Multi-strategy collaborative allocation and route scheduling method for automotive parts inbound logistics

By constructing a multi-strategy collaborative allocation and path scheduling method, the problems of accuracy and automation in the selection of automotive parts inbound logistics mode in existing technologies are solved, and the comprehensive optimization of transportation costs and loading efficiency is achieved, ensuring the feasibility and economy of the logistics solution.

CN123117972APending Publication Date: 2026-11-17HEFEI UNIV OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611635999.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-22
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

The existing automotive parts inbound logistics solutions lack a unified quantitative decision-making model, making it difficult to accurately judge the overall advantages and disadvantages of different inbound modes. Furthermore, human experience makes it difficult to systematically compare the differences between multiple modes, resulting in local optima and failing to simultaneously optimize transportation costs and loading efficiency.

Method used

A multi-strategy collaborative allocation and route scheduling method is constructed. By building a logistics feature vector of multi-dimensional attributes of parts and components, and combining the feasibility verification of direct delivery and circular pickup modes, a multi-mode collaborative optimization model is generated to uniformly solve the direct delivery candidate scheme and circular pickup route, taking into account transportation costs and loading capacity.

Benefits of technology

It improved the accuracy and automation of the selection of logistics modes for parts entering the factory and the planning of circular pickup routes, reduced low vehicle loading rates and repeated transportation, reduced total freight costs and node operation costs, and ensured the feasibility of the scheduling plan.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123117972A_ABST
    Figure CN123117972A_ABST
Patent Text Reader

Abstract

This invention relates to the field of logistics planning technology and discloses a multi-strategy collaborative allocation and route scheduling method for automotive parts inbound logistics. The method acquires basic data related to automotive parts inbound logistics and converts the original demand for various parts into the demand for standard transport packaging units; using parts as the basic decision-making object, it determines the feasibility of direct delivery for various parts using a direct delivery mode; it calculates the required number of transport trips and generates corresponding direct delivery candidate schemes; it generates initial circular pickup routes; it integrates the direct delivery candidate schemes and the verified circular pickup candidate routes, constructing a multi-mode collaborative optimization model with the goal of minimizing overall logistics costs; it solves the optimization model and outputs the final inbound logistics mode allocation scheme and circular pickup route scheduling scheme for each part. This invention improves the accuracy and automation of parts inbound logistics mode selection and circular pickup route planning, thereby saving logistics costs.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of logistics planning technology, specifically a multi-strategy collaborative allocation and route scheduling method for the logistics of automotive parts entering the factory, as well as a computer terminal and computer-readable storage medium for applying this method. Background Technology

[0002] The automotive manufacturing process is characterized by a wide variety of products, multiple suppliers, high frequency of production, and strong cyclical coordination. OEMs continuously source various components from external suppliers during production, including seats, wiring harnesses, dashboards, bumpers, interior parts, braking components, electronic modules, stamped parts, and standard parts. Since vehicle production lines typically operate continuously at a fixed pace, untimely supply of critical components can lead to material shortages at the production line, production delays, and even line stoppages. Therefore, the inbound logistics system not only serves a transportation function but also undertakes tasks such as coordinating supply rhythms, controlling transportation costs, organizing vehicle resources, and ensuring production continuity. Common inbound logistics models for automotive parts include direct delivery to the OEM, cyclical pickup to the OEM, direct delivery to regional distribution centers, and cyclical pickup to regional distribution centers, with multiple inbound logistics models often coexisting.

[0003] However, existing inbound logistics solutions still rely heavily on manual experience and phased decision-making. Logistics planners typically determine whether a particular type of component should be delivered directly, via a loop pickup system, or via an RDC (Regional Distribution Center) based on factors such as component volume, supplier distance, historical shipping methods, single-trip demand, or on-site experience. They then proceed with vehicle route planning, supplier visit sequence scheduling, and transportation cost calculations. This approach is feasible in small-scale scenarios, but as the number of suppliers, component types, and vehicle models and packaging specifications increase, manual rules struggle to systematically compare the comprehensive differences between different inbound models. For example, a direct delivery model might have a shorter route but could result in lower vehicle loading rates; a loop pickup model allows for consolidated transport but is limited by vehicle capacity, route time, and packaging compatibility; and an RDC-related model enables transit consolidation and subsequent delivery organization but incurs RDC processing and secondary delivery costs. Without a unified quantitative decision-making model, it is difficult to accurately assess the overall advantages and disadvantages of different models.

[0004] Existing vehicle routing methods often assume that the transportation mode is already determined. For example, they may pre-assuming all suppliers use circular pickup before solving the routing problem, or pre-assigning some parts to RDCs and then calculating their transportation costs separately. While these methods can optimize routes or reduce transportation mileage to some extent, they fail to address the fundamental decision-making problem of "which delivery mode should be used for each type of part," and they also struggle to simultaneously characterize the substitution relationships between direct delivery, circular pickup, and RDC transfer. Furthermore, this sequential approach can lead to local optima. For instance, a part might have a high cost for direct delivery alone, but if its supplier is located close to other suppliers, including it in a circular pickup route might result in a lower overall cost. Conversely, while a part could be included in a circular pickup route, poor packaging, volume, or loading adaptability could reduce the overall vehicle loading efficiency of the route. If transportation mode selection and route planning are treated separately, it is difficult to obtain a comprehensively optimal solution.

[0005] Therefore, in the context of automotive parts inbound logistics, there is an urgent need for a technical solution that can take parts as the basic decision-making object, comprehensively consider multiple factors, and automatically generate logistics mode allocation and cyclic pickup route schemes. Summary of the Invention

[0006] To address the technical problems existing in the prior art, this invention provides a multi-strategy collaborative allocation and route scheduling method for automotive parts inbound logistics, aiming to improve the accuracy and automation of parts inbound logistics mode selection and cyclic pickup route planning, thereby saving logistics costs.

[0007] To achieve the above objectives, the present invention provides the following technical solution: This invention discloses a multi-strategy collaborative allocation and path scheduling method for inbound logistics of automotive parts, comprising the following steps: S1. Obtain basic data related to the logistics of automotive parts entering the factory, and convert the original demand for various parts into the demand for standard transport packaging units; S2. Based on the components as the decision-making object, construct a logistics feature vector containing multi-dimensional attributes of the components. Based on the logistics timeliness requirements and operational adaptation conditions in the logistics feature vector, determine the feasibility of direct delivery of various components using the direct delivery mode. S3. For components that are feasible for direct delivery, calculate the required number of transport trips and generate corresponding direct delivery candidate schemes based on the demand for their corresponding transport packaging units and the loading parameters of the candidate vehicle models. S4. Using each supplier as a candidate access node, generate an initial cyclic pickup route that sequentially visits at least one candidate access node and ends at the OEM or regional distribution center, based on the spatial location information of each supplier. S5. Verify the running time, actual logistics timeliness and multi-dimensional loading capacity of the initial circular pickup route, and determine the feasibility of circular pickup mode for various parts, and obtain the verified circular pickup route. S6. Combining the aforementioned direct delivery candidate schemes and the verified cyclic pickup candidate routes, a multi-mode collaborative optimization model is constructed with the goal of minimizing the overall inbound logistics cost within the planned period. S7. Solve the multi-mode collaborative optimization model and output the final inbound logistics mode allocation scheme and the circular pickup route scheduling scheme corresponding to each component; wherein, the inbound logistics mode includes: direct delivery to the OEM, circular pickup to the OEM, direct delivery to the regional distribution center, and circular pickup to the regional distribution center; different components under the same supplier can be independently allocated to different inbound logistics modes according to the solution results.

