Methods, apparatus, equipment, storage media, and products for generating financing schemes

Abstract

This application discloses a method, apparatus, equipment, storage medium, and product for generating financing schemes, relating to the field of financial services technology. The disclosed method includes: acquiring the default probabilities and supply chain data of each enterprise in the supply chain; using the default probabilities and supply chain data of each enterprise as input values ​​for a simulated annealing algorithm, performing simulated annealing calculations with the financing objectives of maximizing liquidity, minimizing financing costs, and minimizing default risk simultaneously, to obtain neighborhood solutions; determining the temperature decrease rate of the current annealing temperature based on the improvement of the neighborhood solutions; adjusting the current annealing temperature based on the temperature decrease rate, and re-performing the simulated annealing calculations based on the adjusted annealing temperature until all financing objectives reach equilibrium; and generating target financing schemes for each enterprise in the supply chain based on the neighborhood solutions corresponding to the equilibrium of each financing objective. This achieves comprehensive optimization of supply chain financing.

Description

Technical Field

[0001] This application relates to the field of financial services technology, and in particular to methods, apparatus, equipment, storage media and products for generating financing schemes. Background Technology

[0002] With the rapid development of supply chain finance, machine learning technology has gradually become an important tool in supply chain financing. Through big data analysis and automated default risk prediction, machine learning helps financial institutions conduct credit assessments and financing decisions more effectively. However, the complexity and dynamism of supply chain financing pose challenges to traditional machine learning algorithms in handling data complexity and multi-objective optimization. Against this backdrop, default probability prediction based on machine learning offers a new approach to optimizing supply chain financing.

[0003] Machine learning technology, through the analysis of vast amounts of historical data, can generate default probability prediction models for each enterprise in the supply chain, thereby helping financial institutions to manage financing risks more accurately. It can effectively solve the problems of strong data dependence and complex manual operations inherent in traditional methods, and improve the efficiency of financing decisions through automated processing. However, existing machine learning algorithms in supply chain financing optimization typically only solve single-objective optimization problems, resulting in insufficiently comprehensive financing solutions. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, device, storage medium, and product for generating financing schemes, aiming to solve the technical problem that current supply chain financing optimization targeting a single financing objective results in incomplete financing schemes.

[0005] To achieve the above objectives, this application proposes a method for generating a financing scheme, comprising:

[0006] Obtain default probabilities and supply chain data for each enterprise in the supply chain;

[0007] Using the default probability and supply chain data of each enterprise as input values ​​for the simulated annealing algorithm, and taking maximizing capital liquidity, minimizing financing costs, and minimizing default risk as the financing objectives of the simulated annealing algorithm, the simulated annealing operation is performed to obtain the neighborhood solution.

[0008] Based on the improvement of the neighborhood solution, determine the rate of temperature decrease of the current annealing temperature in the simulated annealing algorithm;

[0009] The current annealing temperature is adjusted based on the rate of temperature decrease, and the simulated annealing calculation is performed again based on the adjusted annealing temperature until all financing objectives reach equilibrium. Then, the target financing plan for each enterprise in the supply chain is generated based on the neighborhood solution corresponding to the equilibrium of each financing objective.

[0010] In one embodiment, determining the rate of temperature decrease of the current annealing temperature in the simulated annealing algorithm based on the improvement of the neighborhood solution includes:

[0011] When the improvement of the neighborhood solution is the first improvement, the first temperature decrease rate is determined to be the temperature decrease rate of the current annealing temperature;

[0012] When the improvement of the neighborhood solution is the second improvement, the second temperature decrease rate is determined to be the temperature decrease rate of the current annealing temperature;

[0013] The rate of temperature decrease in the first temperature is less than the rate of temperature decrease in the second temperature.

[0014] In one embodiment, the supply chain data includes the financing amount, credit score, and cash flow of each enterprise. The default probability of each enterprise and the supply chain data are used as input values ​​for the simulated annealing algorithm. The simulated annealing algorithm is performed with the financing objectives of maximizing liquidity, minimizing financing costs, and minimizing default risk, resulting in neighborhood solutions including:

[0015] During the simulated annealing process, the financing cost of each enterprise is determined based on its credit score. Based on the financing costs and financing amounts of each enterprise, the total financing cost of all enterprises is determined. Minimizing the financing cost is the minimum total financing cost.

[0016] Obtain the cash flow of each enterprise, and determine the total liquidity of all enterprises based on the cash flow and financing amount of each enterprise. Maximizing liquidity means maximizing the total liquidity.

[0017] Based on the default probability and financing amount of each enterprise, the total default risk of all enterprises is determined. Minimizing the default risk is the minimum total default risk.

[0018] The neighborhood solution is obtained based on the total financing cost, total liquidity, and total default risk of all firms.

[0019] In one embodiment, the default probability and supply chain data of each enterprise are used as input values ​​for the simulated annealing algorithm. The simulated annealing operation is performed with the financing objectives of maximizing liquidity, minimizing financing costs, and minimizing default risk simultaneously. After obtaining the neighborhood solution, the algorithm further includes:

[0020] The current total liquidity, current total financing cost, and current total default risk are obtained from the neighborhood solution.

[0021] Determine the optimal performance of each financing objective among the current total liquidity, current total financing cost, and current total default risk, and identify the financing objectives that are not sufficiently optimized based on the optimal performance of each financing objective.

[0022] A perturbation is set for insufficiently optimized financing targets so that they are optimized first during the simulated annealing process.

[0023] In one embodiment, the default probability and supply chain data of each enterprise are used as input values ​​for the simulated annealing algorithm. The simulated annealing operation is performed with the financing objectives of maximizing liquidity, minimizing financing costs, and minimizing default risk simultaneously. After obtaining the neighborhood solution, the algorithm further includes:

[0024] The current total liquidity, current total financing cost, and current total default risk are obtained from the neighborhood solution.

[0025] Determine the degree of optimization for each financing objective among current total liquidity, current total financing cost, and current total default risk;

[0026] Based on the degree of optimization corresponding to each financing objective, determine the adjustment amount of the current weight parameters of each financing objective;

[0027] Based on the adjustment amount of the current weight parameters of each financing objective, the current weight parameters of the corresponding financing objective are adjusted, and the simulated annealing operation is re-performed based on the financing objective with adjusted weight parameters.

