Engineering material price management system based on supply chain

Abstract

The invention discloses an engineering material price management system based on a supply chain, relates to the technical field of engineering material management, and solves the technical problems that a purchase management platform in the prior art is difficult to comprehensively evaluate materials needing to be purchased before purchasing the materials, the purchase quality is affected, and the efficiency of a purchase scheme is low. Inputting the material comprehensive data into a price prediction model to obtain a corresponding material prediction price; generating a current supplier material score and a predicted supplier material score according to the current material price and the predicted material price of the supplier data; screening the supplier material scores to obtain supplier material score labels; generating a supplier comprehensive score according to the supplier material score label and the supplier material score; the purchasing scheme is generated according to the comprehensive score of the supplier, thereby improving the efficiency of the purchasing scheme.

Description

Technical Field

[0001] The present application belongs to the field of engineering material management, and specifically is an engineering material price management system based on a supply chain. Background Art

[0002] With the acceleration of global economic integration, the complexity of engineering projects is increasing, which puts higher requirements on the price management of engineering materials. At present, there are some systems for engineering material price management on the market. These systems mainly provide price references for purchasers by collecting market price information and establishing price databases. However, most of these systems remain at the level of data collection and display, lacking in-depth price forecasting and procurement optimization functions.

[0003] The prior art (invention patent application with publication number CN115496511A) discloses a raw material supply chain procurement information review method and a procurement management platform, including: receiving raw material procurement information for each raw material type input by a procurement management account, the raw material procurement information including multiple supplier names and sales prices corresponding to the supplier names; sending the target raw material procurement information to multiple production management accounts; receiving target feedback information for the target supplier name input by all production management accounts, the target feedback information including feedback evaluation scores corresponding to multiple procurement feedback items; calculating and generating a comprehensive evaluation score based on multiple feedback evaluation scores in the target feedback information; generating a procurement evaluation form based on multiple comprehensive evaluation scores corresponding to the target raw material procurement information, and sending the procurement evaluation form to the company management account.

[0004] The above scheme generates a procurement plan based on the raw material prices and corresponding supplier information provided by the current supply chain, purchases the supplied goods based on the current supply chain status information, and then provides feedback and scoring to each supplier after procurement; the supplier's product quality, product output and supplier reputation are not considered before procurement, so when the product does not meet the procurement standards, it is necessary to return the product, etc.; this leads to a waste of time, affects the progress of the construction period, and thus leads to a low efficiency of the procurement plan; therefore, the engineering material price management system of the supply chain still needs to be further improved. Summary of the invention

[0005] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a supply chain-based engineering material price management system to solve the technical problem that the existing procurement management platform is difficult to conduct a comprehensive evaluation of the materials to be purchased before purchasing the materials, which affects the procurement quality and leads to low efficiency of the procurement plan.

[0006] To achieve the above-mentioned purpose, the first aspect of the present application provides a supply chain-based engineering material price management system, comprising: a data acquisition module, a data analysis module, a human-computer interaction module and an engineering material management cloud platform;

[0007] The data acquisition module: obtains the procurement requirements through the human-computer interaction module connected thereto, queries the supply chain data and market data of the corresponding engineering materials in the engineering material cloud management platform according to each engineering material in each procurement item in the procurement requirements; and performs data preprocessing on the supply chain data and market data;

[0008] The data analysis module: obtains supplier data and market data from the supply chain data after data preprocessing corresponding to each engineering material; integrates the material historical price, raw material historical price and recording time in the supplier data and the market data with the same recording time into comprehensive material data and inputs it into the price prediction model to obtain the corresponding material prediction price;

[0009] and, generating a current supplier material score and a predicted supplier material score corresponding to each engineering material according to the current material price and the predicted material price corresponding to each supplier data and the supplier score in the supplier data;

[0010] Filter the supplier material scores corresponding to each engineering material to obtain the supplier material score labels;

[0011] Generate a comprehensive score for each supplier in the corresponding procurement project based on the supplier material score label and supplier material score; generate a procurement plan based on the supplier comprehensive score;

[0012] The human-computer interaction module is used to obtain procurement requirements and display procurement plans.