[0008] As a further improvement to the above scheme, step S2 specifically includes the following process: Extract the demand for transport packaging units of components, the weight of a single package, the length, width and height of a single package, the theoretical volume of a single package, the bottom area of ​​a single package, the historical demand fluctuation coefficient, the stockout risk level, the spatial clustering degree of suppliers, the maximum allowable logistics time, and the adaptation parameters to the operating conditions of the regional distribution center, and construct the logistics feature vector. Calculate the logistics time for direct delivery of parts to the OEM. If the logistics time is not greater than the maximum allowable logistics time for the part, then the feasibility of direct delivery of the part to the OEM is confirmed. Determine whether the regional distribution center can receive and process the packaging of parts; if so, determine that the operating condition adaptation parameters meet the requirements. Calculate the total logistics time of the parts via the regional distribution center. If the operating condition adaptation parameters meet the requirements and the total logistics time is not greater than the maximum allowable logistics time of the parts, then the feasibility of direct delivery of the parts to the regional distribution center is determined.

[0009] As a further improvement to the above scheme, step S5 specifically includes the following process: Calculate the total running time of the initial cyclic pickup route, and remove initial cyclic pickup routes whose total running time exceeds the preset maximum allowable running time; For the retained initial cyclic pickup route, the actual logistics time from the completion of pickup to arrival at the destination is calculated based on the actual access order of each candidate access node; A comprehensive adaptation parameter for components and routes is constructed. The comprehensive adaptation parameter is deemed valid when the initial cyclic pickup route meets all of the following conditions: C1. The route passes through the supplier of the component; C2. The component has been prepared when the vehicle arrives at the supplier; C3. The latest arrival time requirement is met; C4. The actual logistics time is not greater than the maximum allowable logistics time; C5. The operating condition adaptation parameter meets the requirements when the route destination is a regional distribution center. If there exists a cyclic pickup route with the OEM as the endpoint and the overall matching parameters are met, then the feasibility of using the cyclic pickup to OEM mode for this part is confirmed. If there exists a cyclic pickup route with the regional distribution center as the endpoint and the overall matching parameters are met, then the feasibility of using the cyclic pickup to regional distribution center mode for this part is confirmed.

[0010] As a further improvement to the above scheme, in step S6, the objective function of the multi-mode collaborative optimization model is: ; In the formula, This refers to the total cost of inbound logistics within the planned period. For parts Vehicle Model The transportation cost of direct delivery to the OEM; To indicate components Whether to use vehicle model 0-1 variables directly sent to the OEM; For parts Vehicle Model Front-end transportation costs for direct delivery to regional distribution centers; To indicate components Whether to use vehicle model 0-1 variables for direct delivery to regional distribution centers; Candidate route instances Execution costs; To represent candidate route instances Whether a 0-1 variable is enabled; For parts Unit cost of processing and subsequent delivery through regional distribution centers; For parts The demand for transport packaging units; To indicate components A 0-1 variable indicating whether to select the direct delivery to regional distribution center mode; To indicate components The 0-1 variable indicating whether to select the cyclic pickup to regional distribution center mode.

[0011] As a further improvement to the above scheme, in step S6, the constraints of the multi-mode collaborative optimization model include: The mode unique selection constraint is used to restrict each type of component to only one inbound logistics mode during the planning cycle. Pattern feasibility constraints are used to limit the model to select only from the set of patterns that have either direct delivery feasibility or cyclic pickup feasibility; The direct delivery vehicle selection constraint is used to restrict that when the parts are selected to be delivered directly to the OEM or to the regional distribution center, only one candidate vehicle must be selected. Demand satisfaction constraints are used to ensure that the number of component packages loaded on the enabled circular pickup route is equal to the total demand for that component under the corresponding circular pickup mode. The component-route association constraint is used to ensure that components can only be loaded into a loop pickup route that passes through its candidate access node, meets its logistics timeliness requirements, and has been activated. The comprehensive loading capacity constraint is used to ensure that the total amount of component packaging loaded on the cyclic pickup route does not exceed the load limit of each dimension of the vehicle model. The comprehensive loading capacity constraint specifically includes load constraint, equivalent volume constraint, and calculated bottom area constraint combined with the effective stacking layer number.

[0012] As a further improvement to the above scheme, in step S7, the multi-mode collaborative optimization model is solved by a mixed integer programming solver, or a candidate column substitution model is established for solving. The step of establishing a candidate column substitution model for solving the problem includes: representing the direct delivery candidate scheme and the verified cyclic pickup route as a unified candidate column, with each candidate column including the logistics mode, vehicle type, corresponding route instance, quantity of transport packaging for each component, and candidate column cost; setting component quantity balance constraints in the multi-mode collaborative optimization model to allow the same component to be transported by multiple candidate columns under the same inbound logistics mode, while prohibiting the same component from being transported across different inbound logistics modes.

[0013] As a further improvement to the above scheme, step S3 specifically includes the following process: For components that are feasible for direct delivery, the minimum number of transport trips required for each candidate vehicle model is calculated from three dimensions: vehicle effective load, equivalent volume, and equivalent floor area. The maximum value among the three calculation results is taken as the required number of transport trips, expressed by the following formula: ; In the formula, For parts Using candidate models The number of transport trips required for direct delivery; For parts The demand for transport packaging units; For parts The corresponding single transport packaging unit The weight; Candidate models Effective load capacity; For parts The corresponding single transport packaging unit The equivalent volume factor of the packaging; For parts The corresponding single transport packaging unit Theoretical volume; Candidate models Effective volume; For parts The corresponding single transport packaging unit The base area; For parts The corresponding single transport packaging unit In candidate models The effective stacking layer; Candidate models The effective base area; Based on the required number of transport trips, the fixed activation cost per trip for candidate vehicle models, and the cost per unit mileage, the corresponding direct delivery candidate schemes are calculated and generated.

[0014] As a further improvement to the above solution, in step S1, obtaining basic data related to the logistics of automotive parts entering the factory includes obtaining distance and running time data between nodes. The specific methods for obtaining this data include: The OEMs, regional distribution centers, and suppliers are uniformly mapped as spatial nodes; By calling the enterprise transportation management system, geographic information system, or navigation interface, the actual road distance and estimated travel time between the spatial nodes can be obtained.

[0015] The present invention also discloses a computer terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics as described above.

[0016] The present invention also discloses a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, it implements the steps of the multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics as described above.

[0017] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention breaks away from the traditional logistics planning process of "fixing the transportation mode first and then scheduling the route" by incorporating all direct delivery candidate schemes and multi-dimensionally validated cyclic pickup routes into a unified collaborative optimization model. This joint decision-making mechanism directly uses specific components as scheduling objects, allowing different specifications of materials from the same supplier to be independently matched with different inbound modes. It groups suitable goods for shared transport into cyclic pickup routes, reducing low vehicle loading rates and redundant transportation caused by decentralized direct delivery, ultimately reducing total freight costs and node operation costs within the planning cycle.

[0018] 2. This invention achieves three-dimensional cross-validation of the loading status of candidate vehicles by introducing a comprehensive constraint strategy in the route access phase, which considers load, equivalent volume, and the calculated base area combined with the effective stacking layer number. This mechanism corrects the bias of simply estimating vehicle dispatching demand based on cumulative volume or weight, and directly eliminates invalid route combinations that only meet the theoretical requirements but cannot be accommodated in the actual physical space at the algorithm level, thus ensuring the feasibility of the vehicle dispatching plan.

[0019] 3. This invention constructs feature vectors by extracting component dimensions, weights, and operational condition adaptation parameters before model calculation. This pre-verification process checks whether the regional distribution center can accept specific packaging and whether the entire supply chain time exceeds the limit. This early screening mechanism filters out alternative solutions that exceed hardware processing capabilities or cannot meet the latest delivery time. While controlling the scale of subsequent mixed-integer programming model solutions, it ensures that the final output scheduling scheme can smoothly adapt to the actual on-site operational conditions. Attached Figure Description

[0020] Figure 1 This is a flowchart of the multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics in Embodiment 1 of the present invention.

[0021] Figure 2 This is a flowchart of the component logistics feature vector and pattern feasibility screening process in Embodiment 1 of the present invention.

[0022] Figure 3 This is a flowchart of candidate route generation, loading, and timeliness verification in Embodiment 1 of the present invention.

[0023] Figure 4 This is a flowchart of the multi-mode collaborative optimization process in Embodiment 1 of the present invention.

[0024] Figure 5 This is the optimized allocation result of the four logistics modes in Embodiment 2 of the present invention.