[0028] In one embodiment, the method for generating a financing plan further includes:

[0029] Obtain the total financing limit, the financing limit for each company, and the financing limit for each company;

[0030] Based on the upper limit of the total financing amount, the upper limit of the financing amount for each enterprise, and the lower limit of the financing amount for each enterprise, constraints are set in the simulated annealing operation to constrain the operation of the simulated annealing algorithm.

[0031] The constraints include that the total financing cost of all enterprises is less than or equal to the upper limit of the total financing amount, and that the financing amount of each enterprise is greater than or equal to the lower limit of the financing amount and greater than or equal to the upper limit of the financing amount.

[0032] Furthermore, to achieve the above objectives, this application also proposes a financing scheme generation apparatus, comprising:

[0033] The acquisition module is used to acquire default probabilities and supply chain data for each enterprise in the supply chain.

[0034] The solution module is used to take the default probability and supply chain data of each enterprise as input values ​​for the simulated annealing algorithm. It performs simulated annealing calculations to maximize capital liquidity, minimize financing costs, and minimize default risk, while also taking these as the financing objectives of the simulated annealing algorithm, and obtains neighborhood solutions.

[0035] The temperature decrease rate determination module is used to determine the temperature decrease rate of the current annealing temperature in the simulated annealing algorithm based on the improvement of the neighborhood solution.

[0036] The financing scheme generation module is used to adjust the current annealing temperature based on the rate of temperature decrease, and to re-perform the simulated annealing calculation based on the adjusted annealing temperature until all financing objectives reach a balance. Based on the neighborhood solution corresponding to the balance of each financing objective, the module generates the target financing scheme for each enterprise in the supply chain.

[0037] In addition, to achieve the above objectives, this application also proposes a financing scheme generation apparatus, the apparatus comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the financing scheme generation method described above.

[0038] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the financing scheme generation method described above.

[0039] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the financing scheme generation method described above.

[0040] This application, by acquiring the default probabilities and supply chain data of each enterprise in the supply chain, sets multiple financing objectives such as maximizing cash flow, minimizing financing costs, and minimizing default risk. It can simultaneously optimize these objectives using a simulated annealing algorithm, maximizing cash flow and minimizing financing costs while ensuring risk control, thus achieving comprehensive optimization of supply chain financing. Furthermore, during the optimization process, the temperature decrease rate of the simulated annealing algorithm is adjusted. Through adaptive temperature adjustment, the annealing speed is dynamically controlled, ensuring that the simulated annealing algorithm retains sufficient randomness during the exploration phase and accelerates convergence when the optimization tends to stabilize. This improves the quality and convergence speed of the solution, resulting in a final financing scheme that is not only comprehensive but also relatively accurate. Attached Figure Description

[0041] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0042] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 A flowchart illustrating the first embodiment of the method for generating the financing scheme of this application;

[0044] Figure 2 A detailed flowchart illustrating step S20 of the first embodiment of the method for generating the financing scheme of this application;

[0045] Figure 3 The flowchart provided in the second embodiment of the method for generating the financing plan for this application is as follows;

[0046] Figure 4 The flowchart provided in the third embodiment of the method for generating the financing plan in this application is as follows;

[0047] Figure 5 The flowchart provided in the fourth embodiment of the method for generating the financing plan of this application is as follows;

[0048] Figure 6 This is a schematic diagram of the module structure of the financing scheme generation device according to an embodiment of this application;

[0049] Figure 7 This is a schematic diagram of the structure of the device for generating the financing scheme in the embodiments of this application.

[0050] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0051] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0052] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0053] With the rapid development of supply chain finance, machine learning technology has gradually become an important tool in supply chain financing. Through big data analysis and automated default risk prediction, machine learning helps financial institutions conduct credit assessments and financing decisions more effectively. However, the complexity and dynamism of supply chain financing pose challenges to traditional machine learning algorithms in handling data complexity and multi-objective optimization. Against this backdrop, default probability prediction based on machine learning offers a new approach to optimizing supply chain financing.

[0054] Machine learning technology, through the analysis of vast amounts of historical data, can generate default probability prediction models for each enterprise in the supply chain, thereby helping financial institutions to manage financing risks more accurately. It can effectively solve the problems of strong data dependence and complex manual operations inherent in traditional methods, and improve the efficiency of financing decisions through automated processing. However, existing machine learning algorithms in supply chain financing optimization typically only solve single-objective optimization problems, resulting in insufficiently comprehensive financing solutions.

[0055] To address the aforementioned issues, this application proposes a method for generating financing schemes. The main technical solution includes: acquiring the default probabilities and supply chain data of each enterprise in the supply chain; using the default probabilities and supply chain data of each enterprise as input values ​​for a simulated annealing algorithm, performing simulated annealing calculations with the financing objectives of maximizing liquidity, minimizing financing costs, and minimizing default risk simultaneously, to obtain neighborhood solutions; determining the temperature decrease rate of the current annealing temperature based on the improvement of the neighborhood solutions; adjusting the current annealing temperature based on the temperature decrease rate, and re-performing the simulated annealing calculations based on the adjusted annealing temperature until all financing objectives reach equilibrium; and generating target financing schemes for each enterprise in the supply chain based on the neighborhood solutions corresponding to the equilibrium of each financing objective.

[0056] This application, by acquiring the default probabilities and supply chain data of each enterprise in the supply chain, sets multiple financing objectives such as maximizing cash flow, minimizing financing costs, and minimizing default risk. It can simultaneously optimize these objectives using a simulated annealing algorithm, maximizing cash flow and minimizing financing costs while ensuring risk control, thus achieving comprehensive optimization of supply chain financing. Furthermore, during the optimization process, the temperature decrease rate of the simulated annealing algorithm is adjusted. Through adaptive temperature adjustment, the annealing speed is dynamically controlled, ensuring that the simulated annealing algorithm retains sufficient randomness during the exploration phase and accelerates convergence when the optimization tends to stabilize. This improves the quality and convergence speed of the solution, resulting in a final financing scheme that is not only comprehensive but also relatively accurate.

[0057] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or a financing scheme generation device capable of achieving the above functions. The following description uses a financing scheme generation device as an example to illustrate this embodiment and the subsequent embodiments.

[0058] It should be noted that supply chain finance is a financing model that integrates core enterprises in the supply chain and their upstream and downstream supporting enterprises, and formulates overall financial solutions based on transaction relationships and industry characteristics.

[0059] Based on this, embodiments of this application provide a method for generating a financing scheme, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the method for generating the financing scheme of this application.

