Neural network model-based water conservancy scientific research purchase management method and system

Abstract

The invention relates to the technical field of project management, in particular to a water conservancy scientific research purchase management method and system based on a neural network model.The method comprises the steps that during water conservancy scientific research project purchase management, suppliers with the largest influence factors are classified, and different scientific research project scoring methods are adopted according to the supplier types; the pertinence and efficiency of scientific research project management are improved; when the number of nodes of the hidden layer of the BP neural network model is determined, establishing a relational expression between the number of training sets of the BP neural network model, the number of indexes of the training sets and the learning rate of the BP neural network model and the number of nodes of the hidden layer of the BP neural network model; in this way, the influence of different numbers of training sets, the complexity of the training sets and the training efficiency of the model learning rate during model training is considered, the training accuracy of the BP neural network model can be effectively guaranteed, and the establishment efficiency of the BP neural network model is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of project management, and particularly to a water conservancy scientific research procurement management method and system based on a neural network model. Background Art

[0002] Scientific research procurement management is a very important part of research institutes, and also an important content of the management of research institutes. Its management situation directly affects the development of research institutes. Whether the procurement management method is reasonable and scientific is closely related to the development of research institutes, and it is a prerequisite for the orderly development of research institutes; in the context of big data, information and data are growing explosively. Facing these massive and fragmented information and data, how to obtain valuable and systematic information for oneself and the industry is a major problem faced by research institutes. Empowering scientific research procurement management with big data technology is an effective way to innovate digital technology to improve the efficiency of scientific research procurement management, enhance the benefits of scientific research procurement management, and can effectively strengthen risk control.

[0003] In the prior art, there are solutions to improve the standardization of procurement and reduce procurement risks through procurement management. For example, Chinese invention patent (CN105243579A) discloses a medical device procurement management system, including a declaration unit for declaring medical device projects to be purchased, and voting on the review of medical device projects to form a procurement plan; an expert database management unit for determining multiple review experts; a supplier management unit for determining multiple suppliers; a procurement review unit for review experts to review the products of suppliers to determine the optimal supplier and the corresponding products; an audit unit for auditing multiple suppliers and their corresponding products, as well as the optimal supplier and the corresponding products, and putting forward audit opinions; a procurement risk management unit for querying, statistics and correlation analysis of multiple suppliers and their corresponding products, as well as the optimal supplier and the corresponding products. However, the above solution uses review experts to review products, resulting in poor pertinence of the review of the solution and inability to meet the procurement review requirements of complex water conservancy scientific research projects. Summary of the Invention

[0004] In view of this, the present invention proposes a water conservancy scientific research procurement management method and system based on a neural network model, which is used to improve the pertinence and efficiency of water conservancy scientific research project procurement management.

[0005] To achieve the above object, a water conservancy scientific research procurement management method based on a neural network model is provided. The method includes: S1: Obtain the supplier information of the water conservancy scientific research procurement project to be managed; S2: Classify the suppliers according to the supplier information to obtain a classification result; Among them, the supplier grading is divided into first - level suppliers and second - level suppliers; S3: Select a scoring model for the procurement plan of the water conservancy scientific research project according to the supplier grading result, and output the scoring result; The specific method of selecting a scoring model for the procurement plan of the water conservancy scientific research project according to the supplier grading result is as follows: If the supplier grading result is a first - level supplier, the expert scoring method is used to score the procurement plan of the water conservancy scientific research project; if the supplier grading result is a second - level supplier, the deep - learning algorithm is used to score the procurement plan of the water conservancy scientific research project; The specific method of using the deep - learning algorithm to score the procurement plan of the water conservancy scientific research project is as follows: Sa: Establish a deep - learning algorithm model; The deep - learning algorithm model is a BP neural network model. Obtain the number m1 of training sets of the BP neural network model, the number m2 of indicators for each training set, and the magnitude L of the learning rate; determine the number m of hidden - layer nodes according to the number m1 of training sets, the number m2 of indicators for each training set, and the magnitude L of the learning rate. The formula for determining the number m of hidden - layer nodes of the deep - learning model is:

[0006] In the formula, a and b are coefficients. In this embodiment, a is 0.01, b is 2.5, and L is 0.001; Sb: Obtain the training set of the BP neural network model; Sc: Train the BP neural network model; Sd: Input the indicators of the procurement plan of the water conservancy scientific research project into the trained BP neural network model to obtain the scoring result; S4: Manage the procurement plan of the water conservancy scientific research project according to the scoring result.

