A real-time allocation method for suppliers in the industrial chain based on quality priority
By adopting a real-time allocation method that prioritizes quality in the industrial chain, optimizing supplier scoring and allocation model, the flexibility and stability of supplier allocation solutions in the existing technology are solved, and a more fair and efficient supplier selection is achieved. , suitable for many-to-many assignment scenarios.
Patent Information
- Application Number
- CN202410130509.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-31
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2044-01-31
AI Technical Summary
The existing industrial chain supplier allocation method lacks flexibility and stability, resulting in unequal resources and differences in service quality. A single qualified supplier bears too many risks, and it is difficult to maximize benefits and serve the new problems introduced by cross-enterprise collaborative distribution. quality.
Adopt the real-time allocation method of industrial chain suppliers based on quality priority. By obtaining the role requirements, maximum capability limitations and scoring information of manufacturers and suppliers, a group role allocation model is built, optimized and calculated, and the priority of excellent suppliers is improved. , reduce the priority of general suppliers, use the qualification matrix adjustment algorithm to optimize the scoring information, and achieve many-to-many real-time assignment of multi-objective optimization.
Improve the flexibility and stability of supplier allocation solutions, solve the problem that the single-target optimization model cannot take into account both profit maximization and service quality, ensure fairness and real-timeness of supplier allocation, and is suitable for part supplier allocation and other Many-to-many assignment scenario.
Smart Images

Figure CN117726140B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of industrial chain supply service technology, and in particular to a real-time allocation method for industrial chain suppliers based on quality priority, a real-time allocation device for industrial chain suppliers based on quality priority, an electronic device and a storage medium. Background Art
[0002] In the industrial chain or logistics chain, the allocation of distribution tasks is a key link. Generally speaking, the parts supplier corresponding to the distribution task is responsible for handling the corresponding logistics tasks. The parts supplier is responsible for coordinating local resources, such as inventory, demand destination and other information, to ensure that the logistics distribution tasks are completed efficiently with the minimum cost.
[0003] However, in reality, in order to maximize profits, manufacturers usually choose to purchase the corresponding parts from suppliers with excellent qualifications, which leads to a shortage of suppliers with excellent qualifications, while other suboptimal suppliers have an oversupply. Similarly, suppliers can also give priority to cooperating with manufacturers with high demand, while other manufacturers need to spend more costs to purchase parts from other suppliers. This can easily lead to problems such as uneven resources.
[0004] In addition, the distribution tasks of the entire industrial chain are completed by a single or a small number of qualified parts suppliers, which lacks flexibility and stability. Once these parts suppliers are unable to provide parts normally due to emergencies such as raw material or machine failure, it will cause the failure of the entire industrial chain.
[0005] At present, cross-enterprise collaborative distribution can assign distribution tasks to partners with similar functions, so as to better adapt to changes in market demand and provide more stable and flexible logistics services. This method can reduce the risks caused by a single or a small number of qualified parts suppliers taking on all distribution tasks in the industrial chain, and is a better method. But correspondingly, cross-enterprise collaborative distribution also introduces new problems. In a group of enterprises, there are differences in the quality of parts produced by different parts suppliers, the delivery distance and the delivery cost, which may lead to differences in the quality of services provided by different enterprises, which is not fair to manufacturers. In other words, the existing supplier allocation plan still lacks flexibility and stability. Summary of the invention
[0006] The present invention provides a real-time allocation method for industrial chain suppliers based on quality priority, a real-time allocation device for industrial chain suppliers based on quality priority, an electronic device and a storage medium, which are used to solve or partially solve the technical problems of lack of flexibility and stability in existing industrial chain supplier allocation methods.
[0007] The present invention provides a real-time allocation method for suppliers in an industrial chain based on quality priority, the method comprising:
[0008] Obtain role requirement information for the producer group, agent maximum capacity limit information for the supplier group, and agent rating information;
[0009] Using the role requirement information, the agent maximum capability limit information, and the agent score information to perform group role allocation calculation to obtain a first real-time allocation result;
[0010] Performing qualification matrix adjustment and optimization on the agent scoring information to obtain optimized agent scoring information;
[0011] Perform group role allocation calculation based on the role requirement information, the agent maximum capability limit information, and the optimized agent score information to obtain a second real-time allocation result;
[0012] The first real-time allocation result and the second real-time allocation result are returned to the display interface for the decision maker to make a decision.
[0013] Optionally, the using the role requirement information, the agent maximum capability limit information and the agent score information to perform group role allocation calculation to obtain a first real-time allocation result includes:
[0014] Constructing a first group role allocation model according to the role requirement information, the agent maximum capability limit information and the agent score information;
[0015] The first group role allocation model is solved to obtain a first real-time allocation result corresponding to the producer group.
[0016] Optionally, constructing a first group role allocation model according to the role requirement information, the agent maximum capability limit information and the agent score information includes:
[0017] Taking cost minimization and benefit maximization as optimization goals, a global optimization objective function is constructed based on the agent scoring information;
[0018] According to the role requirement information and the agent maximum capacity limit information, constructing constraint conditions corresponding to the global optimization objective function, wherein the constraint conditions include supply and demand balance constraints, manufacturer demand satisfaction constraints, and supplier supply capacity constraints;
[0019] A first group role allocation model is constructed based on the global optimization objective function, the supply-demand balance constraint, the manufacturer demand satisfaction constraint, and the supplier supply capacity constraint.
[0020] Optionally, the performing qualification matrix adjustment and optimization on the agent scoring information to obtain optimized agent scoring information includes:
[0021] Taking profit maximization and service quality maximization as optimization goals, a joint optimization objective function is constructed;
[0022] Solving the joint optimization objective function by using the NSGA-II algorithm to obtain a set of parameter solutions;
[0023] The parameter solution set is traversed, and the proxy score information is optimized using a qualification matrix adjustment algorithm to obtain optimized proxy score information.