[0060] In this embodiment, the method for generating the financing plan includes steps S10 to S40:

[0061] Step S10: Obtain the default probability and supply chain data of each enterprise in the supply chain.

[0062] Default probability is a key indicator for quantifying credit risk, reflecting a company's creditworthiness and repayment ability. Different companies have corresponding default probabilities, and these probabilities may vary.

[0063] Supply chain data includes each company's financing amount, credit score, and cash flow, as well as its revenue and profit. This supply chain data originates from multi-dimensional company information in supply chain financing. The data collection scope includes financial information, transaction data, credit scores, and external risk factors, comprehensively covering a company's operational performance within the supply chain. Data types include company name, revenue, profit, cash flow, and debt ratio.

[0064] After collecting multi-dimensional enterprise information data, the collected data is preprocessed to obtain preprocessed supply chain data. Then, based on the preprocessed supply chain data, the default probability of each enterprise in the supply chain is calculated.

[0065] In one feasible implementation, during the data preprocessing of the collected multi-dimensional enterprise information data, different types of supply chain data are processed using different preprocessing methods according to their type.

[0066] For example, for numerical supply chain data such as revenue, profit, cash flow, debt ratio, transaction amount, transaction frequency, and number of employees, the median can be used to impute missing values ​​to avoid extreme values ​​affecting the overall data distribution. During standardization, numerical data can be standardized to have a mean of 0 and a standard deviation of 1 to prevent excessive differences in feature value ranges from affecting model training. For categorical supply chain data such as industry codes, the mode can be used to impute missing values, i.e., using the category with the highest frequency to fill in the missing values. During one-hot encoding, categorical data can be converted into binary vectors to facilitate the model's identification of the independence between categories. For continuous supply chain data such as credit scores and external risk factors, the mean can be used to impute missing values ​​to maintain data continuity. During standardization, continuous data can be standardized to ensure a uniform data distribution. The following example illustrates the preprocessing steps for industry codes, which are categorical supply chain data:

[0067] First, missing values ​​for industry codes are handled. Mode imputation is used, which selects the most common category among the industry codes as the imputation items for missing values. The specific formula is as follows:

[0068] x 填补 =mode(x).

[0069] Where x is the industry code column, and mode(x) represents the category that appears most frequently in this column.

[0070] To make industry codes suitable for the improved logistic regression model, this application employs one-hot encoding to convert categorical data into binary vectors. For example, assuming the industry codes contain four categories: A, B, C, and D, the result of one-hot encoding is:

[0071]

[0072] This encoding method avoids the issue of sequentiality between categorical supply chain data, ensuring that the model can correctly understand the independence of different categories. Assume the industry code column is... x The encoding process is as follows:

[0073]

[0074] This encoding method converts each category into a numerical form, which can be effectively applied to subsequent model training.

[0075] In another feasible implementation, after obtaining the preprocessed supply chain data, the default probability of each enterprise in the supply chain is obtained based on the preprocessed supply chain data, including: predicting the default probability using an improved logistic regression model on the preprocessed supply chain data. After the supply chain data preprocessing is completed, the next step is to predict the default rate of enterprises based on the improved logistic regression model. The logistic regression model is suitable for classification problems, calculating the default probability of enterprises by analyzing the characteristic data of each enterprise in the supply chain. This embodiment improves upon the traditional logistic regression model by introducing regularization and ensemble learning techniques to enhance the robustness and prediction accuracy of the model.

[0076] The core of logistic regression models is using a logistic function to fit the probability output of a classification problem. For a given firm's feature X, its default probability P(y = 1|X) is expressed as:

[0077]

[0078] Where: X = (X1, X2 , ..., X n ) represents the firm's characteristic vector, such as revenue, profit, and debt ratio. β0 is the bias term. β1, β2, ..., β n These are the weight parameters of the model. e is the natural constant. The logistic regression model learns to fit the data by training on the weights β, ultimately predicting the probability of corporate default.

[0079] In another feasible implementation, after obtaining the preprocessed supply chain data, default probability can be predicted using classification algorithms such as decision trees and support vector machines.

[0080] In another feasible implementation, when processing high-dimensional supply chain data, L2 regularization, also known as ridge regression, is introduced to constrain the complexity of the logistic regression model. To avoid overfitting, especially when dealing with high-dimensional supply chain data, L2 regularization, also known as ridge regression, is introduced to constrain the complexity of the logistic regression model and prevent excessive weighting. L2 regularization adds a regularization term to the loss function, and its optimization objective becomes:

[0081]

[0082] Where m is the number of samples. λ is the regularization parameter used to control model complexity. β j It is the regression coefficient of the j-th feature. Regularization term. The weights are limited to prevent the coefficients of certain features from becoming too large, which could lead to overfitting of the model.

[0083] In summary, traditional logistic regression models are prone to overfitting when dealing with high-dimensional and complex data. This application enhances the model's generalization ability by introducing L2 regularization, avoiding overfitting caused by excessive feature weights, thereby improving the prediction accuracy of default probability. Furthermore, compared to traditional logistic regression models, the improved logistic regression model in this application can better handle high-dimensional and complex data in supply chain finance, especially multi-dimensional financial, credit, and operational data, thus enabling more accurate predictions of default risk.

[0084] In practical applications, after training with the logistic regression model, given a new corporate feature vector X, the logistic regression model will output the corporate default probability P(y=1|X).

[0085]

[0086] Where β0 is the bias term, β j It is feature X j The weights, X j This represents the standardized data of each feature value in the input feature matrix, such as revenue, profit, and debt ratio.

[0087] After training the logistic regression model, the obtained β parameter is used to calculate the specific default probability P(y=1|X) for each enterprise. This default probability is a value between 0 and 1, and is used in the subsequent supply chain financing optimization process. After improving the default rate prediction of the logistic regression model, the default probability of each enterprise is obtained.

[0088] Step S20: Using the default probability and supply chain data of each enterprise as input values ​​for the simulated annealing algorithm, and taking maximizing capital liquidity, minimizing financing costs, and minimizing default risk as the financing objectives of the simulated annealing algorithm, the simulated annealing operation is performed to obtain the neighborhood solution.

[0089] Simulated annealing is a probabilistic optimization algorithm designed to find the global optimum within a vast solution space. Its principle is based on the solid-state annealing process, where the internal atoms of a solid undergo a change from disordered to ordered arrangement as it is heated and slowly cooled to room temperature. By setting an initial annealing temperature and gradually decreasing it, and by accepting a new solution with a certain probability at each step according to the Metropolis criterion (which may be worse than the current solution), the algorithm aims to escape local optima and explore the global optimum.