[0013] This application obtains procurement requirements, queries the supply chain data and market data of the corresponding engineering materials in the engineering materials cloud management platform according to each engineering material in each procurement project in the procurement requirements, and performs preprocessing operations; integrates the material historical prices, raw material historical prices and recording times in the supplier data after data preprocessing, and the market data with the same recording time into material comprehensive data, and inputs them into the price prediction model to obtain the corresponding material prediction prices; generates the current supplier material score and predicted supplier material score corresponding to each engineering material according to the current material price and material prediction price corresponding to each supplier data and the supplier score in the supplier data; screens the supplier material score corresponding to each engineering material to obtain the supplier material score label; generates the comprehensive score of each supplier in the corresponding procurement project according to the supplier material score label and the supplier material score; generates a procurement plan according to the supplier comprehensive score, conducts a comprehensive evaluation of each supplier and each engineering material before purchasing the engineering materials, and then makes a reasonable plan for the procurement suppliers; thereby improving the efficiency of the procurement plan.

[0014] Furthermore, the data preprocessing includes: data cleaning and data normalization.

[0015] Furthermore, the price prediction model is constructed by using an LSTM model, including:

[0016] Obtain the historical prices of several materials, the historical prices of raw materials and their corresponding recording time and the market data corresponding to the recording time for each supplier ID in the engineering materials required for procurement;

[0017] Performing data preprocessing on several material historical prices, raw material historical prices and their corresponding recording times and market data corresponding to the recording times to obtain preprocessed several material historical prices, raw material historical prices and their corresponding recording times and market data corresponding to the recording times;

[0018] Integrate the material historical prices, raw material historical prices and market data with the same recording time and the same engineering material and supplier ID into a time feature vector; the market data includes market prices and market supply and demand ratios;

[0019] Using the principal component analysis method to reduce the characteristics of a number of time characteristic vectors, a reduced time characteristic vector is obtained; the reduced characteristic vector is a characteristic vector obtained by the principal component analysis method after the cumulative contribution rate reaches 90%;

[0020] Integrate each time feature vector into several prediction sequences according to the order of recording time;

[0021] Divide several prediction sequences into training set, test set and validation set in chronological order;

[0022] Select the LSTM model as the base model;

[0023] Train the basic model with the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain the pre-trained model;

[0024] By verifying the pre-trained model on the test set, we finally obtained a price prediction model whose input is the comprehensive data of engineering materials and whose output is the corresponding predicted price of materials.

[0025] Further, generating the current supplier material score and predicted supplier material score corresponding to each engineering material according to the current material price and predicted material price corresponding to each supplier data and the supplier score in the supplier data includes:

[0026] Obtain the current material price DJi, material forecast price CYJi, current material planned production volume DJCi, forecast material planned production volume YJCi, total transportation cost YFi, supplier rating GPi and material demand volume CX of the corresponding engineering materials in the procurement requirements for several supplier IDs in the corresponding engineering materials;

[0027] Determine whether the material demand is greater than the current material planned production:

[0028] Yes, calculate the current supplier material score DGCPi through the formula, DGCPi=0;

[0029] No, calculate the current supplier material score DGCPi through the formula;

[0030] DGCPi=α1×exp(-(DYJi+YFi) / BF)+α2×1 / (1+exp(-(DJCi-CX) / DC))+α3×GPi / max(GPi);

[0031] Determine whether the material demand is greater than the predicted material planned production volume:

[0032] Yes, the supplier material score YGCPi is predicted by the formula, YGCPi=0;

[0033] No, the predicted supplier material score YGCPi is calculated by formula;

[0034] YGCPi=α1×exp(-(CYJi+YFi) / BF)+α2×1 / (1+exp(-(YJCi-CX) / DC))+α3×GPi / max(GPi);

[0035] Among them, i represents the number of the supplier of the corresponding engineering material; α1, α2 and α3 are weight coefficients, α1, α2 and α3∈(0,1); BF is the standard cost, and DC is the unit output.