[0024] Optionally, the parameter solution set includes an excellent qualification benchmark value, a degree of deviation from the excellent qualification value, and an acceptability degree of the qualification value score, and the traversing the parameter solution set and optimizing the proxy score information using a qualification matrix adjustment algorithm to obtain the optimized proxy score information includes:
[0025] Traversing the parameter solution set, and in the traversal process, performing qualification judgment on the proxy score information by using the excellent qualification benchmark value and the degree of deviation from the excellent qualification value, so as to determine the proxy score of each supplier in the supplier group as the first qualification value score or the second qualification value score;
[0026] According to the acceptability of the qualification value score, the priority of the supplier with the first qualification value score is increased, and the priority of the supplier with the second qualification value score is decreased;
[0027] Based on the priority adjustment result, the proxy score information is updated to obtain optimized proxy score information.
[0028] Optionally, performing group role allocation calculation based on the role requirement information, the agent maximum capability limit information, and the optimized agent score information to obtain a second real-time allocation result includes:
[0029] Constructing a second group role allocation model according to the role requirement information, the agent maximum capability limit information and the optimized agent score information;
[0030] The second group role allocation model is solved, and the solution result is filtered by the maximum loss acceptability to obtain a second real-time allocation result corresponding to the producer group.
[0031] Optionally, the method further comprises:
[0032] Establishing a real-time allocation database, and establishing a role requirement table, an agent maximum capacity limit table, an agent scoring table, and a supplier allocation table in the real-time allocation database;
[0033] Wherein, the role requirement table is used to collect and store the role requirement information of the producer group in real time;
[0034] The agent maximum capacity limit table is used to collect and store agent maximum capacity limit information of the supplier group in real time;
[0035] The agent scoring table is used to collect and store agent scoring information of the supplier group in real time;
[0036] The supplier allocation table is used to store the real-time allocation results of suppliers corresponding to the manufacturer group.
[0037] The present invention also provides a real-time allocation device for suppliers in the industrial chain based on quality priority, comprising:
[0038] The module for obtaining information to be processed is used to obtain the role requirement information of the manufacturer group, the agent maximum capacity limit information of the supplier group, and the agent rating information;
[0039] A first real-time allocation result calculation module, configured to use the role requirement information, the agent maximum capability limit information, and the agent score information to perform group role allocation calculation to obtain a first real-time allocation result;
[0040] A qualification matrix adjustment and optimization module is used to adjust and optimize the qualification matrix of the agent scoring information to obtain optimized agent scoring information;
[0041] A second real-time allocation result calculation module is used to perform group role allocation calculation based on the role requirement information, the agent maximum capability limit information and the optimized agent score information to obtain a second real-time allocation result;
[0042] The real-time allocation result returning module is used to return the first real-time allocation result and the second real-time allocation result to the display interface for decision makers to make decisions.
[0043] The present invention also provides an electronic device, the device comprising a processor and a memory:
[0044] The memory is used to store program code and transmit the program code to the processor;
[0045] The processor is used to execute the real-time allocation method of industrial chain suppliers based on quality priority as described in any of the above items according to the instructions in the program code.
[0046] The present invention also provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the real-time allocation method of industrial chain suppliers based on quality priority as described in any of the above items.
[0047] It can be seen from the above technical solutions that the present invention has the following advantages:
[0048] A real-time allocation method for parts suppliers in an industrial chain based on quality priority is provided. First, the role requirement information of the manufacturer group, the agent maximum capacity limit information and the agent score information of the supplier group are obtained; then the role requirement information, the agent maximum capacity limit information and the agent score information are used to calculate the group role allocation to obtain the first real-time allocation result; the qualification matrix is adjusted and optimized for the agent score information to obtain the optimized agent score information; then the group role allocation calculation is performed based on the role requirement information, the agent maximum capacity limit information and the optimized agent score information to obtain the second real-time allocation result. Therefore, in view of the cross-enterprise collaborative distribution problem existing in the industrial chain, the real-time allocation process of parts suppliers for group role allocation is extended based on the qualification matrix adjustment and optimization, so as to realize the multi-objective optimized many-to-many real-time assignment while solving the problem that only some excellent parts suppliers supply parts due to algorithm defects. Finally, the first real-time allocation result and the second real-time allocation result are returned to the display interface for decision makers to make decisions based on actual application scenarios or actual needs, so as to ensure the stability of the obtained allocation scheme while greatly improving the flexibility of scheme selection. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0050] Figure 1 A flowchart of the steps of a real-time allocation method for suppliers in the industrial chain based on quality priority;
[0051] Figure 2 It is a schematic diagram of the overall process of a real-time allocation method of suppliers in the industrial chain based on quality priority;
[0052] Figure 3 A schematic diagram of a real-time allocation scheme obtained through group role allocation calculation when no qualification matrix adjustment is adopted;
[0053] Figure 4 A schematic diagram of a real-time allocation scheme obtained by calculating group role allocation under qualification matrix adjustment;
[0054] Figure 5 The present invention is a structural block diagram of a real-time allocation device for suppliers in an industrial chain based on quality priority. DETAILED DESCRIPTION
[0055] Embodiments of the present invention provide a real-time allocation method for industrial chain suppliers based on quality priority, a real-time allocation device for industrial chain suppliers based on quality priority, an electronic device and a storage medium, which are used to solve or partially solve the technical problems of lack of flexibility and stability in existing industrial chain supplier allocation methods.
[0056] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below 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.
[0057] In order to facilitate better understanding by those skilled in the art, the relevant technical features or technical means involved in the technical solution of the present invention are first explained and illustrated.
[0058] GMRA (Group Multirole Assignment): also known as group role assignment, is a group multi-task assignment solution. By building a group role assignment model (i.e., GMRA model) based on the relevant information of manufacturers and suppliers and solving it, the corresponding assignment results can be obtained.