[0025] Figure 6 This describes the logistics model for each supplier's components in Embodiment 2 of the present invention.

[0026] Figure 7 This is the optimized cyclic pickup route diagram in Embodiment 2 of the present invention.

[0027] Figure 8 This refers to the multi-dimensional loading utilization rate of the selected cyclic pickup route in Embodiment 2 of the present invention.

[0028] Figure 9 This is a comparison of the overall logistics costs before and after optimization in Embodiment 2 of the present invention.

[0029] Figure 10 This is a schematic diagram of the structure of the computer terminal in Embodiment 3 of the present invention. Detailed Implementation

[0030] 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.

[0031] Example 1 This embodiment provides a multi-strategy collaborative allocation and route scheduling method for automotive parts inbound logistics. This method takes parts as the basic decision-making object and incorporates the following four inbound logistics modes into a unified decision-making framework: direct delivery to the OEM, cyclic pickup to the OEM, direct delivery to the regional distribution center, and cyclic pickup to the regional distribution center.

[0032] The direct-to-OEM model refers to suppliers transporting parts directly to the OEM's receiving area or production line area. This model has a simple path and fast response, making it suitable for high-value, high-volume, time-sensitive, or urgent parts. However, when there are many suppliers and the shipment volume at a single point is small, it can easily lead to problems such as low vehicle loading rate, duplicate transportation, frequent vehicle arrivals at the factory, and congestion at the receiving point.

[0033] The milk-run model, also known as the circular pickup model, refers to a system where the OEM or carrier arranges vehicles to visit one or more suppliers sequentially along a pre-set route, consolidating small batches of goods from all suppliers and transporting them to the OEM. This model reduces redundant transportation and empty runs caused by suppliers delivering goods separately, improves vehicle loading utilization, and reduces the frequency of OEM receiving vehicles. It is particularly suitable for parts suppliers that are geographically concentrated, have small individual shipment volumes, and stable demand patterns.

[0034] A Regional Distribution Center (RDC) is a transit, collection, sorting, and distribution node located between suppliers and OEMs. In automotive parts inbound logistics, the RDC can handle supplier collection, material storage, demand buffering, and subsequent distribution organization. For parts with large demand fluctuations, high frequency of direct deliveries to the OEM, high on-site receiving pressure at the OEM, or suitable for centralized management, suppliers can first deliver to the RDC, which then organizes subsequent distribution according to the OEM's production plan. If multiple suppliers are located in close proximity, a cyclical pickup method can be used to consolidate goods and deliver them to the RDC, which then distributes them to the OEM.

[0035] This invention mainly includes the following steps: 1. Collect data on component demand, packaging, suppliers, vehicles, OEMs, RDCs, and transportation costs; 2. Convert the demand for bare components into the demand for transport packaging units; 3. Construct feature vectors for component logistics and determine the feasibility of the direct delivery model; 4. Generate direct delivery candidate solutions for different vehicle models; 5. Generate candidate routes for cyclic pickup with the OEM and RDC as the final destinations, respectively; 6. Verify the running time, component efficiency, and loading capacity of the candidate routes; 7. Confirm the final feasibility of the circular pickup model based on the candidate routes; 8. Establish and solve a multi-mode collaborative optimization model; 9. Output the parts logistics model, vehicle usage, cyclical pickup routes, loading details, and cost results.

[0036] To highlight the collaborative optimization of parts logistics mode allocation, direct delivery vehicle selection, and cyclic pickup route, this invention adopts the following modeling boundaries.

[0037] First, each individual transport packaging unit of various components can accommodate any candidate vehicle model, and all supplier loading and unloading sites allow access to candidate vehicle models. Therefore, no 0-1 compatibility parameters are set between packaging and vehicle model. Differences in loading capacity between different vehicle models are reflected through effective load, effective volume, effective floor area, number of stacking layers, and number of transport trips.

[0038] Second, all components can, in principle, participate in the generation of candidate routes for cyclic pickup to the OEM, without setting a separate parameter for "whether business rules allow cyclic pickup". The feasibility of the cyclic pickup mode is determined by the existence of candidate routes that meet the requirements of supplier access and logistics timeliness; whether the route can carry the corresponding goods is determined uniformly by subsequent loading constraints.

[0039] Third, for parts transiting through RDC, only it is determined whether RDC can receive and process their packaging, and the RDC processing and subsequent delivery time and cost are calculated; no optimization is made for RDC cycle throughput or buffer capacity.

[0040] Fourth, this invention does not impose constraints on the number of vehicles, vehicle working hours, vehicle shifts, or maximum number of trips per vehicle type. It assumes that the transportation needs of the optimized plan can be met within the planning period using the company's own vehicles, carrier vehicles, or temporary transportation resources. Vehicle type is only used to reflect differences in single-trip transportation capacity and transportation costs.

[0041] Please see Figure 1 The multi-mode collaborative allocation and path scheduling method of the present invention specifically includes the following steps, namely S1 to S7.

[0042] S1. Obtain basic data related to the logistics of automotive parts entering the factory, and convert the original demand for various parts into the demand for standard transport packaging units.

[0043] The system first collects and reads the OEM's production plan, bill of materials, supplier information, parts demand information, packaging attributes, vehicle parameters, RDC parameters, transportation cost parameters, and node distance and running time data.

[0044] The collected data mainly includes: part number, name and supplier; original and historical demand for parts in the planning cycle; quantity per package; type of transport packaging unit, length, width and height of the package, weight of the package and maximum number of stacking layers; spatial location of suppliers, OEMs and RDCs; part preparation completion time and latest arrival time; vehicle vehicle effective load capacity, effective volume, cargo box size and effective floor area; vehicle vehicle fixed start-up cost and unit mileage cost; types of transport packaging units that the RDC can accept, loading and unloading equipment, processing time, unit processing cost and subsequent delivery time.

[0045] The standardization preprocessing in step S1 includes the following four aspects (1.1~1.4).

[0046] 1.1 Calculation of Transport Packaging Unit Requirement The demand and loading quantities in the model are uniformly measured in transport packaging units.

[0047] The original demand for components is converted into the demand for transport packaging units: ; In the formula, This indicates the converted demand for transport packaging units of the parts, subscript. Indicates the part number; Indicates components The original demand quantity within the planning period; Indicates parts Corresponding single packaging The number of parts it can hold; This indicates rounding up to the nearest integer.

[0048] 1.2 Packaging Attribute Calculation The theoretical volume of a single transport packaging unit is: ; In the formula, Represents a single transport packaging unit Theoretical volume; , , These represent the length, width, and height of the packaging unit, respectively.

[0049] The base area of ​​a single transport packaging unit is: ; In the formula, Represents a single transport packaging unit The base area.

[0050] The total transport weight and theoretical total volume of the components during the planned period are respectively and .in, Indicates components Corresponding packaging unit The weight.

[0051] 1.3 Distance and Running Time Processing The system maps OEMs, RDCs, and suppliers as spatial nodes and obtains the road distances and estimated travel times between nodes based on node coordinates, enterprise transportation management systems, geographic information systems, or navigation interfaces.

[0052] In a simulation environment, the Euclidean distance can also be calculated based on the spatial coordinates of the nodes, and then converted into a simulated road distance using a road correction factor.

[0053] 1.4 Calculation of Effective Stacking Layers Although all types of packaging can fit all candidate vehicle models, the height of the vehicle compartments varies between models, and the number of stacking layers that can be achieved may differ.

[0054] Packaging unit In the model Effective stacking layers for: ; in, This indicates rounding down. This parameter only reflects the stackability of the package across different vehicle models and is not used to determine whether a single package can fit into a vehicle. Indicates packaging unit Maximum number of stacking layers allowed; Indicates vehicle model The effective height of the carriage.

[0055] S2. Based on the components as the decision-making objects, construct a logistics feature vector containing multi-dimensional attributes of the components. Based on the logistics timeliness requirements and operational adaptation conditions in the logistics feature vector, determine the feasibility of direct delivery of various components using the direct delivery mode.

[0056] Please see Figure 2 Step S2 specifically includes the following processes, namely S21 to S24.