[0007] Preferably, the BP neural network consists of an input layer, a hidden layer, and an output layer.

[0008] Preferably, the number of nodes in the input layer is 4.

[0009] Preferably, in Sc, it is determined whether the BP neural network model is trained through a loss function; the loss function is the mean - square error loss function.

[0010] Preferably, in Sd, the indicators of the procurement plan of the water conservancy scientific research project are the supplier risk of the procurement project, the plan risk, the technical risk of the procurement project, and the process risk of the procurement project.

[0011] Preferably, in S4, if the scoring result is greater than the preset threshold, then the water conservancy project procurement plan is adopted; otherwise, the water conservancy project procurement plan is set as not passed.

[0012] Preferably, the preset threshold is 0.8.

[0013] Preferably, S2 is specifically as follows: S2.1: Obtain the features for supplier grading; S2.2: Create a supplier classification model based on the k-mean algorithm; S2.3: Grade the suppliers according to the supplier information to obtain a grading result.

[0014] Preferably, S2.2.1: Randomly select k points as the clustering centers; S2.2.2: Calculate the distances between the supplier samples and the clustering centers; S2.2.3: Allocate all supplier samples to the nearest supplier classes according to the distances to form 2 supplier classes; S2.2.4: Re-determine the 2 clustering centers; S2.2.5: Repeat S2.2.4 until the clustering centers no longer change; S2.2.6: Output the clustering result, thereby realizing the creation of the supplier classification model.

[0015] According to another aspect of the present invention, there is provided a water conservancy scientific research procurement management system based on a neural network model. The system adopts the above-mentioned water conservancy scientific research procurement management method based on a neural network model. The system includes: A supplier information collection module, which is used to obtain the supplier information of the water conservancy scientific research procurement project to be managed; A supplier grading module, which is used to grade the suppliers according to the supplier information to obtain a supplier grading result; A procurement project scoring module, which is used to select a water conservancy scientific research project procurement plan scoring model according to the supplier grading result and output a scoring result; A procurement project management module, which is used to manage the water conservancy scientific research project procurement plan according to the scoring result.