[0090] In this embodiment, simulated annealing is a stochastic optimization algorithm suitable for solving multi-objective optimization problems. It searches for the optimal solution step by step by simulating the physical annealing process. In multi-objective programming, simulated annealing can be used to find the optimal financing scheme among multiple objectives.

[0091] In supply chain financing optimization, by analyzing the default probability of enterprises and related supply chain data, multi-objective programming can be constructed to help financial institutions weigh multiple objectives and optimize financing decisions. Minimizing financing costs means minimizing total financing costs to ensure the lowest possible cost of capital utilization; maximizing liquidity means maximizing total liquidity while optimizing capital liquidity to meet enterprise needs; minimizing default risk means minimizing total default risk by rationally allocating financing amounts based on predicted default rates to reduce overall financing risk.

[0092] The neighborhood solution can include the total financing cost of all firms, the total liquidity of all firms, and the total default risk of all firms.

[0093] In one feasible implementation, the default probability and supply chain data of each enterprise are used as input values ​​for the simulated annealing algorithm to maximize capital liquidity, minimize financing costs, and minimize default risk, while also serving as the financing objectives of the simulated annealing algorithm. The initial annealing temperature is used as the current annealing temperature for the simulated annealing algorithm to perform simulated annealing calculations and obtain neighborhood solutions.

[0094] In another feasible implementation, the default probabilities and supply chain data of each enterprise are used as input values ​​for the simulated annealing algorithm. The goal of the simulated annealing algorithm is to maximize capital liquidity, minimize financing costs, and minimize default risk. The initial annealing temperature is used as the current annealing temperature for the simulated annealing algorithm to obtain a new solution. The objective function value of the new solution is determined and compared with the objective function value of the initial solution. If the objective function value of the new solution is greater than or equal to the objective function value of the initial solution, the new solution is taken as a neighborhood solution. If the objective function value of the new solution is less than the objective function value of the initial solution, the new solution is accepted with a certain probability according to the Metropolis criterion.

[0095] By acquiring default probabilities and supply chain data from various companies in the supply chain, and setting multiple financing objectives such as maximizing cash flow, minimizing financing costs, and minimizing default risk, the simulated annealing algorithm can simultaneously optimize these objectives. This ensures that cash flow is maximized and financing costs are minimized while maintaining risk control, thus achieving comprehensive optimization of supply chain financing.

[0096] Step S30: Determine the rate of temperature decrease of the current annealing temperature in the simulated annealing algorithm based on the improvement of the neighborhood solution.

[0097] The improvement of the neighborhood solution includes a first improvement scenario and a second improvement scenario. The first improvement scenario is continuously obtaining better solutions. For example, it can be determined that continuously obtaining better solutions means that the objective function value of the neighborhood solution is greater than or equal to the objective function value of the initial solution for a consecutive number of times. The second improvement scenario is that the solution cannot be significantly improved. For example, it can be determined that the solution cannot be significantly improved means that the objective function value of the neighborhood solution is less than the objective function value of the initial solution.

[0098] The temperature decrease rate is not fixed but dynamically adjusted based on the improvement of neighborhood solutions. The temperature decrease rate of the simulated annealing algorithm varies depending on the improvement of neighborhood solutions. Because traditional simulated annealing algorithms use a fixed temperature decrease method, they are prone to prematurely getting trapped in local optima during the exploration space. This application uses adaptive temperature adjustment to dynamically control the annealing rate, ensuring sufficient randomness during the exploration phase and accelerating convergence when the optimization tends to stabilize, thereby improving the quality of the solution and the convergence speed.

[0099] In one feasible implementation, when the improvement of the neighborhood solution is the first improvement case, the temperature decrease rate of the current annealing temperature is slowly reduced to maintain the exploration space; when the improvement of the neighborhood solution is the second improvement case, the temperature decrease rate of the current annealing temperature is increased to accelerate convergence.

[0100] Step S40: Adjust the current annealing temperature based on the rate of temperature decrease, and perform simulated annealing calculation again based on the adjusted annealing temperature until all financing targets reach a balance. Then, generate the target financing plan for each enterprise in the supply chain based on the neighborhood solution corresponding to the balance of each financing target.

[0101] After determining the temperature drop rate, the current annealing temperature is adjusted based on the temperature drop rate, so that the current annealing temperature can be adjusted based on the temperature drop rate. Subsequently, the simulated annealing calculation is performed again based on the adjusted annealing temperature. In each iteration, the simulated annealing is performed in the above manner until all financing objectives reach a balance. Based on the neighborhood solutions corresponding to the balance of each financing objective, the target financing plan for each enterprise in the supply chain is generated.

[0102] Different companies have different target financing plans. The optimal financing plan, optimized by the simulated annealing algorithm, combines key data such as the default probability, financing limit, and financing cost of each company, and provides the optimal financing allocation plan.

[0103] In this embodiment, by acquiring the default probability and supply chain data of each enterprise in the supply chain, and setting multiple financing objectives such as maximizing cash flow, minimizing financing costs, and minimizing default risk, the simulated annealing algorithm can simultaneously optimize multiple financing objectives. This ensures that cash flow is maximized and financing costs are minimized while maintaining risk control, achieving comprehensive optimization of supply chain financing. Furthermore, during the optimization process, the rate of temperature decrease in the simulated annealing algorithm is adjusted. Through adaptive temperature adjustment, the annealing speed is dynamically controlled, ensuring that the simulated annealing algorithm retains sufficient randomness during the exploration phase and accelerates convergence when the optimization tends to stabilize. This improves the quality and convergence speed of the solution, resulting in a financing scheme that is not only comprehensive but also relatively accurate.

[0104] In some embodiments of this application, the traditional simulated annealing algorithm uses a fixed temperature decrease method, which can easily lead to premature entrapment in local optima during the exploration space. To address this problem, this application determines the temperature decrease rate of the current annealing temperature for the simulated annealing algorithm based on the improvement of neighborhood solutions, including:

[0105] Step S31: When the improvement of the neighborhood solution is the first improvement, determine the first temperature decrease rate as the temperature decrease rate of the current annealing temperature.

[0106] Step S32: When the improvement of the neighborhood solution is the second improvement, determine the second temperature decrease rate as the temperature decrease rate of the current annealing temperature.

[0107] The rate of temperature decrease in the first temperature is less than the rate of temperature decrease in the second temperature.

