Intelligent capacity subcontracting system driven by multi-dimensional data analysis

Abstract

The present invention discloses an intelligent capacity subcontracting system driven by multidimensional data analysis, which belongs to the technical field of capacity subcontracting, and specifically comprises: obtaining order information of orders to be produced in a target area, and addresses and capacity of all subcontractors in the target area; calculating the Euclidean distance between each order to be produced and all subcontractors, marking each subcontractor address as the initial centroid of clustering, and allocating each order to be produced to the initial centroid closest to generate a category cluster; calculating the virtual distance between each order to be produced and the initial centroid according to a duration coefficient and a quantity coefficient; and iterating the centroid of each category cluster according to the virtual distance, and judging the offset of the centroid in each iteration until the centroid offset is less than a preset threshold; obtaining the virtual distance between any order to be produced after clustering and the final centroid, allocating the capacity ratio of the order to be produced according to the virtual distance, and calculating the allocated capacity of the order to be produced according to the capacity ratio and the subcontractor capacity.

Description

Technical Field

[0001] The present invention relates to the technical field of capacity subcontracting, and in particular to an intelligent capacity subcontracting system driven by multidimensional data analysis. Background Art

[0002] Capacity subcontracting is a production management strategy that refers to the allocation of a company's production tasks to external suppliers or partners. This subcontracting model is usually based on specific production capabilities, technical expertise or cost-effectiveness considerations, and aims to optimize resource allocation, improve production efficiency and flexibility. Through capacity subcontracting, companies can focus on their core competencies while using external resources to meet production needs, especially in the face of order fluctuations or changes in market demand. Capacity subcontracting can more effectively adjust the scale and speed of production.

[0003] In the process of enterprise development, with the continuous expansion of market demand and the extension of business scope, enterprises often expand their product business to different regions. Especially in some remote areas, there is a strong demand for products. However, there is often a high transportation cost to transport products from the original factory site to these remote areas. Therefore, in order to reduce transportation costs and improve market response speed, enterprises usually consider establishing factories or production bases locally.

[0004] However, before building a factory locally, enterprises usually cooperate with local subcontractors. This cooperation model can make full use of the resources and production capacity of local subcontractors to quickly meet market demand while reducing the investment risk and operating costs of enterprises. However, when there are many orders for products and the production capacity of local subcontractors is limited, the allocation of production capacity becomes a complex problem that needs to be solved.

[0005] In this case, companies need to conduct reasonable capacity planning and resource allocation to ensure that orders can be delivered on time and with guaranteed quality. This may involve negotiating and cooperating with multiple subcontractors, or finding other solutions to increase capacity. In addition, companies also need to consider factors such as how to optimize supply chain management, improve production efficiency, and reduce operating costs. Summary of the invention

[0006] The purpose of the present invention is to provide an intelligent capacity subcontracting system driven by multidimensional data analysis to solve the following technical problems:

[0007] How to allocate production capacity when there are many orders for a product and the local subcontractor's production capacity is limited.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] Intelligent capacity subcontracting system driven by multi-dimensional data analysis, including:

[0010] A data statistics module is used to obtain order information of pending orders in the target area, and mark the number of pending orders as N. The order information includes address, delivery time T and product quantity O, as well as the address and production capacity P of all subcontractors in the target area, and mark the number of subcontractors as M;

[0011] The clustering initialization module is used to mark the address of the i-th order to be produced as (X oi ,Y oi ), mark the address of the j-th subcontractor as (X pj ,Y pj ), i∈[1,N], j∈[1,M], calculate the Euclidean distance L between each pending production order and all subcontractors in turn according to the Euclidean distance formula, mark each subcontractor address as the initial centroid of the cluster, and assign each pending production order to the initial centroid with the closest distance to generate a category cluster;