[0016] The advantages and beneficial effects of the present invention are as follows: When the present invention manages the procurement of water conservancy scientific research projects, it first classifies the suppliers with the greatest influence factors, and adopts different scientific research project scoring methods according to the supplier categories, improving the pertinence and efficiency of scientific research project management; When determining the number of nodes in the hidden layer of a BP neural network model, a relational expression is established between the number of training sets of the BP neural network model, the number of metrics of the training sets, the learning rate of the BP neural network model, and the number of nodes in the hidden layer of the BP neural network model. This takes into account the influence of different numbers of training sets and the complexity of the training sets during model training, as well as the training efficiency of the model learning rate, and can effectively ensure the training accuracy of the BP neural network model and improve the establishment efficiency of the BP neural network model. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the present invention or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a flowchart of a water conservancy scientific research procurement management method based on a neural network model provided by an embodiment of the present invention; Figure 2 It is a flowchart of scoring the procurement plan of the water conservancy scientific research project by using a deep learning algorithm provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0020] As shown in the Figure 1 drawings, a water conservancy scientific research procurement management method based on a neural network model, the method includes the following steps: S1: Obtain the supplier information of the water conservancy scientific research procurement project to be managed; Among them, the supplier information of the water conservancy scientific research project to be managed includes supplier qualifications, supplier credit, etc.; In this embodiment, the supplier can be a single supplier; at the same time, for some complex water conservancy scientific research procurement projects, the supplier can be multiple suppliers, and the above supplier information is the average value of each supplier's information; S2: Classify the suppliers according to the supplier information to obtain a classification result; Among them, the supplier grading is divided into first-level suppliers and second-level suppliers; the first-level suppliers are those with relatively high comprehensive scores, and the second-level suppliers are those with relatively low comprehensive scores; In this embodiment, the k-mean algorithm is used to grade the suppliers; Specifically, the S2 is specifically as follows: S2.1: Obtain the features for supplier grading; In this step, the features are the information of suppliers collected by big data methods, including: supply quality, return rate, delivery days, service level, supplier qualification, and supplier historical credit; Among them, the supply quality is used to reflect the supplier's product quality control ability. The higher the characteristic value of the supply quality, the stronger the supplier's product quality control ability; the return rate is the ratio of the number of returns to the quantity of goods supplied by the supplier. Among them, the higher the characteristic value of the return rate, the higher the product quality. Among them, the characteristic value of the return rate is not the same concept as the return rate; the delivery days are the number of days required from the signing of the contract by the supplier to the shipment. Among them, the higher the characteristic value of the delivery days, the faster the supplier supplies the product; the service level is the timeliness and satisfaction of the supplier's after-sales business answers. Among them, the higher the characteristic value of the service level, the higher the supplier's service level; the supplier qualification is whether the supplier meets quality certifications such as ISO9001. If so, the supplier qualification is assigned 100 points, otherwise it is assigned 0 points; the supplier historical credit is the production scale, development prospect, and whether there are any bad credit records of the supplier, etc.; the higher the characteristic value of the supplier historical credit, the stronger the supplier's performance ability; S2.2: Create a supplier classification model based on the k-mean algorithm; Among them, creating a supplier classification model based on the k-mean algorithm is specifically as follows: S2.2.1: Randomly select k points as cluster centers; Among them, in this embodiment, the suppliers are divided into first-level suppliers and second-level suppliers, and the k is 2; S2.2.2: Calculate the distance between the sample in the supplier sample and the cluster center; S2.2.3: According to the distance, allocate all supplier samples to the nearest supplier class to form 2 supplier classes; S2.2.4: Re-determine 2 cluster centers; Among them, in this step, the average value of the characteristic values of each feature of each supplier sample in each supplier class is calculated to obtain 2 new cluster centers; S2.2.5: Repeat the S2.2.4; until the cluster centers no longer change; S2.2.6: Output the clustering results, thus realizing the creation of the supplier classification model; S2.3: Classify the suppliers according to the said supplier information to obtain the classification result; In this step, input the obtained supplier information into the supplier classification model to obtain the supplier classification result; Among them, the supplier classification result is that the supplier belongs to a first-class supplier or a second-class supplier; S3: Select the scoring model for the procurement plan of water conservancy scientific research projects according to the supplier classification result and output the scoring result; Among them, the selection of the scoring model for the procurement plan of water conservancy scientific research projects according to the supplier classification result is specifically as follows: if the supplier classification result is a first-class supplier, the expert scoring method is used to score the procurement plan of the water conservancy scientific research project; if the supplier classification result is a second-class supplier, the deep learning algorithm is used to score the procurement plan of the water conservancy scientific research project; The specific method of using the expert scoring method to score the supplier is as follows: randomly select n experts from the expert library, send the procurement plan of the water conservancy scientific research project to the scoring experts, and then fill in the scoring form for the procurement plan of the water conservancy scientific research project. Among them, the scoring form is divided into four dimensions: supplier risk of the procurement project, plan risk, technical risk of the procurement project, and process risk of the procurement project. Take the average value of the scoring values of the n experts to obtain the score of the procurement plan of the water conservancy scientific research project; As attached Figure 2 As shown, the specific method of using the deep learning algorithm to score the procurement plan of the water conservancy scientific research project is as follows: Sa: Establish a deep learning algorithm model; Among them, the deep learning algorithm model is a BP neural network model. The BP neural network consists of an input layer, a hidden layer, and an output layer. Among them, the number of nodes in the input and output layers and the hidden layer of the BP neural network is closely related to the specific problem to be processed. In this embodiment, it is closely related to the procurement plan of the water conservancy scientific research project; In this embodiment, the input indicators of the input layer include: supplier risk of the procurement project, plan risk, technical risk of the procurement project, and process risk of the procurement project; therefore, the number of nodes in the input layer is 4; The number of nodes in the hidden layer should be set according to the requirements of the BP neural network model and the complexity of the problem. If the number of hidden layer nodes is small, the problem-solving information obtained by the BP neural network will be less; if the number of hidden nodes is too large, it will affect the learning speed of the BP neural network model and cause prediction errors; In the prior art, when selecting the number of hidden layer nodes, it is generally determined by manual experiments. That is, first, a relatively small number of nodes is set to establish a model, and then the learning effect of the model, that is, the learning accuracy rate, is detected. After that, the number of nodes is gradually increased until there is no significant fluctuation in the accuracy rate. The above method requires repeated experiments, resulting in low efficiency. This embodiment proposes a method for determining the number of hidden layer nodes of a BP neural network model to improve the establishment efficiency of the BP neural network model. Specifically, the method for determining the number of hidden layer nodes m of the BP neural network model is as follows: Obtain the number m1 of training sets of the BP neural network model, the number m2 of indicators of each training set, and the magnitude L of the learning rate; determine the number m of hidden layer nodes according to the number m1 of training sets, the number m2 of indicators of each training set, and the magnitude L of the learning rate. Among them, the determination formula for the number m of hidden layer nodes is:

[0021] In the formula, a and b are coefficients. In this embodiment, a is 0.01, b is 2.5, and L is 0.001. In this embodiment, when determining the number of hidden layer nodes of the BP neural network model, a relational expression is established between the number of training sets of the BP neural network model, the number of indicators of the training sets, and the learning rate of the BP neural network model and the number of hidden layer nodes of the BP neural network model. In this way, it takes into account the influence of different numbers of training sets and the complexity of the training sets during model training and the training efficiency of the magnitude of the model learning rate, can effectively ensure the training accuracy of the BP neural network model, and improves the establishment efficiency of the BP neural network model. Sb: Obtain the training set of the BP neural network model; Among them, the training set of the BP neural network model is obtained by collecting and sorting out the big data of historical water conservancy scientific research procurement projects. Scientific research project procurement management is an indispensable part of the operation of scientific research institutes. Nowadays, the requirements for the accuracy and efficiency of procurement decisions are getting higher and higher. As a powerful decision-making support tool, big data analysis can help scientific research institutes extract valuable information from huge data, optimize the decision-making process, and improve the efficiency and accuracy of scientific research project procurement management.

[0022] Sc: Train the BP neural network model; Determine whether the BP neural network model is trained completed through a loss function; in this embodiment, the loss function is the mean square error loss function. Sd: Input the indicators of the water conservancy scientific research project procurement plan into the trained BP neural network model to obtain a scoring result. Among them, the indicators of the procurement plan for the water conservancy scientific research project are the risks of suppliers of procurement projects, plan risks, technical risks of procurement projects, and process risks of procurement projects; S4: Manage the procurement plan for the water conservancy scientific research project according to the scoring result; Among them, if the scoring result is greater than the preset threshold, the procurement plan for the water conservancy project is passed; otherwise, the procurement plan for the water conservancy project is set as not passed.

[0023] In this embodiment, the preset threshold is 0.8.

[0024] When the present invention manages the procurement of water conservancy scientific research projects, it first classifies the suppliers with the greatest impact factors, and adopts different scoring methods for scientific research projects according to the supplier categories, which improves the pertinence and efficiency of scientific research project management.

[0025] Embodiment 2. This embodiment includes a water conservancy scientific research procurement management system based on a neural network model. The system adopts a water conservancy scientific research procurement management method based on a neural network model in Embodiment 1. The system includes: A supplier information collection module, which is used to obtain the supplier information of the water conservancy scientific research procurement project to be managed; A supplier grading module, which is used to grade the suppliers according to the supplier information to obtain a grading result; A procurement project scoring module, which is used to select a scoring model for the procurement plan of the water conservancy scientific research project according to the supplier grading result and output a scoring result; The specific method of selecting a scoring model for the procurement plan of the water conservancy scientific research project according to the supplier grading result is as follows: if the supplier grading result is a first-level supplier, the expert scoring method is used to score the procurement plan of the water conservancy scientific research project; if the supplier grading result is a second-level supplier, the deep learning algorithm is used to score the procurement plan of the water conservancy scientific research project; A procurement project management module, which is used to manage the procurement plan of the water conservancy scientific research project according to the scoring result.

[0026] Embodiment 3. This embodiment includes a computer-readable storage medium, on which a data processing program is stored, and the data processing program is executed by a processor to implement a water conservancy scientific research procurement management method based on a neural network model in Embodiment 1.

[0027] Those skilled in the art should understand that the embodiments herein can be provided as a method, an apparatus (device), or a computer program product. Therefore, the present disclosure can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. including but not limited to RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those of ordinary skill in the art, communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

[0028] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments herein. It should be understood that each flow and / or block in the flowchart illustrations and / or block diagrams, and combinations of flows and / or blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general purpose computer, a special purpose computer, an embedded processor, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device produce means for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0029] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including instruction means for implementing the steps of the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or combinations of blocks.