[0059] Pulp: A linear programming calculation package in Python language, used for establishing and calculating linear programming models.
[0060] NSGA-II (Non-Dominated Sorting Genetic Algorithm-II): An algorithm for parameter search. Its main core is: for optimization problem solving, an elite strategy is used to prevent the loss of excellent individuals, and a set of parameter solutions is obtained by mixing all individuals of the parent and child generations and performing non-dominated sorting.
[0061] QMA (Qualification Matrix Adjustments): An optimization solution algorithm proposed in the embodiment of the present invention. Its calculation principle is as follows: first, the optimization goal is defined, that is, the maximization of the benefits and service quality of the distribution task is taken as the optimization goal; secondly, the defined optimization goal is solved by the NSGA-II algorithm to obtain a set of parameter solutions (the corresponding optimization results can be obtained through this set of parameter solutions); finally, by setting the maximum acceptable loss level, the target allocation plan is obtained, and the target allocation plan is stored and returned for decision makers to make decisions.
[0062] As an example, cross-enterprise collaborative distribution can assign distribution tasks to partners with similar functions, so as to better adapt to changes in market demand and provide more stable and flexible logistics services. This method can reduce the risks caused by a single or a small number of well-qualified parts suppliers taking on all distribution tasks in the industrial chain, and is a better method. But correspondingly, cross-enterprise collaborative distribution also introduces new problems. In a group of enterprises, there are differences in the quality of parts produced by different parts suppliers, the delivery distance and delivery cost, which may lead to differences in the quality of services provided by different enterprises, which is not fair to manufacturers. In other words, the existing supplier allocation scheme still lacks flexibility and stability.
[0063] In modern logistics management, while giving priority to maximizing the benefits of distribution tasks, the industrial and public management sectors are paying more and more attention to other factors, which often conflict with the maximization of benefits. Therefore, for the industrial chain, it is of great significance to balance the relationship between maximizing the benefits of distribution tasks and service quality. However, the current mainstream many-to-many assignment scheme is usually a single-objective optimization model, which cannot maximize the benefits of distribution tasks while taking into account service quality. This model has a low evaluation frequency, cannot reflect the latest changes of suppliers, and has poor real-time performance. In addition, the method based on supplier evaluation technology is subjective and the results are poorly operational. If it is not designed properly, it is very likely to lose potential suppliers.
[0064] Therefore, one of the core invention points of the embodiment of the present invention is: taking into account the relationship between service quality and profit maximization in the distribution task, a real-time allocation method for parts suppliers in the industrial chain based on quality priority is proposed. In view of the cross-enterprise collaborative distribution problem existing in the industrial chain, the real-time allocation process of parts suppliers of the GMRA model is extended by the QMA algorithm. By increasing the priority of excellent parts suppliers and reducing the priority of parts suppliers with general qualifications, a stable, efficient and flexible parts supplier allocation scheme is provided. While realizing multi-objective optimization of many-to-many real-time assignment, the problem that only some excellent parts suppliers supply parts due to algorithm defects is solved. After obtaining the supplier allocation scheme by calculation, the decision maker can select the parts supplier allocation scheme that meets the actual scenario according to actual needs, thereby solving the problem that the single-objective optimization model cannot take into account the relationship between maximizing benefits and service quality in the cross-enterprise collaborative distribution problem, as well as other problems caused by differences in the service quality provided by enterprises. And the technical solution provided by the embodiment of the present invention is suitable for but not limited to the parts supplier allocation scheme, and can also be used in other many-to-many assignment application scenarios, with certain extensibility.
[0065] Reference Figure 1 , shows a flowchart of a method for real-time allocation of suppliers in an industrial chain based on quality priority provided by an embodiment of the present invention, which may specifically include the following steps:
[0066] Step 101, obtaining role requirement information of the manufacturer group, agent maximum capacity limit information of the supplier group and agent rating information;
[0067] Before actual data processing, a real-time allocation database for storing data can be established, and four tables can be established in the real-time allocation database, namely, the role requirement table, the agent maximum capacity limit table, the agent scoring table and the supplier allocation table. The information of these four tables can be associated through task numbers.
[0068] The role requirement table is used to collect and store the role requirement information of the manufacturer group in real time, specifically storing the number of parts required by each manufacturer in the manufacturer group.
[0069] The agent maximum capacity limit table is used to collect and store the agent maximum capacity limit information of the supplier group in real time, specifically storing the maximum number of parts that each supplier in the supplier group can provide.
[0070] The agent scoring table is used to collect and store the agent scoring information of the supplier group in real time, specifically to store the comprehensive score value of each supplier in the supplier group, which can be obtained by comprehensively calculating factors such as average delivery distance and delivery cost.
[0071] The supplier allocation table is used to store the real-time allocation results of suppliers corresponding to the manufacturer group. Specifically, after the supplier allocation results of the real-time allocation task are obtained through relevant allocation calculations, the allocation results are stored.
[0072] In actual applications, the relevant information required in the role requirement table, agent maximum capacity limit table and agent scoring table (such as the role requirement information of the manufacturer group, the agent maximum capacity limit information of the supplier group and the agent scoring information) can be obtained in real time. Then the decision maker can send the corresponding data to the background through the user display interface. The background can collect the relevant data and store it in the corresponding form of the real-time allocation database. If the input data is empty, it means that the data has not been successfully obtained. At this time, the above data acquisition process needs to be repeated until the data is correctly transmitted.
[0073] When executing the supplier allocation task, the corresponding data can be obtained by reading the role requirement table, the agent maximum capacity limit table and the agent score table, and the corresponding data can be stored in the form of an array.
[0074] Among them, the role requirement table is an n-dimensional array, which can be represented by L, and the index in the array represents the corresponding j-th producer.
[0075] The agent maximum capacity limit table is an m-dimensional array, which can be represented by L a Indicates that the index in the array represents the corresponding i-th supplier.