[0012] The clustering optimization module is used to normalize the delivery time and product quantity of all pending orders respectively, obtain the normalized time coefficient α and quantity coefficient β of each pending order, and calculate the virtual distance L' between each pending order and the initial centroid according to the time coefficient and quantity coefficient, L'=L(α+1)(2-β); and update the centroid of each category cluster according to the virtual distance, and judge the offset of the centroid in each iteration until the centroid offset is less than a preset threshold;

[0013] The capacity calculation module is used to obtain the virtual distance D between any pending production order and the final centroid after cluster optimization in any category cluster. i , according to the virtual distance D i Allocate the capacity ratio K of the production order to be produced i , , mark the subcontractor capacity of the category cluster where the production order is located as P j , then the allocated capacity p of the pending production order i =P j K i .

[0014] As a further solution of the present invention: the capacity calculation module is also used to count the total amount of all orders to be produced in each category cluster, compare the total amount with the capacity of the subcontractor in the category cluster, mark the subcontractor whose capacity is greater than the total amount as a redundant capacity subcontractor, and mark the subcontractor whose capacity is less than the total amount as a shortage subcontractor; for any category cluster where a redundant subcontractor is located, directly mark the allocated capacity of all orders to be produced as the number of products, and then mark the redundant capacity of each redundant capacity subcontractor as a redundant capacity pool; and redistribute the capacity in the redundant capacity pool to the orders to be produced of nearby shortage subcontractors according to the distance priority until the redundant capacity pool is fully allocated, and the distance priority is that the shortage subcontractor who is closer to the redundant capacity subcontractor is allocated first.

[0015] As a further solution of the present invention: in the capacity calculation module, the secondary allocation is specifically:

[0016] For any shortage subcontractor, its original capacity is P j , obtain the received redundant capacity and mark it as P k , then the capacity of the shortage subcontractor after secondary allocation is P j =P j +P k .

[0017] As a further solution of the present invention: in the capacity calculation module, after the secondary allocation, for any category cluster, the allocated capacity p is screened out. i Greater than the corresponding product quantity O i The pending production orders are marked as overcapacity orders, and the allocated capacity of overcapacity orders is changed from p i Replace with O i , and aggregate the excess capacity to generate an excess capacity pool;

[0018] Screening and allocation capacity p i Less than the corresponding product quantity O i The pending production orders are marked as under-capacity orders, and the capacity is allocated to the under-capacity orders three times in proportion from the excess capacity pool. If the allocated capacity of the under-capacity orders is still lower than the number of products after three allocations, the difference between the allocated capacity and the number of products will be supplemented by production from the original factory.

[0019] As a further solution of the present invention: in the clustering initial module, the calculation formula of the Euclidean distance L is:

[0020] ;

[0021] Among them, L ij Represents the distance from the address of the i-th pending production order to the address of the j-th subcontractor.

[0022] As a further solution of the present invention: in the cluster optimization module, the calculation formula of the duration coefficient α and the quantity coefficient β is:

[0023] ; ;

[0024] Where T i represents the delivery time of the i-th pending production order, T min Represents the minimum delivery time of all pending production orders, T max Represents the maximum delivery time of all pending production orders, where O i represents the number of products to be produced in the i-th order, O min Indicates the minimum quantity of all products to be produced, O max Indicates the maximum quantity of all products to be produced.

[0025] As a further solution of the present invention: in the data statistics module, the production capacity P represents the number of products produced per unit time.

[0026] As a further solution of the present invention: in the capacity calculation module, when the allocated capacity value p of any pending production order i If it is not an integer, then p i Round down.