[0030] The above-described embodiments and / or implementation manners are merely used to illustrate the preferred embodiments and / or implementation manners for realizing the technology of the present invention, and do not impose any formal restrictions on the implementation manners of the technology of the present invention. Any person skilled in the art, without departing from the scope of the technical means disclosed in the content of the present invention, may make some changes or modifications to other equivalent embodiments, but should still be regarded as the same technology or embodiment as the present invention in essence. Note that the above is only the preferred embodiment of the present invention and the technical principles applied. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described here, and various obvious changes, re-adjustments and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments only. Without departing from the concept of the present invention, more other equivalent embodiments can be included, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A water conservancy scientific research procurement management method based on a neural network model, characterized in that: The method comprises the following steps: S1: Obtain supplier information of water conservancy scientific research procurement projects to be managed; S2: Classify suppliers according to the supplier information to obtain supplier classification results; wherein the supplier classification results are divided into first-level suppliers and second-level suppliers; S3: Select a scoring model for the water conservancy scientific research project procurement plan according to the supplier grading result, and output the scoring result; the scoring model for the water conservancy scientific research project procurement plan selected according to the supplier grading result is specifically: if the supplier grading result is a first-level supplier, the water conservancy scientific research project procurement plan is scored using an expert scoring method; if the supplier grading result is a second-level supplier, the water conservancy scientific research project procurement plan is scored using a deep learning algorithm; The use of deep learning algorithm to score the water conservancy scientific research project procurement plan is specifically as follows: Sa: Build deep learning algorithm model; The deep learning algorithm model is a BP neural network model, and the number of training sets m1, the number of indicators m2 of each training set, and the size L of the learning rate of the BP neural network model are obtained; the number m of hidden layer nodes is determined according to the number m1 of training sets, the number m2 of indicators of each training set, and the size L of the learning rate; The formula for determining the number m of hidden layer nodes of the deep learning algorithm model is: In the formula, a and b are coefficients, a is 0.01, b is 2.5, and L is 0.001; Sb: Obtain the training set of the BP neural network model; Sc: training the BP neural network model; Sd: Input the indicators of the water conservancy scientific research project procurement plan into the trained BP neural network model to obtain the scoring result; S4: Manage the procurement plan of the water conservancy scientific research project according to the scoring results.

2. According to the water conservancy scientific research procurement management method based on the neural network model of claim 1, it is characterized in that: The BP neural network includes three parts: input layer, hidden layer and output layer.

3. According to claim 2, a water conservancy scientific research procurement management method based on a neural network model is characterized in that: The number of nodes in the input layer is 4.

4. The water conservancy scientific research procurement management method based on a neural network model according to claim 1 is characterized in that: In the Sc, whether the BP neural network model is trained is determined by a loss function; the loss function is a mean square error loss function.

5. The water conservancy scientific research procurement management method based on a neural network model according to claim 1 is characterized in that: In the Sd, the indicators of the water conservancy scientific research project procurement plan are procurement project supplier risk, planning risk, procurement project technical risk and procurement project process risk.

6. The water conservancy scientific research procurement management method based on a neural network model according to claim 1 is characterized in that: In S4, if the scoring result is greater than a preset threshold, the water conservancy project procurement plan is approved; otherwise, the water conservancy project procurement plan is set to be rejected.

7. The water conservancy scientific research procurement management method based on a neural network model according to claim 6 is characterized in that: The preset threshold is 0.

8.

8. The water conservancy scientific research procurement management method based on a neural network model according to claim 1 is characterized in that: The S2 is specifically: S2.1: Obtain characteristics for supplier grading; S2.2: Create a supplier classification model based on the k-mean algorithm; S2.3: Classify the suppliers according to the supplier information to obtain a classification result.

9. The water conservancy scientific research procurement management method based on a neural network model according to claim 8 is characterized in that: The S2.2 is specifically: S2.2.1: Randomly select k points as cluster centers; S2.2.2: Calculate the distance between samples in the supplier sample and the cluster center; S2.2.3: Allocate all supplier samples to the closest supplier class based on the distance, forming 2 supplier classes; S2.2.4: Re-determine the two cluster centers; S2.2.5: Repeat S2.2.4 until the cluster center no longer changes; S2.2.6: Output clustering results to create a supplier classification model.

10. A water conservancy scientific research procurement management system based on a neural network model, characterized in that: The system adopts a water conservancy scientific research procurement management method based on a neural network model as described in any one of claims 1 to 9, and the system comprises: Supplier information collection module, used to obtain supplier information of water conservancy scientific research procurement projects to be managed; A supplier grading module, used to grade the suppliers according to the supplier information to obtain supplier grading results; A procurement project scoring module is used to select a water conservancy scientific research project procurement plan scoring model according to the supplier classification results and output the scoring results; The procurement project management module is used to manage the procurement plan of the water conservancy scientific research project according to the scoring results.

Images