[0057] S21. Extract the demand for transport packaging units of components, the weight of a single package, the length, width and height of a single package, the theoretical volume of a single package, the bottom area of ​​a single package, the historical demand fluctuation coefficient, the stockout risk level, the spatial clustering degree of suppliers, the maximum allowable logistics time, and the adaptation parameters to the operating conditions of the regional distribution center, and construct the logistics feature vector, expressed as: ; In the formula, Indicates components The logistics feature vector; Indicates the historical demand volatility coefficient; Indicates the level of risk of stockout; Indicates the degree of spatial agglomeration of suppliers with the OEM as the endpoint, superscript Represents the OEM node; Indicates the degree of spatial clustering of suppliers with RDC as the endpoint, superscript Represents an RDC node; Indicates the maximum allowed logistics time; This indicates the adaptation parameters for RDC operation conditions.

[0058] The demand volatility coefficient can be calculated based on the ratio of the historical demand standard deviation to the average value; the stockout risk level can be determined based on the criticality of the components and the maximum allowable logistics time; the spatial agglomeration degree can be calculated based on the number of other suppliers within a certain range around the supplier, their directional consistency, and the degree of detour.

[0059] Demand fluctuations, stockout risks, spatial clustering, and demand volume are primarily used to determine the order of candidate solution generation, supplier portfolio priority, the number of candidate routes to retain, and to rank the candidate routes. These characteristics are not used as the sole exclusion criteria to avoid prematurely eliminating potentially superior solutions due to overly stringent fixed thresholds.

[0060] Let the physical mode set be... for: ; in, This indicates direct delivery to the OEM (Original Equipment Manufacturer). This indicates that goods are repeatedly picked up and delivered to the OEM. This indicates direct delivery to the RDC. This indicates that goods will be picked up repeatedly and delivered to the RDC.

[0061] S22. Calculate the logistics time for direct delivery of parts to the OEM. If the logistics time is not greater than the maximum allowable logistics time for the part, then the feasibility of direct delivery of the part to the OEM is confirmed.

[0062] In this embodiment, the feasibility expression for the direct delivery to the OEM mode is: ; The logistics time for direct delivery to the OEM includes the time required for the vehicle to arrive at the supplier, loading, transportation from the supplier to the OEM, and unloading at the factory.

[0063] In the formula, Indicates components Feasibility parameters for using Mode A (direct to OEM) are superscripted. Corresponding mode type; This indicates the logistics time for direct delivery to the OEM.

[0064] S23. Determine whether the regional distribution center can receive and process the packaging of parts; if so, determine that the operating condition adaptation parameters meet the requirements.

[0065] RDC adaptation parameters are: ; The operating condition adaptation parameters of the regional distribution center are determined based on the type, length, width, height, and weight of the corresponding transport packaging unit of the parts, as well as the receiving and loading / unloading conditions of the regional distribution center. When the type, size, and weight of the transport packaging unit meet the receiving and loading / unloading conditions of the regional distribution center, the operating condition adaptation parameter is determined to be 1; otherwise, it is determined to be 0.

[0066] S24. Calculate the total logistics time of the parts via the regional distribution center. If the operating condition adaptation parameters meet the requirements and the total logistics time is not greater than the maximum allowable logistics time of the parts, then the feasibility of the direct delivery mode of the parts to the regional distribution center is determined.

[0067] In this embodiment, the feasibility expression for the direct RDC mode is: ; In the formula, Indicates the feasibility parameters for using Mode C (direct to RDC), superscript. Corresponding mode type; This indicates the total logistics time via RDC. Among them, This includes the transportation time from the supplier to the RDC, the RDC receiving and processing time, and the subsequent delivery time from the RDC to the OEM.

[0068] S3. For components that are feasible for direct delivery, calculate the required number of transport trips based on the demand for their corresponding transport packaging units and the loading parameters of the candidate vehicle models, and generate corresponding direct delivery candidate schemes (specifically, 3.1~3.2).

[0069] 3.1 Candidate solutions directly sent to OEMs For satisfying For each candidate vehicle model, the system generates a solution that is directly sent to the OEM.

[0070] The number of transport trips required for direct delivery is: ; In the formula, For parts Using candidate models The number of transport trips required for direct delivery; For parts The demand for transport packaging units; For parts The corresponding single transport packaging unit The weight; Candidate models Effective load capacity; For parts The corresponding single transport packaging unit The equivalent volume factor of the packaging; For parts The corresponding single transport packaging unit Theoretical volume; Candidate models Effective volume; For parts The corresponding single transport packaging unit The base area; For parts The corresponding single transport packaging unit In candidate models The effective stacking layer; Candidate models The effective base area.

[0071] The formula calculates the required number of transport trips from three dimensions: load capacity, equivalent volume, and equivalent base area, and takes the maximum value among the three.

[0072] The cost of the direct delivery solution to the OEM is: ; If the company uses round-trip mileage for billing, then Includes the distance from the supplier to the OEM and the return trip; if a one-way transportation quote is used, the return trip cost is considered to be included in the fixed activation cost or unit mileage cost.

[0073] 3.2 Candidate solutions for direct delivery to RDC For satisfying For the components, the system generates direct delivery candidate solutions to the RDC for each vehicle model, with the following front-end transportation costs: ; In the formula, Indicates components Vehicle Model Front-end transportation costs for direct delivery to RDC; This indicates the billing mileage for a single trip directly to the RDC.

[0074] This cost only includes the initial transportation cost from the supplier to the RDC; the RDC processing and subsequent delivery costs are calculated uniformly in the comprehensive objective function.

[0075] S4. Using each supplier as a candidate access node, generate an initial cyclic pickup route that sequentially visits at least one candidate access node and ends at the OEM or regional distribution center, based on the spatial location information of each supplier (specifically, it may include 4.1~4.2).

[0076] 4.1 Circular pickup to OEM candidate route The system uses all suppliers as candidate nodes and generates a set of candidate route instances with the OEM as the final destination, based on the supplier's spatial location, component demand, spatial clustering, route direction, and maximum allowable detour distance. .

[0077] When vehicles are centrally dispatched by the OEM, the route adopts a closed-loop structure: ; In the formula, Indicates the OEM node; Indicates the number of visits in sequence To the There are 10 candidate access nodes (i.e., supplier nodes).

[0078] 4.2 Circular pickup to RDC candidate routes For suppliers with RDC-compatible components, the system generates a set of candidate route instances whose endpoint is RDC. .

[0079] When vehicles are centrally dispatched by the RDC, the route adopts a closed-loop structure: ; In the formula, This indicates a Regional Distribution Center (RDC) node.

[0080] The complete set of candidate route instances is: ; In the formula, This represents the set of all candidate circular pickup route instances.

[0081] If the company's vehicles actually depart from the carrier's base, the route departure node can be replaced with the carrier's base, and all actual billing segments can be included in the route distance.

[0082] Each candidate route instance represents an independent transport execution. When the same supplier access sequence needs to be executed multiple times within the planning period, multiple candidate route instances with different numbers can be generated.

[0083] Any candidate route instance is represented as: ; In the formula, Indicates the number of the candidate route instance; Indicates the starting node of the route, subscript All of these indicate that the parameter belongs to the route. ; Indicates the order in which suppliers are accessed; Indicates the final delivery node; This indicates the vehicle model that is being linked.

[0084] The order of access to its suppliers is represented as follows: ; In the formula, These represent the first to nth supplier nodes visited sequentially by route r.

[0085] The total distance of the route is: ; In the formula, This represents the total distance traveled on route r; This represents the distance parameter between two corresponding spatial nodes.

[0086] The route cost is: ; In the formula, Indicates the execution cost of the route; Indicates the vehicle model that is bound. The fixed cost per activation; Indicates the vehicle model that is bound. The cost per unit mileage.

[0087] S5. Verify the running time, actual logistics timeliness, and multi-dimensional loading capacity of the initial circular pickup route, and determine the feasibility of circular pickup using the circular pickup mode for various parts, and obtain the verified circular pickup route.