[0076] The proxy score table is an m*n array, which can be represented by Q, indicating the proxy score value of the parts provided by the i-th supplier to the j-th manufacturer.
[0077] Step 102, using the role requirement information, the agent maximum capability limit information and the agent score information to perform group role allocation calculation to obtain a first real-time allocation result;
[0078] Then, based on the acquired data, a group role allocation model (i.e., GMRA model) can be constructed and solved to perform real-time allocation calculation processing, calculate the allocation plan with maximum benefit, and obtain the overall performance score and variance value under the allocation plan for calculation in subsequent steps.
[0079] Specifically, the process of using the role requirement information, the agent maximum capability limit information and the agent score information to perform group role allocation calculation and obtain the first real-time allocation result can be achieved by executing the following steps S01 to S02:
[0080] Step S01: constructing a first group role allocation model according to role requirement information, agent maximum capability limit information and agent score information;
[0081] Furthermore, this step can be implemented by executing the following sub-steps S011 to S013:
[0082] Step S011: Taking cost minimization and benefit maximization as optimization goals, a global optimization objective function is constructed based on agent rating information;
[0083] In the embodiment of the present invention, the benefits are maximized by minimizing the costs. Therefore, the smaller the proxy score is, the better, and accordingly, the smaller the overall performance score is, the better.
[0084] Taking cost minimization and benefit maximization as optimization goals, the established global optimization objective function is as follows:
[0085]
[0086] in, represents cost minimization, Q[i, j] represents the agent score value; T[i, j] represents the allocation plan to be solved. After the allocation plan is calculated, the data of T[i, j] can be stored in the supplier allocation table.
[0087] Step S012: constructing constraint conditions corresponding to the global optimization objective function according to the role requirement information and the agent maximum capacity limit information, wherein the constraint conditions include supply and demand balance constraint, manufacturer demand satisfaction constraint and supplier supply capacity constraint;
[0088] Specifically, the supply and demand balance constraint expression is as follows:
[0089]
[0090] The supply-demand balance constraint states that each supplier provides less than or equal to the number of parts required by each producer.
[0091] The producer's demand satisfies the constraint expression as follows:
[0092]
[0093] The producer demand satisfaction constraint means that during the allocation process, each producer can obtain the target number of parts.
[0094] The supplier supply capacity constraint expression is as follows:
[0095]
[0096] The supplier supply capacity constraint means that during the allocation process, each supplier cannot provide a number of parts that exceeds its own capacity.
[0097] Step S013: constructing a first group role allocation model based on the global optimization objective function, the supply-demand balance constraint, the manufacturer's demand satisfaction constraint, and the supplier's supply capacity constraint.
[0098] Taking the global optimization objective function as the objective function to be solved, and the supply-demand balance constraint, the manufacturer's demand satisfaction constraint and the supplier's supply capacity constraint as the constraint conditions, the corresponding first group role allocation model is constructed.
[0099] Step S02: Solve the first group role allocation model to obtain a first real-time allocation result corresponding to the manufacturer group.
[0100] In practical applications, steps S01 to S02 are modeled and calculated using the Pulp package in Python, and the first real-time allocation result corresponding to the manufacturer group can be obtained, and the overall performance score and variance value corresponding to the first real-time allocation result can also be obtained. Then the corresponding real-time allocation result can be stored in the supplier allocation table.
[0101] Step 103, performing qualification matrix adjustment and optimization on the agent scoring information to obtain optimized agent scoring information;
[0102] One of the improvements of the embodiment of the present invention is that the real-time allocation process of parts suppliers of the group role allocation model is expanded by adopting a qualification matrix adjustment algorithm (ie, QMA algorithm).
[0103] Specifically, in this step, the process of adjusting and optimizing the qualification matrix of the agent scoring information to obtain the optimized agent scoring information can be implemented by executing the following steps S11 to S13:
[0104] Step S11: Taking profit maximization and service quality maximization as optimization goals, constructing a joint optimization objective function;
[0105] First, it is necessary to define the optimization goal. In the embodiment of the present invention, the main definition is to maximize the benefits of the delivery task and maximize the service quality of the delivery task. The specific definition is as follows:
[0106]
[0107]
[0108] in, and They correspond to the overall benefit of the distribution task and the overall service quality. The comprehensive understanding of the overall service quality is to maximize the service quality under the condition of minimizing the cost.
[0109] and The calculation expressions are as follows:
[0110]
[0111]
[0112] in, represents the variance value obtained by solving the first group role allocation model in the above step; Indicates the variance value to be solved, which can be calculated through subsequent steps; represents the overall performance score obtained by solving the first group role allocation model in the previous step; Represents the overall performance score of the problem to be solved, which can be calculated in subsequent steps.
[0113] Step S12: Solve the joint optimization objective function by using the NSGA-II algorithm to obtain a set of parameter solutions;
[0114] By solving the two optimization objectives in step S12 through the NSGA-II algorithm, a set of parameter solutions for the eligibility matrix adjustment algorithm can be obtained.
[0115] Step S13: traverse the parameter solution set, and use the qualification matrix adjustment algorithm to optimize the agent scoring information to obtain the optimized agent scoring information.
[0116] Then, the parameter solution set can be traversed, and the data in the agent score table (i.e., the agent score value of each supplier) can be recalculated using the qualification matrix adjustment algorithm. The qualification matrix adjustment algorithm contains three parameters in total, and accordingly, the parameter solution set can also include three parameter values, namely, the excellent qualification benchmark value , the degree of deviation of excellent qualification value and the acceptability of the qualification score .
[0117] Specifically, the process of traversing the parameter solution set, optimizing the proxy scoring information using the qualification matrix adjustment algorithm, and obtaining the optimized proxy scoring information can be achieved by executing the following sub-steps S131 to S133:
[0118] Step S131: traverse the parameter solution set, and in the traversal process, make a qualification judgment on the proxy score information by using the excellent qualification benchmark value and the degree of deviation from the excellent qualification value, so as to determine the proxy score of each supplier in the supplier group as the first qualification value score or the second qualification value score;
[0119] First, through the excellent qualification benchmark value And the degree of deviation of excellent qualification value The agent scoring information is qualified to classify the qualification value scores into excellent (first qualification value score) and general (second qualification value score).