[0027] Beneficial effects of the present invention:

[0028] The present invention aims to solve the complex capacity allocation problem faced by enterprises when expanding their business across regions. The order and subcontractor data are obtained through the data statistics module. In the clustering initial module, the Euclidean distance formula is used to calculate the distance between each order to be produced and all subcontractors, and the order is allocated to the nearest subcontractor. This step ensures that the order can be processed quickly, reducing transportation time and cost. In the clustering optimization module, the system normalizes the delivery time and product quantity of the order, and calculates the virtual distance to update the centroid of the iterative category cluster. In this way, the system continuously optimizes the matching degree between orders and subcontractors, improves the overall production efficiency, and ensures that each order can have a certain capacity allocation, and focuses on maintaining customer orders with a large number of products. In addition, when the system handles redundant capacity between subcontractors and within subcontractors, it further optimizes resource allocation and reduces waste through secondary and tertiary allocation mechanisms. This intelligent capacity subcontracting system provides enterprises with a flexible and efficient solution to help them maintain their leading position in a highly competitive market environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The present invention will be further described below in conjunction with the accompanying drawings.

[0030] Figure 1 It is a schematic diagram of a module of the present invention;

[0031] Figure 2 It is a schematic diagram of the process of clustering production orders in the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] See also Figure 1-Figure 2 As shown, the present invention is an intelligent capacity subcontracting system driven by multi-dimensional data analysis, comprising:

[0034] The data statistics module is one of the core components of the intelligent capacity subcontracting system. Its main function is to comprehensively collect and organize detailed information on all relevant orders and subcontractors in the target area. This module ensures that enterprises can obtain the latest and most accurate market dynamics and production resource status through efficient data capture and processing technology.

[0035] In the data statistics module, the production orders are first recorded in detail. Each order contains multiple key information, including the order's shipping address, expected delivery time (T), and product quantity (O). This information is crucial for subsequent order allocation and production planning. For example, address information can help the system calculate the distance between each subcontractor, thereby optimizing logistics costs; delivery time is related to the urgency and priority of the order; and the number of products directly affects the calculation of production capacity requirements.

[0036] At the same time, the data statistics module is also responsible for collecting detailed information on all potential subcontractors in the target area. This includes but is not limited to the specific location of the subcontractor, production capacity (P), etc. Among them, production capacity is an important indicator, which indicates the number of products that a subcontractor can produce per unit time. Knowing the production capacity of each subcontractor helps the system to plan resources more reasonably when allocating orders and avoid overcapacity or undercapacity. In addition, the number of subcontractors is also recorded and marked as M so that the system can manage and schedule them uniformly in the subsequent processing process.

[0037] The clustering initialization module is used to assign orders to manufacturers by clustering. We need to cluster based on the distance between the order address and the manufacturer address. The specific steps are as follows:

[0038] Calculate distance: First, you need to calculate the geographical distance between each order and each manufacturer.

[0039] Clustering method selection: Select an appropriate clustering algorithm, such as K-means or hierarchical clustering.

[0040] Allocate orders: Based on the clustering results, assign the orders to the nearest manufacturers.

[0041] The specific steps are:

[0042] Mark the address of the i-th order to be produced as (X oi ,Y oi ), mark the address of the j-th subcontractor as (X pj ,Y pj ), i∈[1,N], j∈[1,M], the Euclidean distance L between each pending production order and all subcontractors is calculated in turn according to the Euclidean distance formula. The calculation formula of the Euclidean distance L is:

[0043] ;

[0044] Among them, L ij Represents the distance from the address of the i-th pending production order to the address of the j-th subcontractor.

[0045] The present invention selects K-means algorithm for clustering. K-means is an iterative algorithm for allocating data points into K clusters so that the distance between data points in the clusters is minimized.

[0046] Mark each subcontractor address as the initial centroid of the cluster, and assign each pending production order to the nearest initial centroid to generate a category cluster;

[0047] The clustering optimization module is used to normalize the delivery time and product quantity of all pending production orders, and obtain the normalized time coefficient α and quantity coefficient β of each pending production order. The calculation formula of the time coefficient α and the quantity coefficient β is:

[0048] ; ;

[0049] Where T i represents the delivery time of the i-th pending production order, T min Represents the minimum delivery time of all pending production orders, T max Represents the maximum delivery time of all pending production orders, where O i represents the number of products to be produced in the i-th order, O min Indicates the minimum quantity of all products to be produced, O maxIndicates the maximum quantity of all products to be produced.