[0088] Please see Figure 3 Step S5 specifically includes the following processes, namely S51~S5: S51. Calculate the total running time of the initial cyclic pickup route, and remove the initial cyclic pickup routes whose total running time exceeds the preset maximum allowable running time.

[0089] If the total running time of candidate route r satisfy If so, the system will directly delete the route instance. This is the maximum allowable running time for a single circular pickup route. It is only used to control the overall running time of a single route and is not used to replace the logistics timeliness requirements of various components themselves.

[0090] S52. For the retained initial cyclic pickup route, calculate the actual logistics time from the completion of pickup to the arrival of the destination for each part based on the actual access order of each candidate access node.

[0091] The system calculates the actual logistics time for parts from pickup by the supplier to arrival at the OEM or RDC based on the supplier's location along the route. .

[0092] Parts delivery time Not equal to the total route time When the supplier is located at the end of the route, the time from pickup to the destination for their parts is usually shorter.

[0093] S53. Construct comprehensive adaptation parameters for parts and routes. When the initial cyclic pickup route meets all of the following conditions, the comprehensive adaptation parameters are determined to be valid: C1. The route passes through the supplier to which the part belongs; C2. The part has been prepared when the vehicle arrives at the supplier; C3. The latest arrival time requirement is met; C4. The actual logistics time is not greater than the maximum allowable logistics time; C5. When the route destination is a regional distribution center, the operational condition adaptation parameters meet the requirements.

[0094] The comprehensive compatibility parameters of components and routes are expressed as follows: ; when At the same time, the following conditions must be met: route After parts The supplier is responsible for the vehicle's availability; the parts are ready when the vehicle arrives at the supplier; the parts can be delivered before the latest arrival time. When the route terminus is RDC, the following conditions are met. .

[0095] Whether a vehicle's load capacity, volume, and floor area can accommodate all the goods on the route is not a factor. Instead of making repeated judgments in the model, we will make unified judgments in the subsequent optimization model.

[0096] If a certain route does not meet the timeliness or RDC compatibility requirements of a certain type of component, then only the corresponding... Instead of deleting the entire route directly.

[0097] S54. If there exists a cyclic pickup route with the OEM as the endpoint, and the overall adaptation parameters are met, then the cyclic pickup of the component using the OEM mode is deemed feasible; if there exists a cyclic pickup route with the regional distribution center as the endpoint, then the cyclic pickup of the component using the regional distribution center mode is deemed feasible.

[0098] The feasibility of a cyclical pickup model to the OEM is as follows: ; In the formula, This indicates the feasibility parameters for using Mode B (cyclic pickup to the OEM), indicated by the superscript. Corresponding mode type.

[0099] The feasibility of the cyclical pickup to RDC model is as follows: ; In the formula, Indicates the feasibility parameters for using Mode D (cyclic pickup to RDC), superscript. Corresponding mode type.

[0100] Whether the cyclic pickup mode is feasible depends directly on whether there are candidate route instances that can serve the part, and no preliminary feasibility parameters for cyclic pickup are set.

[0101] S6. Combining the aforementioned direct delivery candidate schemes and the verified cyclic pickup candidate routes, a multi-mode collaborative optimization model is constructed with the goal of minimizing the overall inbound logistics cost within the planned period.

[0102] Please see Figure 4The multi-mode collaborative optimization model simultaneously determines: the logistics mode adopted for each type of component; the vehicle model used for direct-delivery components; the instances of activated circular pickup routes; and the number of component packages loaded on each route.

[0103] In step S6, the multi-mode collaborative optimization model aims to minimize the overall cost of inbound logistics within the planning period, and the objective function is: ; In the formula, This refers to the total cost of inbound logistics within the planned period. For parts Vehicle Model The transportation cost of direct delivery to the OEM; To indicate components Whether to use vehicle model 0-1 variables directly sent to the OEM; For parts Vehicle Model Front-end transportation costs for direct delivery to regional distribution centers; To indicate components Whether to use vehicle model 0-1 variables for direct delivery to regional distribution centers; Candidate route instances Execution costs; To represent candidate route instances Whether a 0-1 variable is enabled; For parts Unit cost of processing and subsequent delivery through regional distribution centers; For parts The demand for transport packaging units; To indicate components A 0-1 variable indicating whether to select the direct delivery to regional distribution center mode; To indicate components The 0-1 variable indicating whether to select the cyclic pickup to regional distribution center mode.

[0104] Note: Superscript symbols in the formula ( ) respectively refer to the pattern determined in the preceding text, and the subscript ( () respectively refer to components, candidate vehicle models, and route examples.

[0105] In the objective function above, the first term is the transportation cost of direct delivery to the OEM, the second term is the transportation cost of direct delivery to the RDC, the third term is the cost of the circular pickup route, and the fourth term is the cost of RDC processing and subsequent delivery.

[0106] In step S6, the constraints of the multi-mode collaborative optimization model include: (1) The unique selection constraint of the pattern is expressed as: ; In the formula, Indicates components Select logistics mode 0-1 variables; subscript Indicates the specific logistics mode category; This represents the collection of all components. Each type of component can only use one logistics mode within the planning period.

[0107] (2) Model feasibility constraints, expressed as: ; In the formula, Indicates components Adopting the pattern Feasibility parameters (corresponding to those calculated above) (etc.). The model can only select logistics models that have undergone feasibility assessment.

[0108] The model can only select logistics modes that have been assessed for feasibility.

[0109] (3) Direct delivery vehicle selection constraint, expressed as: ; ; In the formula, Represents the set of candidate vehicle models; Indicates components Select the 0-1 variable for the direct delivery to the OEM mode; Indicates components Select 0-1 variables for direct RDC mode.

[0110] When choosing between direct delivery to the OEM or direct delivery to the RDC for parts, only one vehicle model can be selected.

[0111] Since this invention assumes that all types of transport packaging units can be fitted with any candidate vehicle type, no direct delivery vehicle type compatibility constraints are set. The loading capacity and transportation costs of different vehicle types are determined by… , and reflect.

[0112] (4) The requirement to pick up goods cyclically to the OEM to meet the constraint is expressed as: ; In the formula, Indicates that route r loads components. The number of packaging units; Indicates components Select 0-1 variables for the cyclic pickup to OEM mode.

[0113] (5) The requirement for cyclic pickup to RDC satisfies the constraint, as shown in: ; In the formula, Indicates components Select 0-1 variables for the cyclic pickup to RDC mode.

[0114] The same component can be transported by multiple routes under the same cyclical pickup mode, but multiple logistics modes cannot be used at the same time.

[0115] (6) The constraint relating components to the route is expressed as: ; This constraint ensures that components can only be loaded onto routes that pass through their respective suppliers, meet their logistics timeliness requirements, and have already been activated.

[0116] (7) Route load constraints, expressed as: ; In the formula, This indicates the vehicle type associated with route r. Effective load capacity.

[0117] (8) The equivalent volume constraint of the route is expressed as: ; In the formula, Indicates vehicle model The effective volume. The equivalent volume factor is used to correct for different packaging forms, gaps between packages, and loading space losses.

[0118] (9) Route equivalent base area constraint: route Vehicle model already bound ,therefore: ; The route's equivalent base area constraint is expressed as: ; In the formula, Indicates components The corresponding packaging unit is on the line The effective stacking layer; Indicates vehicle model The effective base area.

[0119] Load capacity, equivalent volume, and equivalent base area constraints jointly control the loading capacity of the cyclic pickup route.

[0120] (10) The running time constraint for a single route is expressed as: ; This constraint controls the overall running time of a single cyclic pickup route. The specific logistics time for different components is determined by... control.

[0121] If all exceed Since all routes have been deleted during the candidate route generation stage, this constraint does not need to be repeated in the optimization model.

[0122] The range of values ​​for the above variables: ; ; In the formula, Represents the set of integers greater than or equal to 0.

[0123] The model does not impose constraints on the number of vehicles, vehicle working hours, or the maximum number of trips per vehicle type. Since all transportation options have positive transportation costs, the objective function will not select redundant routes that are not loaded with goods or have no practical use, provided that the demand is met.