[0120] This step can be simply understood as: if the target value (i.e. the agent score value of the supplier to be judged as excellent or general) is greater than the excellent qualification benchmark value , and within the range of deviation from the excellent qualification value (i.e. the degree of deviation from the excellent qualification value ), the corresponding agent rating value of the supplier is determined as average. If the target value is less than the excellent qualification benchmark value , and in the range of deviation from the excellent qualification value If the supplier's corresponding agent rating value is within , the supplier's corresponding agent rating value is determined to be excellent.
[0121] Step S132: according to the acceptability of the qualification value score, the priority of the supplier with the first qualification value score is increased, and the priority of the supplier with the second qualification value score is decreased;
[0122] Secondly, for suppliers with excellent qualification scores, the acceptability of the qualification scores is Increase the corresponding priority value, that is, divide the current agent score of the supplier by the acceptability of the qualification score. , obtain a new proxy score value to increase its priority.
[0123] For suppliers with average qualification scores, the acceptability of the qualification scores is determined by the qualification score. Reduce the corresponding priority value, that is, multiply the current agent score of the supplier by the acceptability of the qualification score. , obtain a new proxy score value to lower its priority.
[0124] In this step, the purpose of adjusting the priority is to ensure the stable supply of excellent suppliers, restrict suppliers with average performance, and allow suppliers in between to have room to perform. Therefore, for suppliers with excellent qualification values, their priority is increased; for suppliers with average qualification values, their priority is restricted; for suppliers between excellent and average, the best ones are selected.
[0125] Step S133: Based on the priority adjustment result, the proxy score information is updated to obtain optimized proxy score information.
[0126] Then, based on the priority adjustment result of step S132, the proxy score information may be updated to obtain optimized proxy score information.
[0127] Step 104, performing group role allocation calculation based on the role requirement information, the agent maximum capability limit information and the optimized agent score information to obtain a second real-time allocation result;
[0128] Specifically, step 104 can be implemented by executing the following sub-steps S21 to S22:
[0129] Step S21: constructing a second group role allocation model according to role requirement information, agent maximum capability limit information and optimized agent score information;
[0130] First, according to the role requirement information, the agent maximum capability limit information and the optimized agent score information, the relevant process in the aforementioned step S01 is called to build a second group role allocation model. Since this step can be implemented with reference to the relevant process in the aforementioned embodiment, it will not be repeated here.
[0131] Step S22: Solve the second group role allocation model, and filter the solution results by the maximum loss acceptance level to obtain the second real-time allocation result corresponding to the manufacturer group.
[0132] Then solve the second group role allocation model, so as to solve the overall performance score corresponding to each set of parameters in the parameter solution set. and variance .
[0133] Then, the overall performance score and variance value corresponding to each group of parameters can be saved as a solution set, and filtered by the maximum acceptable loss level to obtain the corresponding target allocation plan, that is, the second real-time allocation result corresponding to the manufacturer group. Among them, the maximum acceptable loss level represents the maximum acceptable loss level during the allocation process, which can be set in the form of a small interval or as a threshold. Those skilled in the art can pre-set it according to the actual situation. Similarly, after the second real-time allocation result is obtained, it can be stored in the supplier allocation table.
[0134] Step 105: Return the first real-time allocation result and the second real-time allocation result to a display interface for decision makers to make decisions.
[0135] Finally, the first real-time allocation result (i.e., the allocation result obtained by calculating only through the group role allocation model) and the second real-time allocation result (i.e., the optimized allocation result obtained after expanding the real-time allocation process of the parts supplier of the group role allocation model by adopting the qualification matrix adjustment algorithm) can be returned to the display interface for decision makers to make decisions.
[0136] The purpose of performing this step is that in actual applications, decision makers often need to consider multiple factors, such as whether to sacrifice a small part of the interests to achieve optimal balance according to the actual situation. If the decision maker believes that the desired effect can be achieved without sacrificing a small part of the interests, the first real-time allocation result can be selected as the final selection scheme. If the decision maker wants to maximize the benefits and service quality, the second real-time allocation result can be selected as the final selection scheme. Or it can be pre-set that after calculating the two real-time allocation results, if the data comparison is not much different, the first scheme is automatically selected to ensure that there is no need to sacrifice a small part of the interests as much as possible. If the data comparison is greatly different, the second scheme is automatically selected to maximize the balance between interests and service quality. Therefore, through the relevant settings of the scheme selection, it can be determined according to the actual situation which allocation scheme is finally selected, which can improve the flexibility of scheme selection.
[0137] It should be noted that, in order to enable those skilled in the art to better distinguish data of the same type but with different actual meanings, some technical features are distinguished by the first and the second in the embodiments of the present invention. The first and the second are only used for data distinction and have no other special meanings. It can be understood that the present invention is not limited to this.
[0138] In addition, with the rapid development of artificial intelligence technology, more and more researchers choose to use deep learning algorithms for data processing. In addition to using the methods in the aforementioned related steps to implement the real-time supplier allocation scheme proposed by the present invention, the evaluation and selection of parts suppliers can also be achieved through appropriate deep learning algorithms. For example, neural networks such as LSTM (Long Short-Term Memory), SRNN (Sliced Recurrent Neural Network), and GRU (Gate Recurrent Unit) can be used. Because the construction and training of neural networks are not the focus of the technical solution of the present invention, they will not be elaborated here. It can be understood that the present invention is not limited to this.