[0108] The improvement of the neighborhood solution includes a first improvement scenario and a second improvement scenario. The first improvement scenario is continuously obtaining better solutions. For example, it can be determined that continuously obtaining better solutions means that the objective function value of the neighborhood solution is greater than or equal to the objective function value of the initial solution for a consecutive number of times. The second improvement scenario is that the solution cannot be significantly improved. For example, it can be determined that the solution cannot be significantly improved means that the objective function value of the neighborhood solution is less than the objective function value of the initial solution.

[0109] When the improvement of the neighborhood solution is the first improvement, that is, when good solutions are obtained continuously and the neighborhood solution is significantly improved, the first temperature decrease rate can be obtained by multiplying the first weight coefficient and the current annealing temperature, and this first temperature decrease rate is determined as the temperature decrease rate of the current annealing temperature. The first weight coefficient can be set to 0.9.

[0110] When the improvement of the neighborhood solution is the second improvement case, that is, the solution cannot be significantly improved, the second temperature decrease rate can be obtained by multiplying the second weighting coefficient and the current annealing temperature, and the second temperature decrease rate can be determined as the temperature decrease rate of the current annealing temperature. The second weighting coefficient can be set to 0.5.

[0111] In this embodiment, the annealing speed is dynamically controlled by adaptive temperature adjustment, ensuring that the simulated annealing algorithm retains sufficient randomness during the exploration phase, while accelerating convergence when the optimization tends to stabilize, thereby improving the quality and convergence speed of the neighborhood solution.

[0112] Reference Figure 2 In some embodiments of this application, supply chain data includes the financing amount, credit score, and cash flow of each enterprise. The default probability of each enterprise and the supply chain data are used as input values ​​for the simulated annealing algorithm. The simulated annealing operation is performed with the financing objectives of maximizing liquidity, minimizing financing costs, and minimizing default risk simultaneously, yielding neighborhood solutions including:

[0113] Step S21: During the simulated annealing operation, the financing cost of each enterprise is determined based on its credit score. Based on the financing costs and financing amounts of each enterprise, the total financing cost of all enterprises is determined. Minimizing the financing cost is the minimum total financing cost.

[0114] Financing cost C i The cost of financing is dynamically adjusted based on the company's credit score and default probability. It is assumed that the higher the company's credit score, the lower the financing cost. The financing cost can be calculated using the following formula, assuming a minimum financing cost of 5% for a credit score of 850 and a maximum financing cost of 15% for a credit score of 300:

[0115]

[0116] For example, if a company has a credit score of 650, then its financing cost is:

[0117]

[0118] Each company can use the above method to calculate its corresponding financing cost.

[0119] After obtaining the financing cost for each enterprise, the financing amount for each enterprise can be determined based on supply chain data. For a single enterprise, its financing cost and corresponding financing amount can be multiplied to obtain that enterprise's financing cost. Finally, the financing costs of each enterprise are summed to obtain the total financing cost of all enterprises. The simulated annealing algorithm achieves the financing cost minimization objective by minimizing the total financing cost of all enterprises.

[0120] For example, the financing objective is to minimize the financing costs for all firms. Assume each firm i has a financing amount of L. i The financing cost is C i Then the objective function for minimizing financing costs is:

[0121]

[0122] Among them, C i It can be dynamically adjusted based on the company's credit score and probability of default.

[0123] Step S22: Obtain the cash flow of each enterprise. Based on the cash flow and financing amount of each enterprise, determine the total liquidity of all enterprises. Maximizing liquidity means maximizing the total liquidity.

[0124] Cash flow F i This is an indicator of a company's cash flow. This indicator is directly taken from pre-processed supply chain data (in thousands of yuan) and used as a weight in the cash flow target during optimization. For example, if a company's cash flow is 30 million yuan, then F... i =30 million yuan.

[0125] After obtaining the cash flow for each enterprise, the financing amount for each enterprise can be determined based on supply chain data. For a single enterprise, its corresponding cash flow and its corresponding financing amount can be multiplied to obtain its liquidity. Finally, the liquidity of each enterprise is summed to obtain the total liquidity of all enterprises. The simulated annealing algorithm maximizes the financing objective of maximizing liquidity by maximizing the total liquidity of all enterprises.

[0126] For example, maximizing cash flow aims to optimize cash allocation and ensure efficient cash flow throughout the supply chain. Let cash flow be related to the firm's cash flow F. i If they are related, then the objective function is:

[0127]

[0128] Among them, liquidity and allocated financing amount L i and the company's cash flow Fi Proportional.

[0129] Step S23: Determine the total default risk of all companies based on their default probability and financing amount. Minimizing the default risk is the minimum total default risk.

[0130] Default probability P i The parameters, predicted by a logistic regression model and ranging from 0 to 1, represent the likelihood of a firm's default. These parameters will be used as input to the simulated annealing algorithm to optimize financing strategies, aiming to maximize liquidity, minimize financing costs, and control default risk.

[0131] After obtaining the default probability for each enterprise, the financing amount for each enterprise can be determined based on supply chain data. For a single enterprise, its default probability and corresponding financing amount can be multiplied to obtain its default risk. Finally, the default risks of each enterprise are summed to obtain the total default risk of all enterprises. The simulated annealing algorithm achieves the financing objective of minimizing default risk by minimizing the total default risk of all enterprises.

[0132] For example, to minimize default risk, financial institutions need to determine the default probability P of a company. i To allocate financing amounts reasonably. The financing objective is to control default risk within an acceptable range, i.e., to minimize total default risk. The objective function for default risk control can be defined as:

[0133]

[0134] Step S24: Based on the total financing cost of all enterprises, the total liquidity of all enterprises, and the total default risk of all enterprises, obtain the neighborhood solution.

[0135] In this embodiment, the total financing cost, total liquidity, and total default risk of all enterprises are determined by the above method. Then, based on the total financing cost, total liquidity, and total default risk of all enterprises, a neighborhood solution is obtained. This process of finding the neighborhood solution considers multiple factors and improves the accuracy of the neighborhood solution.

[0136] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. On this basis, refer to Figure 3Using the default probabilities and supply chain data of each enterprise as input values ​​for the simulated annealing algorithm, and taking maximizing cash flow, minimizing financing costs, and minimizing default risk as the financing objectives of the simulated annealing algorithm, the simulated annealing operation is performed. After obtaining the neighborhood solution, the following steps are also included:

[0137] Step S110: Obtain the current total liquidity, current total financing cost, and current total default risk based on the neighborhood solution.