[0124] S7. Solve the multi-mode collaborative optimization model and output the final inbound logistics mode allocation scheme and the circular pickup route scheduling scheme corresponding to each component; wherein, the inbound logistics mode includes: direct delivery to the OEM, circular pickup to the OEM, direct delivery to the regional distribution center, and circular pickup to the regional distribution center; different components under the same supplier can be independently allocated to different inbound logistics modes according to the solution results.

[0125] The system inputs the above objective function and constraints into a mixed-integer programming solver to search for the lowest overall cost solution among all feasible combinations of logistics modes, vehicle types, and routes.

[0126] After the solution is completed, the system will obtain at least the following: the logistics mode selected for each type of component; the vehicle type and number of transport trips used for direct delivery of components; the activated circular pickup routes; the departure node, supplier access order and final delivery node for each route; the number of components and packages loaded on each route; and the total cost of the planning cycle and its composition.

[0127] In an alternative implementation, the system can represent both direct delivery and cyclic pickup loading schemes as a single candidate column.

[0128] Each candidate column must include at least: the logistics mode, vehicle type, corresponding route example, quantity of each component's transport packaging, and candidate column cost.

[0129] The objective function for the candidate column model is: ; In the formula, Represents the set of candidate columns; Indicates the index of the candidate column in the set; Indicates candidate columns The overall cost; Indicates candidate columns A 0-1 variable indicating whether it has been selected.

[0130] The unique selection constraint of the pattern is: ; The component quantity balance constraint is: ; In the formula, This indicates that it belongs to the logistics model. The set of candidate columns; Indicates candidate columns Transportation components The number of packaging units. This constraint allows the same component to be transported together by multiple candidate columns under the same mode, but prohibits the same component from being transported across different modes.

[0131] The feasibility constraints of the model are: ; The unique load combination constraint for the same route instance is: ; In the formula, Indicates candidate columns Does it correspond to a route instance? The parameter (1 if it is true, 0 otherwise).

[0132] The same route instance can generate multiple candidate columns for different loading combinations, but the model can only select one at most.

[0133] If the same geometric route is allowed to be executed multiple times within a planning period, the system pre-generates it as multiple route instances with different numbers.

[0134] The range of values ​​for the variable is: ; For direct candidate columns, there are usually: This indicates that the candidate column has completed all planned cycle requirements for the components.

[0135] For the candidate column of recurring pickup, there can be This allows multiple candidate columns with the same pattern to jointly fulfill all the requirements of the same component.

[0136] The candidate column model does not set constraints on the number of vehicles, vehicle working hours, maximum number of transport trips for vehicle type, and RDC cycle processing capacity. It mainly controls the number of parts, mode selection, and the uniqueness of loading schemes for the same route instance.

[0137] After the model is solved, the system outputs the following results: Parts logistics mode allocation table, direct delivery scheme vehicle type and transportation trip table, circular pickup route table, route parts loading details table, cost summary table, supplier spatial distribution map, direct delivery to OEM route map, direct delivery to RDC route map, circular pickup to OEM route map, circular pickup to RDC route map, multi-mode overall route map, and route weight, equivalent volume and floor area utilization rate map.

[0138] The parts logistics mode allocation table should include at least the parts number, the supplier, the original parts demand, the transport packaging unit demand, the transport packaging unit type, the recommended logistics mode, the selected vehicle type, the number of direct delivery trips, and the circular pickup route number.

[0139] The cyclic pickup route table should include at least the route number, departure node, final delivery node, vehicle type, supplier access order, route distance, route time, route cost, and number of packages loaded.

[0140] The route loading details list should include at least the route number, component number, supplier, quantity of loaded packages, loading weight, equivalent volume, and equivalent floor area.

[0141] When changes occur in production plans, component requirements, supplier status, packaging specifications, transportation distance, vehicle type transportation costs, or RDC processing costs, the system re-executes data preprocessing, logistics mode feasibility assessment, candidate solution generation, candidate route verification, and optimization solution to obtain an updated inbound logistics mode allocation and path scheduling scheme.

[0142] Example 2 This embodiment sets up one OEM node, one regional distribution center (RDC), and 11 supplier nodes, encompassing a total of 60 types of automotive parts. Based on the multi-mode collaborative allocation and route scheduling method described in Embodiment 1, four inbound logistics modes—direct delivery to the OEM, cyclic pickup to the OEM, direct delivery to the RDC, and cyclic pickup to the RDC—are incorporated into a unified optimization framework. The logistics mode and cyclic pickup route for each part are determined according to part demand, supplier location, packaging attributes, vehicle loading capacity, route travel time, RDC compatibility conditions, and transportation costs. The specific process includes steps S101 to S110.

[0143] S101: Read and preprocess basic data. The system first reads the spatial locations of 11 suppliers, the planned cycle requirements and packaging attributes of 60 types of parts, and simultaneously reads data on OEMs, RDCs, candidate vehicle models, and transportation costs. For each type of part, the demand is uniformly converted into transportation packaging unit requirements based on the original demand quantity and the capacity of a single transportation packaging unit; the theoretical volume, bottom area, and total transportation weight are calculated based on the length, width, height, and weight of the packaging and the weight of a single packaging unit; further, the effective stacking layers for different vehicle models are calculated by combining the effective height of the vehicle model and the maximum allowable stacking layers of the packaging. The system generates road distance and running time parameters between nodes based on the spatial locations of OEMs, RDCs, and suppliers, and converts the above data into standardized input parameters required for subsequent mode feasibility assessments and multi-mode optimization models.

[0144] S102: Constructing Parts Logistics Characteristics and Determining the Feasibility of Direct Delivery Mode. For 60 types of parts, the system constructs a parts logistics characteristic vector based on planned demand, historical demand fluctuations, packaging size, single package weight, stockout risk, and supplier spatial distribution. For the direct delivery to OEM mode, the system calculates the logistics time for each part to be transported directly from its supplier to the OEM and compares it with its maximum allowable logistics time; parts meeting the timeliness requirements are included in the direct delivery to OEM candidate solution set. For the direct delivery to RDC mode, the system further determines the compatibility between the parts and RDC operating conditions and calculates the total logistics time for supplier-to-RDC transportation, RDC processing, and subsequent delivery from RDC to the OEM. Only when both RDC compatibility and logistics timeliness requirements are met is this mode selected as an option for the parts. For parts meeting the direct delivery conditions, the system calculates the number of transport trips required to complete the planned cycle demand and the corresponding transportation costs for different candidate vehicle models.

[0145] S103: Generate candidate routes for cyclic pickup and perform timeliness verification. For the cyclic pickup mode, the system generates candidate routes with the OEM and RDC as the destination, respectively, based on the supplier's spatial location, the proximity between suppliers, the route direction, and the allowed detour range. Each candidate route corresponds to a specific route instance and is bound to the corresponding vehicle model. The system calculates the total travel distance, total running time, and route execution cost of the candidate routes, and deletes candidate routes that exceed the maximum allowed running time for a single route. For the retained candidate routes, the system further calculates the actual logistics time of the parts from pickup completion to the OEM or RDC destination based on the actual access order of each supplier in the route, and determines whether the route passes through the supplier to which the part belongs, whether the goods have been prepared at the time of pickup, whether the goods can be delivered before the latest arrival time, and whether the RDC adaptation conditions are met, thereby establishing a comprehensive adaptation relationship between the parts and the candidate routes.

[0146] S104: Establish and solve a multi-mode collaborative optimization model. After generating candidate schemes and routes, input feasible logistics modes for each component, direct delivery candidate vehicle models, and cyclic pickup candidate routes into the multi-mode collaborative optimization model. The model, according to the objective function and constraints described in Part III, simultaneously determines the logistics mode adopted for each type of component, the vehicle model selection under the direct delivery mode, the cyclic pickup routes to be activated, and the actual number of component packages transported on each route. The model aims to minimize the overall inbound logistics cost during the planning cycle and imposes constraints such as unique mode selection, mode feasibility, direct delivery vehicle model selection, cyclic pickup demand satisfaction, component-route association, vehicle load, equivalent volume, converted base area, and route travel time. After solving the problem using a mixed-integer programming solver, the multi-mode collaborative allocation and cyclic pickup path scheme of this embodiment is obtained.