[0050] And the virtual distance L' between each order to be produced and the initial centroid is calculated according to the duration coefficient and the quantity coefficient, L'=L(α+1)(2-β); in the present invention, when the delivery time of the order is shorter, the production priority of the order is higher, and the corresponding distance to the initial centroid is closer, the initial Euclidean distance is adjusted by the normalized duration coefficient, and α+1 is used to prevent the virtual distance from being 0 due to the coefficient value; in the present invention, when the number of products in the order is more, the production priority of the order is higher, and the corresponding distance to the initial centroid is closer, the Euclidean distance is adjusted by the normalized quantity coefficient, because the higher the order quantity, the smaller the corresponding quantity coefficient β, so 2-β is used to convert the size relationship, and prevent the virtual distance from being 0 due to the coefficient value;

[0051] Iterating the centroid of each category cluster according to the virtual distance, and determining the offset of the centroid in each iteration until the centroid offset is less than a preset threshold;

[0052] The formula for calculating the centroid is:

[0053] ;

[0054] Where Ai represents the i-th order, μk represents the centroid, Ck is the set of orders assigned to the k-th centroid, and |Ck| is the size of the set. The K-means algorithm continuously optimizes the division of clusters and the location of cluster centroids in an iterative manner. In each iteration, the data points are first assigned to the nearest cluster based on the current cluster centroid, and then the mean of all data points in each cluster is recalculated as the new cluster centroid. This process is repeated until the stopping condition is met.

[0055] Capacity calculation module,The capacity calculation module is a vital part of the intelligent capacity subcontracting system. After the clustering optimization is completed, it is responsible for reasonably allocating the capacity ratio according to the virtual distance between each order to be produced and the centroid of the final category cluster, so as to ensure the maximization of production efficiency and resource utilization.

[0056] In the capacity calculation module, we first need to obtain the virtual distance D between any pending production order after clustering optimization and the centroid of the final category cluster. i This virtual distance is calculated by comprehensively considering the order's delivery time, product quantity and other relevant factors, and it reflects the degree of match between the order and the subcontractor.

[0057] Based on the virtual distance D i , the system will further calculate the capacity ratio K of the pending production order i , The capacity ratio is a key indicator that determines how much capacity an order should be allocated to. Specifically, if an order is closer to a subcontractor (i.e., the virtual distance is smaller), then the order is more likely to be allocated to this subcontractor, thereby obtaining a higher capacity ratio. Conversely, if the distance is farther, the capacity ratio may be lower.

[0058] Next, the subcontractor capacity of the category cluster where the production order is located is marked as P j Here P j It indicates the number of products that the subcontractor can produce per unit time. i Capacity P of subcontractors j Multiplying them together, we can get the specific allocated capacity p of the pending production order. i , that is, p i =P j K i .

[0059] However, in the actual production process, due to various reasons (such as equipment limitations, human resources, etc.), sometimes the allocated capacity value p i is not an integer. To solve this problem, the system will i This means that if the calculated p i If the value contains a decimal part, the decimal part will be discarded directly, and only the integer part will be retained as the final allocated capacity. This is done to ensure the feasibility of capacity allocation and the simplicity of actual operation.

[0060] The capacity calculation module in the intelligent capacity subcontracting system is not only responsible for allocating capacity ratios based on virtual distances, but also bears the important responsibility of statistics and resource optimization. Through detailed data analysis, this module ensures that the production orders in each category cluster receive reasonable capacity allocation and effectively manages redundant and short capacity, thereby improving overall production efficiency and resource utilization.

[0061] First, the capacity calculation module counts the total number of all pending orders in each category cluster. This step is crucial because it provides basic data for subsequent capacity comparisons. By summarizing the product quantities of all orders, the system can clearly understand the actual demand scale of each category cluster.