[0138] Step S120: Determine the optimal performance of each financing objective among the current total liquidity, current total financing cost, and current total default risk, and determine the financing objectives that are not sufficiently optimized based on the optimal performance of each financing objective.

[0139] Step S130: Set a perturbation amount for the insufficiently optimized financing target so that the insufficiently optimized financing target will be optimized first during the simulated annealing operation.

[0140] In the improved simulated annealing algorithm, the generation of neighborhood solutions incorporates a multi-objective-oriented weight adjustment strategy, aiming to prioritize the optimization of financing objectives that currently perform poorly. For example, if the financing cost of the current solution is high, new solutions are generated preferentially in the direction of reducing financing costs.

[0141] Neighborhood solution generation formula:

[0142] L′ i =L i +w i ·Δ i .

[0143] Among them, w i It is a multi-objective-oriented weighting system that, based on the performance of the current solution, prioritizes adjustments to financing objectives that have not been sufficiently optimized. Δ i It is a perturbation quantity used to generate neighborhood solutions in the solution space.

[0144] By adding perturbations to under-optimized financing objectives, the priority of those objectives can be increased.

[0145] In one feasible implementation, determining an underoptimized financing objective can be done by: obtaining the neighborhood solutions after two consecutive optimizations, comparing the results corresponding to each financing objective in the currently optimized neighborhood solution with the results corresponding to each financing objective in the previously optimized neighborhood solution, and then determining the optimization performance of each financing objective in the neighborhood solution based on the comparison results.

[0146] Specifically, the difference between the results for each financing objective in the currently optimized neighborhood solution and the results for the corresponding financing objective in the previously optimized neighborhood solution can be used to characterize the optimization performance of each financing objective in the neighborhood solution. A larger difference indicates a worse current optimization performance of the financing objective, while a smaller difference indicates a better current optimization performance. This allows for the identification of insufficiently optimized financing objectives based on their optimization performance.

[0147] In this embodiment, the optimal performance of each financing objective among current total liquidity, current total financing cost, and current total default risk is determined, and insufficiently optimized financing objectives are identified based on their performance. Perturbations are set for these insufficiently optimized financing objectives to prioritize their optimization during simulated annealing. By comparing the optimization performance of each financing objective, it is possible to identify which objectives have not yet reached their optimal state, thereby determining their optimization priority. Identifying insufficiently optimized financing objectives allows companies to develop more targeted improvement measures to enhance financing efficiency.

[0148] Based on the first embodiment of this application, in the third embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description and will not be repeated hereafter. On this basis, the neighborhood solution generation process of the traditional simulated annealing algorithm is a random perturbation and does not specifically consider the direction of multi-objective optimization. To address this problem, refer to... Figure 4 Using the default probabilities and supply chain data of each enterprise as input values ​​for the simulated annealing algorithm, and taking maximizing cash flow, minimizing financing costs, and minimizing default risk as the financing objectives of the simulated annealing algorithm, the simulated annealing operation is performed. After obtaining the neighborhood solution, the following steps are also included:

[0149] Step S210: Obtain the current total liquidity, current total financing cost, and current total default risk based on the neighborhood solution.

[0150] Step S220: Determine the degree of optimization for each financing objective among the current total liquidity, current total financing cost, and current total default risk.

[0151] Step S230: Determine the adjustment amount of the current weight parameters of each financing objective based on the degree of optimization corresponding to each financing objective.

[0152] Step S240: Based on the adjustment amount of the current weight parameters of each financing target, adjust the current weight parameters of the corresponding financing target, and re-perform the simulated annealing calculation based on the financing target with adjusted weight parameters.

[0153] To further balance the optimization of multiple objectives, a dynamic weight adjustment strategy is adopted, which adaptively adjusts the current weight parameters of each financing objective based on the degree of optimization of the financing objective during the optimization process. For example, if the default risk is high, the weight of the default risk financing objective is increased accordingly, so that more attention is paid to optimizing this financing objective in subsequent iterations.

[0154] The adjustment formula for the current weight parameters of each financing target is as follows:

[0155]

[0156] in, It is the first j The financing target is in the first k The value in the next iteration This represents the expected value of the financing target. Combining the improved simulated annealing algorithm described above, the multi-objective optimization model of this application is defined as follows:

[0157]

[0158] Among them, α1, α2, and α3 are weight parameters of the dynamically adjusted financing objectives, corresponding to financing costs, liquidity, and default risk, respectively.

[0159] In this embodiment, a multi-objective-oriented neighborhood solution generation strategy is used to adjust the weights of multiple optimization objectives such as financing costs, liquidity, and default risk, so that the optimization process can find a balanced solution that satisfies multiple objectives more efficiently.

[0160] Based on the first and second embodiments of this application, in the fourth embodiment of this application, the content that is the same as or similar to the above embodiments can be referred to the above description, and will not be repeated hereafter. On this basis, refer to Figure 5 The methods for generating financing plans also include:

[0161] Step S310: Obtain the total financing limit, the financing limit for each enterprise, and the financing limit for each enterprise.

[0162] Total financing limit L max This is the maximum funding limit set by financial institutions for the entire supply chain financing pool. This value can be set based on factors such as market size and the size of the financing pool. Let's assume the maximum financing amount provided in the supply chain is 1 billion RMB (i.e., 100,000,000 yuan).

[0163] L max =100,000,000 yuan.

[0164] Financing limit for a single enterprise L max_iThe maximum amount of financing a company can obtain is determined based on factors such as its credit score, financial performance, and cash flow. Typically, a company's maximum financing amount can be linked to its revenue or cash flow. Assuming the financing amount does not exceed 50% of the company's annual revenue, the maximum financing amount L for a single company is... max_i It can be calculated using the following formula:

[0165] L max_i = 0.5 × company revenue.

[0166] For example, if a company has an annual revenue of 10 million yuan, its financing limit is 5 million yuan.

[0167] To prevent providing too little financing to small businesses, a minimum financing limit L is set for each individual business. min For example, setting a minimum financing amount of 500,000 yuan ensures that each enterprise can obtain certain financial support.

[0168] L min =500,000 yuan.

[0169] Step S320: Based on the upper limit of the total financing amount, the upper limit of the financing amount for each enterprise, and the lower limit of the financing amount for each enterprise, set the constraints in the simulated annealing operation to constrain the operation of the simulated annealing algorithm based on the constraints.

