single ring supply area is 8000 ÷ 4 = 2000 tickets, which is in the range of 1000 tickets < 2000 tickets ≤ 2500 tickets. The number of slots allocated is 4. At the same time, the average ticket quantity coefficient of the package organization is introduced as 1.12. After correction, the baseline value 1 = 1000 × 1.12 = 1120 tickets, and the baseline value 2 = 2500 × 1.12 = 2800 tickets. The allocation results are re-verified to ensure that they are adapted to the sorting capacity of the package organization. After all mandatory package types are allocated, 60 slots are occupied. Replaceable package type allocation, surplus slots = 180 - 60 = 120, replaceable package type allocation is initiated, the logic of mandatory package type allocation is reused, the final number of allocated slots is 100, and there are 20 surplus slots remaining; Additional allocation is performed, sorted in descending order by average number of slots, and additional slots are allocated to the already allocated packet types until the surplus slots are exhausted. Finally, 2 additional slots are added to a certain high-volume packet type, reaching 6 slots.

[0037] Bag compartment arrangement: The basic layout is grouped according to "unpacking organization → terminal center → city → province → region". Package types in the East China region are arranged in a concentrated manner. Package types from the same unpacking organization are assigned to adjacent compartments. A certain package type is assigned to 3 compartments, distributed in supply areas A, B, and C respectively. The next package type is assigned starting from supply area D, realizing cross-distribution of supply areas. Priority is given to allocating the inner 100 compartments. After the inner circle is exhausted, the outer circle compartments are allocated. For traffic balance fine-tuning, calculate the total traffic of each supply area. The total traffic of supply area A is 12,000 tickets, and the total traffic of supply area D is 8,000 tickets. The difference of 400 tickets is less than the threshold of 500 tickets, so no fine-tuning is required. Easy-to-distribute grid optimization: If a certain easy-to-distribute grid is currently allocated 3000 packages, select the package type with the smallest traffic (500 packages) under the same end center in the same supply area and switch grids to reduce the risk of mis-distribution; The visual display shows that the package types in East China are marked in blue and those in South China are marked in red. Hovering the mouse over a cell displays "Package type code: ZD001, traffic: 8000 packages, supply area: A, average daily processing volume: 267 packages". Cells prone to floating are marked with an orange border. Dynamic adjustment: During operation, if a fault occurs in a certain supply area B, the equipment monitoring system will provide real-time feedback and automatically trigger the recalculation of the layout plan. The package types originally assigned to supply area B will be adjusted to other supply areas. The adjustment takes 25 minutes and is synchronized to the package creation system and pushed to the administrator. In special scenarios, during e-commerce promotions, traffic to a certain type of bag suddenly increases, with the predicted traffic deviating by 15% from the original forecast. The system automatically triggers a fine-tuning process, adding one more compartment to the bag type to ensure balanced traffic. Through the method and system of the present invention, the sorting efficiency fluctuation of the cross belt sorter is controlled within 4%, the missorting rate is reduced to 1.2%, the grid utilization rate reaches 90%, and the time for bag type adjustment is shortened from the original 4.5 hours to 25 minutes, which significantly improves sorting efficiency and management level. Example 2: Initial Layout of New Equipment A newly launched cross-belt sorting machine has no historical layout data and a total of 150 slots. Available grid initialization: The 130 grids that were not occupied by the package creation scheme in the last outbound "grid scheme" before going online are the initial available grids, and the initial abnormal grid limit is set to 1; Package type filtering: 12 essential package types, 6 high-traffic direct-access package types, 3 package types with potential growth, and 15 replaceable package types, for a total of 36 package types; The allocation of storage compartments is based on the principle of prioritizing mandatory construction. The mandatory bag type is allocated 45 compartments. Replaceable bag types are then allocated 60 compartments, leaving 25 spare compartments. Additional compartments are allocated based on the average number of bags dropped into the compartment. Basic layout and fine-tuning: The basic layout is completed according to the principle of grouping by geographical priority and even distribution of supply areas. Flow balancing fine-tuning is performed to generate the initial layout plan. Visualization and synchronization: The initial layout plan is synchronized to the visualization page and related systems. Administrators can verify the rationality of the plan through the visualization page, and the equipment can be put into operation after one-click application.

[0038] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for intelligent arrangement of package types in sorting equipment based on intelligent algorithms, characterized in that, Includes the following steps: S1. Optimal package type calculation: Integrate multi-dimensional data to construct a two-dimensional traffic prediction model and generate package type traffic prediction results; Establish a digital twin model of sorting equipment to adapt and analyze the physical attributes and operating status of the equipment; Establish a package type priority system based on business rules and equipment capabilities to select the optimal package type set; S2. Packet type grid allocation: Based on the principles of priority for mandatory construction, dynamic adaptation and closed-loop verification, the grid allocation of mandatory packet types and replaceable packet types is completed in sequence. The allocation is accurately carried out through differential routing rules, dynamic coefficient correction mechanism and single-area grid upper limit verification, and the grid resource closed-loop control is triggered by the allocation stop mechanism. S3. The arrangement and distribution of package-type compartments are based on the goals of prioritizing geographical association, equipment adaptation, and optimizing flow balance. The basic layout rules are executed to complete the initial allocation of compartment positions. Fine-tuning is achieved through flow balance adjustments in the supply area and visualization and dynamic linkage updates of easily floating compartments.

2. The intelligent bag arrangement method for sorting equipment based on intelligent algorithms according to claim 1, characterized in that: In step S1, the dual-dimensional flow prediction model is built based on the LSTM time series algorithm. It integrates the historical flow data of sorting equipment over the past 12 months, seasonal fluctuation coefficients, holiday delivery patterns, and regional e-commerce promotion cycles. It also connects to the express delivery business system in real time to dynamically correct the prediction results and outputs the peak, average, and fluctuation range of each package type flow in the next 30 days.