[0139] In the embodiment of the present invention, the relationship between service quality and profit maximization in the distribution task is taken into account, and a real-time allocation method for parts suppliers in the industrial chain based on quality priority is proposed. In view of the cross-enterprise collaborative distribution problem existing in the industrial chain, the real-time allocation process of parts suppliers of the GMRA model is extended by the QMA algorithm. By increasing the priority of excellent parts suppliers and reducing the priority of parts suppliers with general qualifications, a stable, efficient and flexible parts supplier allocation scheme is provided. While realizing multi-objective optimization of many-to-many real-time assignment, the problem that only some excellent parts suppliers supply parts due to algorithm defects is solved. After obtaining the supplier allocation scheme by calculation, the decision maker can select the parts supplier allocation scheme that meets the actual scenario according to actual needs, thereby solving the problem that the single-objective optimization model cannot take into account the relationship between maximizing benefits and service quality in the cross-enterprise collaborative distribution problem, as well as other problems caused by differences in the service quality provided by enterprises. In addition, the technical solution provided by the embodiment of the present invention is suitable for but not limited to the parts supplier allocation scheme, and can also be used in other many-to-many assignment application scenarios, with certain extensibility.
[0140] For better explanation, refer to Figure 2 , showing an overall flow diagram of a method for real-time allocation of suppliers in an industrial chain based on quality priority provided by an embodiment of the present invention. It should be pointed out that this embodiment only briefly describes the general process of real-time allocation of suppliers in the industrial chain, and the specific implementation process of each step can be understood by referring to the relevant content in the aforementioned embodiment, which will not be described here. It can be understood that the present invention does not limit this.
[0141] 1. First, establish a real-time allocation database and create four tables in the database, namely the role requirement table, the agent maximum capacity limit table, the agent scoring table and the supplier allocation table. These four tables are associated with information through task numbers (not drawn to simplify the process).
[0142] 2. Collect data in real time and determine whether the data is collected successfully. If not, jump to step 3; if yes, jump to step 4.
[0143] 3. Determine whether to end the allocation. If not, jump to step 2; if yes, end the process;
[0144] 4. Store the collected data in the corresponding table in the database, and perform GMRA model calculation to obtain the first real-time allocation result;
[0145] 5. Use the QMA algorithm to expand the real-time allocation process of the GMRA model, adjust and optimize the qualification matrix based on the QMA algorithm, calculate the allocation results, and obtain the second real-time allocation result;
[0146] 6. Return the first real-time allocation result and the second real-time allocation result to the user display interface for the decision maker to select, and end the process.
[0147] For example, the two schemes calculated in step 4 and step 5 in this example are returned to the decision maker for decision making. The schematic diagrams of the allocation schemes are as follows: Figure 3 and Figure 4 shown.
[0148] Among them, step 4 corresponds to Figure 3 The real-time allocation scheme obtained by group role allocation (GMRA) calculation without using qualification matrix adjustment (QMA) is shown. It represents the supplier allocation scheme obtained under the premise of maximizing distribution benefits (the allocation results are different in different scenarios).
[0149] Step 5 corresponds to Figure 4 The real-time allocation scheme shown in the figure is obtained by calculating the group role allocation (GMRA) under the qualification matrix adjustment (QMA). It represents a comprehensive scheme that balances the relationship between the maximization of benefits of the delivery task and the quality of service, sacrificing a small part of the benefits (the allocation results are different in different scenarios).
[0150] Reference Figure 5 , shows a structural block diagram of a real-time allocation device for suppliers in an industrial chain based on quality priority provided by an embodiment of the present invention, which may specifically include:
[0151] The to-be-processed information acquisition module 501 is used to acquire the role requirement information of the manufacturer group, the agent maximum capacity limit information of the supplier group, and the agent rating information;
[0152] A first real-time allocation result calculation module 502 is used to perform group role allocation calculation using the role requirement information, the agent maximum capability limit information and the agent score information to obtain a first real-time allocation result;
[0153] The qualification matrix adjustment and optimization module 503 is used to adjust and optimize the qualification matrix of the agent scoring information to obtain optimized agent scoring information;
[0154] A second real-time allocation result calculation module 504 is used to perform group role allocation calculation based on the role requirement information, the agent maximum capability limit information and the optimized agent score information to obtain a second real-time allocation result;
[0155] The real-time allocation result returning module 505 is used to return the first real-time allocation result and the second real-time allocation result to the display interface for the decision maker to make a decision.
[0156] In an optional embodiment, the first real-time allocation result calculation module 502 includes:
[0157] A first group role allocation model construction module, configured to construct a first group role allocation model according to the role requirement information, the agent maximum capability limit information and the agent score information;
[0158] The first group role allocation model solving module is used to solve the first group role allocation model to obtain a first real-time allocation result corresponding to the manufacturer group.
[0159] In an optional embodiment, the first group role allocation model construction module includes:
[0160] A global optimization objective function construction module is used to construct a global optimization objective function based on the agent scoring information, taking cost minimization and benefit maximization as optimization objectives;
[0161] A constraint condition construction module, used to construct the constraint conditions corresponding to the global optimization objective function according to the role requirement information and the agent maximum capacity limit information, wherein the constraint conditions include supply and demand balance constraints, manufacturer demand satisfaction constraints and supplier supply capacity constraints;
[0162] The first group role allocation model construction submodule is used to construct a first group role allocation model based on the global optimization objective function, the supply and demand balance constraint, the manufacturer demand satisfaction constraint and the supplier supply capacity constraint.
[0163] In an optional embodiment, the qualification matrix adjustment and optimization module 503 includes:
[0164] A joint optimization objective function construction module is used to construct a joint optimization objective function with profit maximization and service quality maximization as optimization objectives;
[0165] A joint optimization objective function solving module is used to solve the joint optimization objective function by using the NSGA-II algorithm to obtain a set of parameter solutions;
[0166] The qualification matrix adjustment module is used to traverse the parameter solution set, optimize the agent scoring information using a qualification matrix adjustment algorithm, and obtain optimized agent scoring information.