[0170] The constraints include that the total financing cost of all enterprises is less than or equal to the upper limit of the total financing amount, and that the financing amount of each enterprise is greater than or equal to the lower limit of the financing amount and greater than or equal to the upper limit of the financing amount.

[0171] During the optimization process, the following constraints must also be met: the total financing amount shall not exceed the upper limit L of the total financing amount. max :

[0172]

[0173] L, the financing amount for each enterprise i Limitations should be imposed based on their needs and risks:

[0174] L min ≤L i ≤L max_i .

[0175] Among them, L min It is the lower limit of the financing amount for each enterprise, L max_i It is the upper limit of financing calculated based on the company's credit status.

[0176] In this embodiment, since each enterprise has its own financing needs and limitations, setting upper and lower limits for financing ensures that the actual situation of each enterprise is fully considered during the optimization process, avoiding over-financing or under-financing. Furthermore, in the simulated annealing algorithm, setting constraints can narrow the solution space and reduce the algorithm's search range, thereby improving the algorithm's convergence speed and accuracy. Simultaneously, these constraints can also help the algorithm avoid getting trapped in local optima during the search process, improving its global search capability.

[0177] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the method of generating the financing scheme of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0178] Based on the same inventive concept, this application also provides a device for generating a financing scheme, please refer to... Figure 6 The financing scheme generation device includes:

[0179] Module 10 is used to acquire the default probability and supply chain data of each enterprise in the supply chain.

[0180] The solution module 20 is used to perform simulated annealing operations with the default probability of each enterprise and supply chain data as input values ​​for the simulated annealing algorithm. The simulated annealing algorithm aims to maximize capital liquidity, minimize financing costs, and minimize default risk, and obtain neighborhood solutions.

[0181] The temperature decrease rate determination module 30 is used to determine the temperature decrease rate of the current annealing temperature of the simulated annealing algorithm based on the improvement of the neighborhood solution.

[0182] The financing scheme generation module 40 is used to adjust the current annealing temperature based on the rate of temperature decrease, and to re-perform the simulated annealing calculation based on the adjusted annealing temperature until all financing targets reach a balance. Based on the neighborhood solution corresponding to the balance of each financing target, the module generates the target financing scheme for each enterprise in the supply chain.

[0183] In one embodiment, the temperature drop rate determination module 30 is further configured to: determine a first temperature drop rate as the temperature drop rate of the current annealing temperature when the improvement of the neighborhood solution is a first improvement; and determine a second temperature drop rate as the temperature drop rate of the current annealing temperature when the improvement of the neighborhood solution is a second improvement; wherein the first temperature drop rate is less than the second temperature drop rate.

[0184] In one embodiment, the supply chain data includes the financing amount, credit score, and cash flow of each enterprise. The solution module 20 is further configured to: determine the financing cost of each enterprise based on its credit score during the simulated annealing operation; determine the total financing cost of all enterprises based on their financing costs and financing amounts, wherein minimizing the financing cost is the minimum total financing cost; obtain the cash flow of each enterprise; determine the total liquidity of all enterprises based on their cash flow and financing amounts, wherein maximizing liquidity is the maximum total liquidity; determine the total default risk of all enterprises based on their default probabilities and financing amounts, wherein minimizing the default risk is the minimum total default risk; and obtain the neighborhood solution based on the total financing cost, total liquidity, and total default risk of all enterprises.

[0185] In one embodiment, the financing scheme generation device is further configured to: obtain the current total liquidity, current total financing cost, and current total default risk based on the neighborhood solution; determine the optimization performance of each financing objective among the current total liquidity, current total financing cost, and current total default risk, and determine the insufficiently optimized financing objectives based on the optimization performance of each financing objective; and set a perturbation amount for the insufficiently optimized financing objectives so as to prioritize the optimization of the insufficiently optimized financing objectives during the simulated annealing operation.

[0186] In one embodiment, the financing scheme generation device is further configured to: obtain the current total liquidity, current total financing cost, and current total default risk based on the neighborhood solution; determine the degree of optimization corresponding to each financing objective among the current total liquidity, current total financing cost, and current total default risk; determine the adjustment amount of the current weight parameter of each financing objective based on the degree of optimization corresponding to each financing objective; adjust the current weight parameter of the corresponding financing objective based on the adjustment amount of the current weight parameter of each financing objective, and re-perform simulated annealing operation based on the financing objective with adjusted weight parameters.

[0187] In one embodiment, the financing scheme generation device is further configured to: obtain the total financing limit upper limit, the financing limit upper limit for each enterprise, and the financing limit lower limit for each enterprise; set constraints in the simulated annealing operation based on the total financing limit upper limit, the financing limit upper limit for each enterprise, and the financing limit lower limit for each enterprise, so as to constrain the operation of the simulated annealing algorithm based on the constraints; wherein the constraints include that the total financing cost of all enterprises is less than or equal to the total financing limit upper limit, and that the financing amount of each enterprise is greater than or equal to the financing limit lower limit and greater than or equal to the financing limit upper limit.

[0188] The financing scheme generation apparatus provided in this application, employing the financing scheme generation method described in the above embodiments, can solve the technical problem that current supply chain financing optimization targeting a single financing objective results in incomplete financing schemes. Compared with the prior art, the beneficial effects of the financing scheme generation apparatus provided in this application are the same as those of the financing scheme generation method provided in the above embodiments, and other technical features in the financing scheme generation apparatus are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0189] Based on the same inventive concept, this application provides a financing scheme generation device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to execute the financing scheme generation method in the above embodiments.

[0190] like Figure 7 As shown, the financing scheme generation device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the financing scheme generation device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the financing scheme generating device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows a financing scheme generating device with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0191] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0192] The financing scheme generation device provided in this application, employing the financing scheme generation method described in the above embodiments, can solve the technical problem that current supply chain financing optimization targeting a single financing objective results in incomplete financing schemes. Compared with the prior art, the beneficial effects of the financing scheme generation device provided in this application are the same as those of the financing scheme generation method provided in the above embodiments, and other technical features in this financing scheme generation device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0193] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0194] Based on the same inventive concept, this application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the financing scheme generation method in the above embodiments.

[0195] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0196] The aforementioned computer-readable storage medium may be included in the equipment for generating the financing scheme; or it may exist separately and not assembled into the equipment for generating the financing scheme.

[0197] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by the financing scheme generation device, the financing scheme generation device can maximize cash flow and minimize financing costs while ensuring risk control, thereby achieving comprehensive optimization of supply chain financing.