[0062] Next, this total is compared with the subcontractor capacity of the category cluster. If a subcontractor's capacity is greater than the total order volume of its category cluster, then the subcontractor is marked as a redundant capacity subcontractor; conversely, if the subcontractor's capacity is less than the total order volume, it is marked as a shortage subcontractor. This classification helps the system quickly identify which subcontractors have sufficient production capacity to handle additional orders and which subcontractors may need external support to meet current production needs.

[0063] For any redundant subcontractor's category cluster, the system will directly mark the allocated capacity of all pending production orders as product quantity. This means that these orders will be fully supported by capacity without further adjustments. Subsequently, each redundant capacity subcontractor marks its unused capacity (i.e. redundant capacity) as a redundant capacity pool. The capacity in this pool can be regarded as a reserve resource to deal with emergencies or optimize the overall capacity allocation.

[0064] In order to use these redundant capacities more efficiently, the system will re-allocate them to the production orders of nearby shortage subcontractors based on distance priority. Specifically, the system will give priority to the shortage subcontractors that are closest to the redundant capacity subcontractors. This geographical location-based allocation strategy helps reduce logistics costs, speed up delivery, and improve customer satisfaction.

[0065] The entire secondary allocation process continues until the redundant capacity is fully allocated. During this process, the system continuously updates the capacity status of each subcontractor to ensure that each allocation is based on the latest data. In addition, if there are still unmet order demands after secondary allocation, the system can also consider further optimizing supply chain management, such as adding new partners or adjusting existing production plans.

[0066] The secondary distribution is specifically:

[0067] For any shortage subcontractor, its original capacity is P j , obtain the received redundant capacity and mark it as P k , then the capacity of the shortage subcontractor after secondary allocation is P j =P j +P k .

[0068] After the secondary allocation, the capacity calculation module screens the production orders in any category cluster to find out those with allocated capacity p i Greater than the corresponding product quantity O i For such orders, the system will allocate too high a capacity p iAdjust to the actual required product quantity O i To ensure the rational use of resources. At the same time, the excess capacity (i.e. the original allocated capacity p i With the adjusted product quantity O i The difference between the two) is aggregated to form a surplus capacity pool. The capacity in this pool can be regarded as a flexible resource reserve for subsequent possible capacity demand or optimization.

[0069] Then, the system again selects the allocated capacity p i Less than the corresponding product quantity O i The system will allocate capacity three times in proportion from the excess capacity pool formed previously. This proportion-based allocation method aims to fairly redistribute excess capacity to meet the needs of all undercapacity orders as much as possible.

[0070] However, if after three allocations, the allocated capacity of some under-capacity orders is still lower than the product quantity required, this means that the current excess capacity cannot fully make up for the gap. In this case, the system will take further measures: the unmet part (that is, the difference between the allocated capacity and the product quantity) will be handed over to the original factory for production replenishment. This arrangement ensures that even in some special circumstances, orders can be completed on time to avoid production delays or customer dissatisfaction caused by insufficient capacity.

[0071] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0072] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0073] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. Intelligent capacity subcontracting system driven by multi-dimensional data analysis, characterized by: include: A data statistics module is used to obtain order information of pending orders in the target area, and mark the number of pending orders as N. The order information includes address, delivery time T and product quantity O, as well as the address and production capacity P of all subcontractors in the target area, and mark the number of subcontractors as M; The clustering initialization module is used to mark the address of the i-th order to be produced as (X oi ,Y oi ), mark the address of the j-th subcontractor as (X pj ,Y pj ), i∈[1,N], j∈[1,M], calculate the Euclidean distance L between each pending production order and all subcontractors in turn according to the Euclidean distance formula, mark each subcontractor address as the initial centroid of the cluster, and assign each pending production order to the initial centroid with the closest distance to generate a category cluster; The clustering optimization module is used to normalize the delivery time and product quantity of all pending orders respectively, obtain the normalized time coefficient α and quantity coefficient β of each pending order, and calculate the virtual distance L' between each pending order and the initial centroid according to the time coefficient and quantity coefficient, L'=L(α+1)(2-β); and update the centroid of each category cluster according to the virtual distance, and judge the offset of the centroid in each iteration until the centroid offset is less than a preset threshold; The capacity calculation module is used to obtain the virtual distance D between any pending production order and the final centroid after cluster optimization in any category cluster. i , according to the virtual distance D i Allocate the capacity ratio K of the production order to be produced i , , mark the subcontractor capacity of the category cluster where the production order is located as P j , then the allocated capacity p of the pending production order i =P j K i .