3. The intelligent bag arrangement method for sorting equipment based on intelligent algorithms according to claim 1, characterized in that: In step S1, the digital twin model of the sorting equipment is used to map the physical attributes and operating status of the equipment. Specifically, it includes the total number of cross belt slots, the available slot threshold, the number of single-ring supply areas, the load-bearing capacity of the supply table, the slot operation efficiency, and the distribution parameters of easily drifting slots. Based on the digital twin model of the sorting equipment, sorting scenarios with different package combinations are simulated to calculate the sorting efficiency, return rate, and slot utilization rate of the equipment under different flow loads, and package combinations that exceed the equipment's load-bearing capacity are excluded. Package type priority system is ordered according to the following hierarchy: The first priority is the mandatory package type that has been marked and confirmed by the package creation system; The second priority is high-traffic direct-to-the-port packets with a predicted traffic share of ≥5% and matching processing capacity at the end-point centers; The third priority is package types with potential growth that have seen a month-on-month increase of ≥15% in traffic over the past 3 months and meet the equipment sorting process requirements; The fourth priority is to select alternative package types that are not mandatory and have low traffic fluctuations and strong compatibility with other package types.

4. The intelligent bag arrangement method for sorting equipment based on intelligent algorithms according to claim 1, characterized in that: In step S2, the differential routing rule triggers different allocation logic based on the difference between the number of packet types and the number of available slots. When the difference is greater than 5, the packet types are allocated in descending order according to the total number of available slots until the slots are exhausted. When the difference is less than 1, the number of slots is allocated by combining the packet type flow range with the comparison results of the average packet type flow in the single-ring supply area and the baseline value.

5. The intelligent bag arrangement method for sorting equipment based on intelligent algorithms according to claim 1, characterized in that: The dynamic coefficient correction mechanism described in step S2 introduces a coefficient for the average number of votes per package within the package organization. The calculation formula for this coefficient is as follows: Average votes per package = Average votes per package in the past 30 days ÷ 30; The calculation results are rounded to two decimal places; when the coefficient is greater than 1, the correction benchmark 1 = 1000 votes × coefficient, and the correction benchmark 2 = 2500 votes × coefficient; when the coefficient is less than or equal to 1, the benchmark 1 and benchmark 2 remain at their default values. The upper limit of single-zone grid slots is accurately allocated. The upper limit threshold for single-package, single-supply zone grid slots is set to 4. Real-time verification is performed during the allocation process. Specifically: If the number of slots in a single supply area calculated according to the rules is ≤4, allocate directly according to the calculation results; If the number of slots in a single supply area calculated according to the rules is greater than 4, it will be forced to be allocated to 4 slots, synchronized to the package creation system, and a slot allocation over-limit reminder will be given.

6. The intelligent bag arrangement method for sorting equipment based on intelligent algorithms according to claim 1, characterized in that, The basic layout rules in step S3 include: Grouping is based on geographic priority, and the package type is grouped according to the hierarchy of unpacking organization → terminal center → city where the terminal center is located → province where the terminal center is located → region. Based on the geographical grouping results and combined with the physical attributes of the sorting equipment, the location of the sorting slots is allocated in descending order of the number of slots in the three-level sorting system: province-city-unpacking organization; multiple slots for the same package type are distributed in different supply areas.

7. The intelligent bag arrangement method for sorting equipment based on intelligent algorithms according to claim 1, characterized in that: In step S3, the flow balance fine-tuning of the supply area aims to minimize the flow difference between each supply area. When the flow difference between the supply area with the largest flow and the supply area with the smallest flow exceeds the configurable threshold, the grid positions of the package type with the largest flow in the supply area with the smallest flow in the supply area with the smallest flow are swapped within the package type set of the same region and city. This process continues until the flow difference is ≤ the threshold or the number of iterations reaches 10.

8. The intelligent arrangement method for package types in sorting equipment based on intelligent algorithms according to claim 1, characterized in that: In step S3, the optimization and fine-tuning of the easy-to-float grid is aimed at the grids at curved corners with a historical misclassification rate of ≥3%. Prioritize replacing the package type with the lowest flow rate under the same end center in the same supply area, and then expand upwards to the same city, the same province, and the same region. If no package type meets the conditions, retain the original solution and mark it as a prompt.

9. The intelligent bag arrangement method for sorting equipment based on intelligent algorithms according to claim 1, characterized in that: The intelligent layout process also includes special scenario handling steps, specifically including, when initializing the layout of new equipment, taking the unoccupied grids in the last outbound grid plan before going online as the initial available grids, completing the allocation of mandatory package types according to the principle of geographical priority grouping and average distribution of supply areas, and then performing flow balancing fine-tuning. When inserting a temporary packet, priority is given to selecting available slots in the same region and city. If no available slots are available, one slot of the low-traffic packet type is temporarily replaced according to the principle of minimum traffic priority. After insertion, the fine-tuning process is triggered again.

10. A sorting equipment package type intelligent layout system based on intelligent algorithms, applied to a sorting equipment package type intelligent layout method based on intelligent algorithms according to any one of claims 1 to 9, characterized in that, The layout system includes: The optimal package type calculation module accurately outputs traffic data through a two-dimensional traffic prediction model, and combines it with the device digital twin model adaptation analysis to filter the optimal package type set according to four levels of priority. The package type slot allocation module completes the allocation of slots for mandatory and replaceable package types through differential routing and dynamic coefficient correction logic; The package-type grid arrangement module completes the basic layout based on geographical association and equipment characteristics. After two levels of fine-tuning and optimization of flow balancing and easy-to-float grids, the results are displayed and dynamically updated through a visual interface.

Images