[0147] S105: Obtain the allocation results for four logistics modes. After model solving, all 60 types of parts were assigned to a unique inbound logistics mode. Figure 5 As can be seen, the cyclic pickup to OEM mode undertakes the main component transportation task in this embodiment, covering 46 types of components, accounting for 76.67% of all components. The direct delivery to OEM, direct delivery to RDC, and cyclic pickup to RDC modes respectively handle the remaining component demand, indicating that the model does not pre-fix a certain logistics mode, but automatically determines the logistics organization method for different components based on component characteristics, route conditions, and overall transportation costs.

[0148] S106: Further analyze the logistics structure according to the suppliers of the components, as shown in the following figures. Figure 6 As shown, among the 11 suppliers, the optimal logistics modes for the components vary significantly. For example, components from suppliers S06, S09, S10, and S11 are all delivered to the OEM in a cyclical pickup manner; suppliers S05, S07, and S08 employ both direct delivery to the OEM and cyclical pickup to the OEM; suppliers S01 to S04, in addition to cyclical pickup to the OEM, also have components delivered directly to the RDC or cyclical pickup to the RDC. Even within the same supplier, different components can be allocated to different logistics modes based on their own needs, packaging, and timeliness attributes, thus reflecting the characteristic of this invention that uses "components" rather than simply "suppliers" as the basic decision-making object.

[0149] S107: After completing the logistics mode allocation, the model ultimately activates 12 circular pickup routes, such as... Figure 7 As shown. Of these, 10 routes terminate at the OEM (Original Equipment Manufacturer). Figure 7 In the legend, P), the endpoint of both routes is RDC (…). Figure 7(See the legend "R" in the diagram). Routes R0047 and R0049 terminate at the RDC, while the remaining routes terminate at the OEM. The final selected circular pickup routes primarily use K2 and K3 vehicle types, with K2 used in most routes and K3 for some transport tasks. Different routes form different access combinations based on the spatial location of suppliers. The system can merge small batches of parts from multiple suppliers onto the same transport route based on node spatial location and parts transport needs, reducing redundant transport caused by scattered direct delivery while meeting parts logistics timeliness requirements.

[0150] S108: Verify the multi-dimensional loading capacity of the circular pickup routes. To verify the actual loading feasibility of the selected routes, the load utilization rate, equivalent volume utilization rate, and equivalent floor area utilization rate of the 12 circular pickup routes were statistically analyzed. The results are as follows: Figure 8 As shown in the figure, the load utilization rate of each selected route is approximately 38%–90%, the equivalent volume utilization rate is approximately 50%–82%, and the overall utilization rate of the equivalent floor area is at a high level, with most routes exceeding 90%, and some routes reaching 100% of the equivalent floor area utilization. The results indicate that using only a single weight constraint is insufficient to reflect the actual loading status of vehicles transporting automotive parts.

[0151] S109: Compare the overall logistics costs before and after optimization. To further verify the effectiveness of the multi-mode collaborative optimization scheme, this embodiment uses "all 60 types of parts are delivered directly to the OEM" as a benchmark, and calculates the benchmark transportation cost under the same planning cycle and cost parameters. Cost comparison before and after optimization is as follows: Figure 9 As shown in the figure. The results indicate that by simultaneously comparing four logistics modes—direct delivery to the OEM, cyclic pickup to the OEM, direct delivery to the RDC, and cyclic pickup to the RDC—and by uniformly optimizing the selection of logistics mode, direct delivery vehicle type, cyclic pickup route, and vehicle loading capacity, the low loading rate and repetitive transportation costs resulting from using decentralized direct delivery can be reduced, thereby lowering the overall inbound logistics cost within the planning period.

[0152] S110: Output the final optimization results. After the model is solved, the system generates the following results: component logistics mode allocation, direct delivery vehicle type and transportation trips, cyclic pickup route results, route component loading details, and planned cycle comprehensive cost results. In this embodiment, 60 types of components are ultimately allocated to four inbound logistics modes: 5 types are directly delivered to the OEM, 46 types are cyclically picked up to the OEM, 6 types are directly delivered to the RDC, and 3 types are cyclically picked up to the RDC. A total of 12 cyclic pickup routes are activated, 10 of which terminate at the OEM and 2 at the RDC. Each route meets the set requirements for load capacity, equivalent volume, converted floor area, and logistics timeliness. Compared with the baseline scheme of all direct delivery to the OEM, the multi-mode collaborative optimization scheme reduces the planned cycle comprehensive logistics cost from 29,073 yuan to 15,182 yuan, a cost reduction of 47.8%.

[0153] When production plans, component demand, supplier status, vehicle transportation costs, road travel time, or RDC processing costs change, the system can reread the updated input data and repeatedly execute the process of mode feasibility judgment, candidate route generation, route loading and timeliness verification, and multi-mode collaborative optimization solution to obtain an updated component inbound logistics mode allocation and route scheduling scheme.

[0154] Example 3 This embodiment provides a computer terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the multi-mode cooperative allocation and path scheduling method as described in Embodiment 1.

[0155] like Figure 10 As shown, the computer terminal provided in this embodiment includes: at least one processor 101, and a memory 102 connected to at least one processor 101. This embodiment does not limit the specific connection medium between the processor 101 and the memory 102. Figure 10 The example shown is the connection between processor 101 and memory 102 via bus 100. Bus 100 is... Figure 10 The connections between other components are shown in bold lines and are for illustrative purposes only, not as limiting information. Bus 100 can be divided into address bus, data bus, control bus, etc., for ease of representation. Figure 10 The bus is represented by a single thick line, but this does not indicate that there is only one bus or one type of bus. Alternatively, the processor 101 may also be called a controller; there is no restriction on the name.

[0156] In this embodiment, the memory 102 stores instructions that can be executed by at least one processor 101. The at least one processor 101 can execute the aforementioned method by executing the instructions stored in the memory 102.

[0157] The processor 101 is the control center of the device. It can connect to various parts of the control device through various interfaces and lines. By running or executing instructions stored in memory 102 and calling data stored in memory 102, the processor can perform various functions and process data, thereby monitoring the device as a whole.

[0158] In one possible design, processor 101 may include one or more processing units. Processor 101 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into processor 101. In some embodiments, processor 101 and memory 102 may be implemented on the same chip; in some embodiments, they may also be implemented on separate chips.

[0159] Processor 101 can be a general-purpose processor, such as a central processing unit (CPU), digital signal processor, application-specific integrated circuit, field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the multi-mode cooperative allocation and path scheduling method disclosed in Embodiment 1 can be directly manifested as execution by the hardware processor, or executed by a combination of hardware and software modules in processor 101.

[0160] Memory 102, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory 102 may include at least one type of storage medium, such as flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic storage, magnetic disk, optical disk, etc. Memory 102 can be any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto. In this embodiment, memory 102 can also be a circuit or any other device capable of implementing storage functions for storing program instructions and / or data.

[0161] By designing and programming the processor 101, the code corresponding to the multi-mode cooperative allocation and path scheduling method described in the foregoing embodiments can be embedded into the chip, thereby enabling the chip to execute the code during runtime. Figure 1 The steps of the multi-mode cooperative allocation and path scheduling method are shown. How to design and program the processor 101 is a technique well-known to those skilled in the art and will not be described further here.

[0162] Example 4 This embodiment provides a computer-readable storage medium storing a computer program thereon. When the program is executed by a processor, it implements the steps of the multi-mode cooperative allocation and path scheduling method as described in Embodiment 1.

[0163] The computer-readable storage medium may include flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the storage medium may be an internal storage unit of a computer device, such as the hard disk or memory of the computer device. In other embodiments, the storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, smart memory card, secure digital card, flash memory card, etc., provided on the computer device. Of course, the storage medium may include both internal storage units and external storage devices of the computer device. In this embodiment, the memory is typically used to store the operating system and various application software installed on the computer device. In addition, the memory can also be used to temporarily store various types of data that have been output or will be output.