[0198] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0199] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0200] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0201] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described financing scheme generation method. This solves the technical problem that current supply chain financing optimization targeting a single financing objective results in incomplete financing schemes. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the financing scheme generation method provided in the above embodiments, and will not be repeated here.

[0202] Based on the same inventive concept, this application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the financing scheme generation method described above.

[0203] The computer program product provided in this application can solve the technical problem that current supply chain financing optimization targeting a single financing objective results in incomplete financing solutions. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the financing solution generation method provided in the above embodiments, and will not be repeated here.

[0204] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for generating a financing plan, characterized in that, The method for generating the financing plan includes: Obtain default probabilities and supply chain data for each enterprise in the supply chain; Using the default probability and supply chain data of each enterprise as input values ​​for the simulated annealing algorithm, and taking maximizing capital liquidity, minimizing financing costs, and minimizing default risk as the financing objectives of the simulated annealing algorithm, the simulated annealing operation is performed to obtain the neighborhood solution. Based on the improvement of the neighborhood solution, the temperature decrease rate of the current annealing temperature in the simulated annealing algorithm is determined; The current annealing temperature is adjusted based on the rate of temperature decrease, and the simulated annealing operation is performed again based on the adjusted annealing temperature until all the financing targets reach a balance. Then, the target financing plan for each enterprise in the supply chain is generated based on the neighborhood solution corresponding to the balance of each financing target.

2. The method for generating a financing scheme as described in claim 1, characterized in that, The step of determining the temperature decrease rate of the current annealing temperature in the simulated annealing algorithm based on the improvement of the neighborhood solution includes: When the improvement of the neighborhood solution is the first improvement, the first temperature decrease rate is determined to be the temperature decrease rate of the current annealing temperature; When the improvement of the neighborhood solution is the second improvement, the second temperature decrease rate is determined to be the temperature decrease rate of the current annealing temperature; The first temperature decrease rate is less than the second temperature decrease rate.

3. The method for generating a financing scheme as described in claim 1, characterized in that, The supply chain data includes the financing amount, credit score, and cash flow of each enterprise. The simulated annealing algorithm uses the default probability and supply chain data of each enterprise as input values ​​to maximize liquidity, minimize financing costs, and minimize default risk, simultaneously serving as the financing objectives of the simulated annealing algorithm. The resulting neighborhood solutions include: During the simulated annealing operation, the financing cost of each enterprise is determined based on its credit score. The total financing cost of all enterprises is determined based on the financing cost and financing amount of each enterprise. The minimized financing cost is the minimum total financing cost. Obtain the cash flow of each of the aforementioned enterprises, and determine the total liquidity of all enterprises based on the cash flow and financing amount of each of the aforementioned enterprises, wherein maximizing liquidity means maximizing the total liquidity; Based on the default probability and financing amount of each enterprise, the total default risk of all enterprises is determined, wherein the minimized default risk is the minimum of the total default risk; The neighborhood solution is obtained based on the total financing cost of all current enterprises, the total liquidity of all current enterprises, and the total default risk of all current enterprises.

4. The method for generating a financing scheme as described in claim 3, characterized in that, The method involves using the default probabilities and supply chain data of each enterprise as input values ​​for the simulated annealing algorithm. The simulated annealing operation aims to maximize liquidity, minimize financing costs, and minimize default risk, while simultaneously serving as the financing objectives of the algorithm. After obtaining the neighborhood solution, the method further includes: The current total liquidity, current total financing cost, and current total default risk are obtained from the neighborhood solution. Determine the optimal performance of each financing objective among the current total liquidity, the current total financing cost, and the current total default risk, and determine the financing objectives that are not sufficiently optimized based on the optimal performance of each financing objective; A perturbation is set for the insufficiently optimized financing target so that it is optimized first during the simulated annealing operation.

5. The method for generating a financing scheme as described in claim 3, characterized in that, The method involves using the default probabilities and supply chain data of each enterprise as input values ​​for the simulated annealing algorithm. The simulated annealing operation aims to maximize liquidity, minimize financing costs, and minimize default risk, while simultaneously serving as the financing objectives of the algorithm. After obtaining the neighborhood solution, the method further includes: The current total liquidity, current total financing cost, and current total default risk are obtained from the neighborhood solution. Determine the degree of optimization for each financing objective among the current total liquidity, the current total financing cost, and the current total default risk; Based on the degree of optimization corresponding to each financing objective, determine the adjustment amount of the current weight parameter of each financing objective; Based on the adjustment amount of the current weight parameters of each financing target, the current weight parameters of the corresponding financing target are adjusted, and the simulated annealing operation is performed again based on the financing target with the adjusted weight parameters.

6. The method for generating a financing scheme as described in any one of claims 1 to 5, characterized in that, The method for generating the financing plan also includes: Obtain the total financing limit, the financing limit for each company, and the financing limit for each company; Based on the total financing limit, the financing limit for each enterprise, and the financing limit for each enterprise, constraints are set in the simulated annealing operation to constrain the operation of the simulated annealing algorithm. The constraints include that the total financing cost of all enterprises is less than or equal to the upper limit of the total financing amount, and that the financing amount of each enterprise is greater than or equal to the lower limit of the financing amount and greater than or equal to the upper limit of the financing amount.

7. A device for generating a financing scheme, characterized in that, The apparatus for generating the financing plan includes: The acquisition module is used to acquire default probabilities and supply chain data for each enterprise in the supply chain. The solution module is used to take the default probability and supply chain data of each enterprise as input values ​​for the simulated annealing algorithm, and to perform simulated annealing operation with the goal of maximizing capital liquidity, minimizing financing costs and minimizing default risk as the financing objectives of the simulated annealing algorithm, so as to obtain the neighborhood solution. The temperature decrease rate determination module is used to determine the temperature decrease rate of the current annealing temperature of the simulated annealing algorithm based on the improvement of the neighborhood solution. The financing scheme generation module is used to adjust the current annealing temperature based on the rate of temperature decrease, and to re-perform the simulated annealing calculation based on the adjusted annealing temperature until all the financing targets reach a balance. Based on the neighborhood solutions corresponding to the balance of each financing target, the module generates the target financing scheme for each enterprise in the supply chain.

8. A device for generating a financing scheme, characterized in that, The device for generating the financing scheme includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the method for generating the financing scheme as described in any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the method for generating a financing scheme as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the method for generating a financing scheme as described in any one of claims 1 to 6.

Images