[0167] In an optional embodiment, the parameter solution set includes an excellent qualification benchmark value, a deviation from the excellent qualification value, and an acceptability of the qualification value score, and the qualification matrix adjustment module includes:
[0168] a qualification judgment module, configured to traverse the parameter solution set, and in the traversal process, perform qualification judgment on the proxy score information by using the excellent qualification benchmark value and the deviation excellent qualification value degree, so as to determine the proxy score of each supplier in the supplier group as the first qualification value score or the second qualification value score;
[0169] A priority adjustment module, used to increase the priority of suppliers with a first qualification value score and reduce the priority of suppliers with a second qualification value score according to the acceptability of the qualification value score;
[0170] The proxy score information updating module is used to update the proxy score information based on the priority adjustment result to obtain optimized proxy score information.
[0171] In an optional embodiment, the second real-time allocation result calculation module 504 includes:
[0172] A second group role allocation model construction module, used to construct a second group role allocation model according to the role requirement information, the agent maximum capability limit information and the optimized agent score information;
[0173] The second group role allocation model solving module is used to solve the second group role allocation model and filter the solution result by the maximum loss acceptance degree to obtain the second real-time allocation result corresponding to the manufacturer group.
[0174] In an optional embodiment, the device further includes:
[0175] A database construction module is used to establish a real-time allocation database, and to establish a role requirement table, an agent maximum capacity limit table, an agent scoring table, and a supplier allocation table in the real-time allocation database;
[0176] Wherein, the role requirement table is used to collect and store the role requirement information of the producer group in real time;
[0177] The agent maximum capacity limit table is used to collect and store agent maximum capacity limit information of the supplier group in real time;
[0178] The agent scoring table is used to collect and store agent scoring information of the supplier group in real time;
[0179] The supplier allocation table is used to store the real-time allocation results of suppliers corresponding to the manufacturer group.
[0180] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the aforementioned method embodiment.
[0181] An embodiment of the present invention further provides an electronic device, the device comprising a processor and a memory:
[0182] The memory is used to store the program code and transmit the program code to the processor;
[0183] The processor is used to execute the real-time allocation method of the industrial chain supplier based on quality priority according to the instructions in the program code of any embodiment of the present invention.
[0184] An embodiment of the present invention further provides a computer-readable storage medium, which is used to store program code, and the program code is used to execute the real-time allocation method of industrial chain suppliers based on quality priority according to any embodiment of the present invention.
[0185] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0186] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.
[0187] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0188] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0189] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.
[0190] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A real-time allocation method for suppliers in an industrial chain based on quality priority, characterized in that: include: Obtain role requirement information for the producer group, agent maximum capacity limit information for the supplier group, and agent rating information; Using the role requirement information, the agent maximum capability limit information, and the agent score information to perform group role allocation calculation to obtain a first real-time allocation result; Performing qualification matrix adjustment and optimization on the agent scoring information to obtain optimized agent scoring information; Perform group role allocation calculation based on the role requirement information, the agent maximum capability limit information, and the optimized agent score information to obtain a second real-time allocation result; Returning the first real-time allocation result and the second real-time allocation result to a display interface for decision makers to make decisions; The step of using the role requirement information, the agent maximum capability limit information, and the agent score information to perform group role allocation calculation to obtain a first real-time allocation result includes: Constructing a first group role allocation model according to the role requirement information, the agent maximum capability limit information and the agent score information; Solving the first group role allocation model to obtain a first real-time allocation result corresponding to the producer group; The step of constructing a first group role allocation model according to the role requirement information, the agent maximum capability limit information and the agent score information includes: Taking cost minimization and benefit maximization as optimization goals, based on the agent rating information, the global optimization objective function is constructed as follows: ; in, represents cost minimization; Q represents the proxy score table, which is an m*n array, Q[i, j] represents the proxy score value of the parts provided by the i-th supplier to the j-th manufacturer; T[i, j] represents the allocation solution to be solved; According to the role requirement information and the agent maximum capacity limit information, constructing constraint conditions corresponding to the global optimization objective function, wherein the constraint conditions include supply and demand balance constraints, manufacturer demand satisfaction constraints, and supplier supply capacity constraints; The supply and demand balance constraint expression is as follows: ; The producer's demand satisfies the constraint expression as follows: ; The supplier supply capacity constraint expression is as follows: ; Among them, L represents the role demand table, which is an n-dimensional array, and the index j in the array represents the corresponding j-th producer; It represents the agent maximum capacity limit table, which is an m-dimensional array. The index i in the array represents the corresponding i-th supplier; Constructing a first group role allocation model based on the global optimization objective function, the supply-demand balance constraint, the manufacturer demand satisfaction constraint, and the supplier supply capacity constraint; The step of adjusting and optimizing the qualification matrix of the agent scoring information to obtain optimized agent scoring information includes: Taking profit maximization and service quality maximization as optimization goals, the joint optimization objective function is constructed as follows: ; ; in, and They correspond to the overall benefit of the distribution task and the overall service quality. The overall service quality refers to maximizing service quality under the condition of minimizing cost. and The calculation expressions are as follows: ; ; in, represents the variance value obtained by solving the first group role allocation model; represents the variance value to be solved; represents the overall performance score obtained by solving the first group role allocation model; represents the overall performance score of the problem to be solved; Solving the joint optimization objective function by using the NSGA-II algorithm to obtain a set of parameter solutions; Traversing the parameter solution set, optimizing the agent scoring information using a qualification matrix adjustment algorithm, and obtaining optimized agent scoring information; The parameter solution set includes an excellent qualification benchmark value, a deviation degree from the excellent qualification value, and an acceptable degree of qualification value scoring. The traversing of the parameter solution set and optimizing the proxy scoring information using a qualification matrix adjustment algorithm to obtain the optimized proxy scoring information include: Traversing the parameter solution set, and in the traversal process, performing qualification judgment on the proxy score information by using the excellent qualification benchmark value and the degree of deviation from the excellent qualification value, so as to determine the proxy score of each supplier in the supplier group as the first qualification value score or the second qualification value score; According to the acceptability of the qualification value scores, the priority of suppliers with the first qualification value scores is increased, and the priority of suppliers with the second qualification value scores is decreased; Based on the priority adjustment result, the proxy score information is updated to obtain optimized proxy score information.