2. The intelligent capacity subcontracting system based on multi-dimensional data analysis and drive according to claim 1 is characterized in that: The capacity calculation module is also used to count the total amount of all orders to be produced in each category cluster, compare the total amount with the capacity of the subcontractors in the category cluster, mark the subcontractors whose capacity is greater than the total amount as redundant capacity subcontractors, and mark the subcontractors whose capacity is less than the total amount as shortage subcontractors; for the category cluster where any redundant subcontractor is located, directly mark the allocated capacity of all orders to be produced as the product quantity, and then mark the redundant capacity of each redundant capacity subcontractor as a redundant capacity pool; The capacity in the redundant capacity pool is redistributed to the production orders of nearby shortage subcontractors according to the distance priority until the redundant capacity pool is fully allocated. The distance priority is that the shortage subcontractor that is closer to the redundant capacity subcontractor will be allocated first.

3. The intelligent capacity subcontracting system based on multi-dimensional data analysis and drive according to claim 2 is characterized in that: In the capacity calculation module, the secondary allocation is specifically: For any shortage subcontractor, its original capacity is P j , obtain the received redundant capacity and mark it as P k , then the capacity of the shortage subcontractor after secondary allocation is P j =P j +P k .

4. The intelligent capacity subcontracting system based on multi-dimensional data analysis and drive according to claim 3 is characterized in that: In the capacity calculation module, after the secondary allocation, for any category cluster, the allocated capacity p is selected. i Greater than the corresponding product quantity O i The pending production orders are marked as overcapacity orders, and the allocated capacity of overcapacity orders is changed from p i Replace with O i , and aggregate the excess capacity to generate an excess capacity pool; Screening and allocation capacity p i Less than the corresponding product quantity O i The pending production orders are marked as under-capacity orders, and the capacity is allocated to the under-capacity orders three times in proportion from the excess capacity pool. If the allocated capacity of the under-capacity orders is still lower than the number of products after three allocations, the difference between the allocated capacity and the number of products will be supplemented by production from the original factory.

5. The intelligent capacity subcontracting system based on multi-dimensional data analysis and drive according to claim 1 is characterized in that: In the clustering initial module, the calculation formula of the Euclidean distance L is: ； Among them, L ij Represents the distance from the address of the i-th pending production order to the address of the j-th subcontractor.

6. The intelligent capacity subcontracting system based on multi-dimensional data analysis and drive according to claim 1 is characterized in that: In the cluster optimization module, the calculation formulas of the duration coefficient α and the quantity coefficient β are as follows: ； ； Where T i represents the delivery time of the i-th pending production order, T min Represents the minimum delivery time of all pending production orders, T max Represents the maximum delivery time of all pending production orders, where O i represents the number of products to be produced in the i-th order, O min Indicates the minimum quantity of all products to be produced, O max Indicates the maximum quantity of all products to be produced.

7. The intelligent capacity subcontracting system based on multi-dimensional data analysis and drive according to claim 1 is characterized in that: In the data statistics module, the production capacity P represents the number of products produced per unit time.

8. The intelligent capacity subcontracting system based on multi-dimensional data analysis and drive according to claim 1 is characterized in that: In the capacity calculation module, when the allocated capacity value p of any pending production order i If it is not an integer, then p i Round down.

Images