[0164] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A multi-strategy collaborative allocation and path scheduling method for inbound logistics of automotive parts, characterized in that, Includes the following steps: S1. Obtain basic data related to the logistics of automotive parts entering the factory, and convert the original demand for various parts into the demand for standard transport packaging units; S2. Based on the components as the decision-making object, construct a logistics feature vector containing multi-dimensional attributes of the components. Based on the logistics timeliness requirements and operational adaptation conditions in the logistics feature vector, determine the feasibility of direct delivery of various components using the direct delivery mode. S3. For components that are feasible for direct delivery, calculate the required number of transport trips and generate corresponding direct delivery candidate schemes based on the demand for their corresponding transport packaging units and the loading parameters of the candidate vehicle models. S4. Using each supplier as a candidate access node, generate an initial cyclic pickup route that sequentially visits at least one candidate access node and ends at the OEM or regional distribution center, based on the spatial location information of each supplier. S5. Verify the running time, actual logistics timeliness and multi-dimensional loading capacity of the initial circular pickup route, and determine the feasibility of circular pickup mode for various parts, and obtain the verified circular pickup route. S6. Combining the aforementioned direct delivery candidate schemes and the verified cyclic pickup candidate routes, a multi-mode collaborative optimization model is constructed with the goal of minimizing the overall inbound logistics cost within the planned period. S7. Solve the multi-mode collaborative optimization model and output the final inbound logistics mode allocation scheme and the circular pickup route scheduling scheme corresponding to each component; wherein, the inbound logistics mode includes: direct delivery to the OEM, circular pickup to the OEM, direct delivery to the regional distribution center, and circular pickup to the regional distribution center; different components under the same supplier can be independently allocated to different inbound logistics modes according to the solution results.

2. The multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics according to claim 1, characterized in that, Step S2 specifically includes the following processes: Extract the demand for transport packaging units of components, the weight of a single package, the length, width and height of a single package, the theoretical volume of a single package, the bottom area of ​​a single package, the historical demand fluctuation coefficient, the stockout risk level, the spatial clustering degree of suppliers, the maximum allowable logistics time, and the adaptation parameters to the operating conditions of the regional distribution center, and construct the logistics feature vector. Calculate the logistics time for direct delivery of parts to the OEM. If the logistics time is not greater than the maximum allowable logistics time for the part, then the feasibility of direct delivery of the part to the OEM is confirmed. Determine whether the regional distribution center can receive and process the packaging of parts; if so, determine that the operating condition adaptation parameters meet the requirements. Calculate the total logistics time of the parts via the regional distribution center. If the operating condition adaptation parameters meet the requirements and the total logistics time is not greater than the maximum allowable logistics time of the parts, then the feasibility of direct delivery of the parts to the regional distribution center is determined.

3. The multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics according to claim 2, characterized in that, Step S5 details The process includes the following: Calculate the total running time of the initial cyclic pickup route, and remove initial cyclic pickup routes whose total running time exceeds the preset maximum allowable running time; For the retained initial cyclic pickup route, the actual logistics time from the completion of pickup to arrival at the destination is calculated based on the actual access order of each candidate access node; Construct comprehensive adaptation parameters for components and routes. The comprehensive adaptation parameters are determined to be valid when the initial cyclic pickup route meets all of the following conditions: C1, the route passes through the supplier of the component; C2, the component has been prepared when the vehicle arrives at the supplier. C3. The latest arrival time requirement is met; C4. The actual logistics time is not greater than the maximum allowable logistics time; C5. When the route destination is a regional distribution center, the operating condition adaptation parameters meet the requirements. If there exists a cyclic pickup route with the OEM as the endpoint and the overall matching parameters are met, then the feasibility of using the cyclic pickup to OEM mode for this part is confirmed. If there exists a cyclic pickup route with the regional distribution center as the endpoint and the overall matching parameters are met, then the feasibility of using the cyclic pickup to regional distribution center mode for this part is confirmed.

4. The multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics according to claim 1, characterized in that, In step S6, the objective function of the multi-mode collaborative optimization model is: In the formula, This refers to the total cost of inbound logistics within the planned period. For parts Vehicle Model The transportation cost of direct delivery to the OEM; To indicate components Whether to use vehicle model 0-1 variables directly sent to the OEM; For parts Vehicle Model Front-end transportation costs for direct delivery to regional distribution centers; To indicate components Whether to use vehicle model 0-1 variables for direct delivery to regional distribution centers; Candidate route instances Execution costs; To represent candidate route instances Whether a 0-1 variable is enabled; For parts Unit cost of processing and subsequent delivery through regional distribution centers; For parts The demand for transport packaging units; To indicate components A 0-1 variable indicating whether to select the direct delivery to regional distribution center mode; To indicate components The 0-1 variable indicating whether to select the cyclic pickup to regional distribution center mode.

5. The multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics according to claim 4, characterized in that, In step S6, the constraints of the multi-mode collaborative optimization model include: The mode unique selection constraint is used to restrict each type of component to only one inbound logistics mode during the planning cycle. Pattern feasibility constraints are used to limit the model to select only from the set of patterns that have either direct delivery feasibility or cyclic pickup feasibility; The direct delivery vehicle selection constraint is used to restrict that when the parts are selected to be delivered directly to the OEM or to the regional distribution center, only one candidate vehicle must be selected. Demand satisfaction constraints are used to ensure that the number of component packages loaded on the enabled circular pickup route is equal to the total demand for that component under the corresponding circular pickup mode. The component-route association constraint is used to ensure that components can only be loaded into a loop pickup route that passes through its candidate access node, meets its logistics timeliness requirements, and has been activated. The comprehensive loading capacity constraint is used to ensure that the total amount of component packaging loaded on the cyclic pickup route does not exceed the load limit of each dimension of the vehicle model. The comprehensive loading capacity constraint specifically includes load constraint, equivalent volume constraint, and calculated bottom area constraint combined with the effective stacking layer number.

6. The multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics according to claim 1, characterized in that, In step S7, the multi-mode collaborative optimization model is solved using a mixed integer programming solver, or a candidate column substitution model is established for solving. The step of establishing a candidate column substitution model for solving the problem includes: representing the direct delivery candidate scheme and the verified cyclic pickup route as a unified candidate column, with each candidate column including the logistics mode, vehicle type, corresponding route instance, quantity of transport packaging for each component, and candidate column cost; setting component quantity balance constraints in the multi-mode collaborative optimization model to allow the same component to be transported by multiple candidate columns under the same inbound logistics mode, while prohibiting the same component from being transported across different inbound logistics modes.

7. The multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics according to claim 1, characterized in that, Step S3 details The process includes the following: For components that are feasible for direct delivery, the minimum number of transport trips required for each candidate vehicle model is calculated from three dimensions: vehicle effective load, equivalent volume, and equivalent floor area. The maximum value among the three calculation results is taken as the required number of transport trips, expressed by the following formula: In the formula, For parts Using candidate models The number of transport trips required for direct delivery; For parts The demand for transport packaging units; For parts The corresponding single transport packaging unit The weight; Candidate models Effective load capacity; For parts The corresponding single transport packaging unit The equivalent volume factor of the packaging; For parts The corresponding single transport packaging unit Theoretical volume; Candidate models Effective volume; For parts The corresponding single transport packaging unit The base area; For parts The corresponding single transport packaging unit In candidate models The effective stacking layer; Candidate models The effective base area; Based on the required number of transport trips, the fixed activation cost per trip for candidate vehicle models, and the cost per unit mileage, the corresponding direct delivery candidate schemes are calculated and generated.

8. The multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics according to claim 1, characterized in that, In step S1, obtaining basic data related to the logistics of automotive parts entering the factory includes obtaining distance and travel time data between nodes. Specific methods for obtaining this data include: The OEMs, regional distribution centers, and suppliers are uniformly mapped as spatial nodes; By calling the enterprise transportation management system, geographic information system, or navigation interface, the actual road distance and estimated travel time between the spatial nodes can be obtained.

9. A computer terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the multi-strategy collaborative allocation and path scheduling method for automotive parts inbound logistics as described in any one of claims 1 to 8.