2. The real-time allocation method for suppliers in the industrial chain based on quality priority according to claim 1 is characterized in that: The performing group role allocation calculation based on the role requirement information, the agent maximum capability limit information and the optimized agent score information to obtain a second real-time allocation result includes: Constructing a second group role allocation model according to the role requirement information, the agent maximum capability limit information and the optimized agent score information; The second group role allocation model is solved, and the solution result is filtered by the maximum loss acceptability to obtain a second real-time allocation result corresponding to the producer group.
3. The real-time allocation method for suppliers in the industrial chain based on quality priority according to claim 1 or 2 is characterized in that: Also includes: Establishing a real-time allocation database, and establishing a role requirement table, an agent maximum capacity limit table, an agent scoring table, and a supplier allocation table in the real-time allocation database; Wherein, the role requirement table is used to collect and store the role requirement information of the producer group in real time; The agent maximum capacity limit table is used to collect and store agent maximum capacity limit information of the supplier group in real time; The agent scoring table is used to collect and store agent scoring information of the supplier group in real time; The supplier allocation table is used to store the real-time allocation results of suppliers corresponding to the manufacturer group.
4. A real-time allocation device for suppliers in the industrial chain based on quality priority, characterized in that: include: The module for obtaining information to be processed is used to obtain the role requirement information of the manufacturer group, the agent maximum capacity limit information of the supplier group, and the agent rating information; A first real-time allocation result calculation module, configured to use the role requirement information, the agent maximum capability limit information, and the agent score information to perform group role allocation calculation to obtain a first real-time allocation result; A qualification matrix adjustment and optimization module is used to adjust and optimize the qualification matrix of the agent scoring information to obtain optimized agent scoring information; A second real-time allocation result calculation module is used to perform group role allocation calculation based on the role requirement information, the agent maximum capability limit information and the optimized agent score information to obtain a second real-time allocation result; A real-time allocation result returning module, used for returning the first real-time allocation result and the second real-time allocation result to a display interface for decision makers to make decisions; The first real-time allocation result calculation module includes: A first group role allocation model construction module, configured to construct a first group role allocation model according to the role requirement information, the agent maximum capability limit information and the agent score information; A first group role allocation model solving module, used to solve the first group role allocation model to obtain a first real-time allocation result corresponding to the producer group; Wherein, the first group role allocation model construction module includes: The global optimization objective function construction module is used to take cost minimization and benefit maximization as optimization objectives. Based on the agent scoring information, the global optimization objective function is constructed as follows: ; in, represents cost minimization; Q represents the proxy score table, which is an m*n array, Q[i, j] represents the proxy score value of the parts provided by the i-th supplier to the j-th manufacturer; T[i, j] represents the allocation solution to be solved; A constraint condition construction module, used to construct the constraint conditions corresponding to the global optimization objective function according to the role requirement information and the agent maximum capacity limit information, wherein the constraint conditions include supply and demand balance constraints, manufacturer demand satisfaction constraints and supplier supply capacity constraints; The supply and demand balance constraint expression is as follows: ; The producer's demand satisfies the constraint expression as follows: ; The supplier supply capacity constraint expression is as follows: ; Among them, L represents the role demand table, which is an n-dimensional array, and the index j in the array represents the corresponding j-th producer; It represents the agent maximum capacity limit table, which is an m-dimensional array. The index i in the array represents the corresponding i-th supplier; A first group role allocation model construction submodule, configured to construct a first group role allocation model based on the global optimization objective function, the supply-demand balance constraint, the manufacturer demand satisfaction constraint, and the supplier supply capacity constraint; The qualification matrix adjustment and optimization module includes: The joint optimization objective function construction module is used to maximize benefits and service quality as optimization goals. The joint optimization objective function is constructed as follows: ; ; in, and They correspond to the overall benefit of the distribution task and the overall service quality. The overall service quality refers to maximizing service quality under the condition of minimizing cost. and The calculation expressions are as follows: ; ; in, represents the variance value obtained by solving the first group role allocation model; represents the variance value to be solved; represents the overall performance score obtained by solving the first group role allocation model; represents the overall performance score of the problem to be solved; A joint optimization objective function solving module is used to solve the joint optimization objective function by using the NSGA-II algorithm to obtain a set of parameter solutions; A qualification matrix adjustment module, used for traversing the parameter solution set, optimizing the agent scoring information by using a qualification matrix adjustment algorithm, and obtaining optimized agent scoring information; The parameter solution set includes an excellent qualification benchmark value, a deviation from the excellent qualification value, and an acceptable degree of qualification value scoring, and the qualification matrix adjustment module includes: a qualification judgment module, configured to traverse the parameter solution set, and in the traversal process, perform qualification judgment on the proxy score information by using the excellent qualification benchmark value and the deviation excellent qualification value degree, so as to determine the proxy score of each supplier in the supplier group as the first qualification value score or the second qualification value score; A priority adjustment module, used to increase the priority of suppliers with a first qualification value score and reduce the priority of suppliers with a second qualification value score according to the acceptability of the qualification value score; The proxy score information updating module is used to update the proxy score information based on the priority adjustment result to obtain optimized proxy score information.
5. An electronic device, characterized in that: The device comprises a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the real-time allocation method for industrial chain suppliers based on quality priority as described in any one of claims 1-3 according to the instructions in the program code.
6. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program code, and the program code is used to execute the real-time allocation method for industrial chain suppliers based on quality priority as described in any one of claims 1-3.