[0036] Furthermore, the supplier material scores corresponding to each engineering material are screened to obtain supplier material score labels, including:

[0037] Obtain the current supplier material score DGCPi and predicted supplier material score YGCPi corresponding to the engineering material;

[0038] Determine in turn whether each current supplier material score DGCPi is greater than the corresponding engineering material score threshold; if yes, mark the supplier material score label corresponding to the current supplier material score as qualified; if no, mark the supplier material score label corresponding to the current supplier material score as unqualified;

[0039] Determine in turn whether each predicted supplier material score YGCPi is greater than the corresponding predicted engineering material score threshold; if yes, mark the supplier material score label corresponding to the current supplier material score as qualified; if no, mark the supplier material score label corresponding to the current supplier material score as unqualified;

[0040] The predicted engineering material scoring threshold is set according to the engineering material scoring threshold.

[0041] Furthermore, the predicted engineering material scoring threshold is set according to the engineering material scoring threshold, including:

[0042] Obtain the scoring threshold GCYi of each engineering material; obtain the current material price DJi, the material forecast price CYJi, the current material planned production volume DJCi, the forecast material planned production volume YJCi and the forecast time interval T;

[0043] The predicted engineering material score threshold YCYi of the corresponding predicted supplier material score of the supplier numbered i is calculated by the formula YCYi=(1+k×exp(-T×(|CYJi-DJi| / DJ+|YJCi-DJCi| / DC)))×GCYi; where DJ is the unit price, DC is the unit output, and k is the adjustment coefficient.

[0044] Furthermore, generating a comprehensive score of each supplier in the corresponding procurement project based on the supplier material score label and the supplier material score includes:

[0045] Get the current supplier material score DGCPi and predicted supplier material score YGCPi corresponding to each supplier; as well as the corresponding supplier material score label;

[0046] Generate several engineering material scores based on each current supplier material score DGCPi, predicted supplier material score YGCPi, and its corresponding supplier material score label;

[0047] The scores of each engineering material are summed up to obtain the supplier's comprehensive score.

[0048] Furthermore, according to each current supplier material score DGCPi, predicted supplier material score YGCPi, and its corresponding supplier material score label, several engineering material scores are generated, including:

[0049] Determine whether the supplier material score label corresponding to the current supplier material score DGCPi and the predicted supplier material score YGCPi of the same engineering material are both qualified;

[0050] If yes, the current excess evaluation value DCPi is calculated by the formula DCPi=(DGCPi-GCYi) / GCYi; the predicted excess evaluation value YCPi is calculated by YCPi=(YGCPi-YCYi) / YCYi; the current supplier material score DGCPi or the predicted supplier material score YGCPi corresponding to the larger value of the current excess evaluation and the predicted excess evaluation is used as the corresponding engineering material score of the supplier;

[0051] No, when the supplier material score label corresponding to the current supplier material score DGCPi of the same engineering material and the predicted supplier material score YGCPi are both qualified, the current supplier material score DGCPi or the predicted supplier material score YGCPi with the qualified supplier material score label will be used as the corresponding engineering material score of the supplier; otherwise, the engineering material score will be set to "0".

[0052] Furthermore, generating a procurement plan based on the comprehensive rating of suppliers includes the following steps:

[0053] Step 1: Obtain the comprehensive supplier score of each supplier ID, as well as the engineering materials and each engineering material in the procurement requirements;

[0054] Step 2: Sort the supplier IDs from largest to smallest according to the supplier's comprehensive score;

[0055] Step 3: Select the supplier ID with the highest comprehensive score as the first supplier, and purchase all engineering materials from the first supplier whose engineering material score is not "0";

[0056] Step 4: Get the remaining unpurchased engineering materials, set the engineering material scores of the purchased engineering materials to "0", and recalculate the comprehensive supplier scores of each supplier; use the supplier ID with the highest comprehensive supplier score as the remaining supplier, and purchase all engineering materials whose engineering material scores are not "0";

[0057] Step 5: Repeat step 4 until all engineering materials in the procurement requirements have been purchased;

[0058] Step 6: All engineering materials and their corresponding supplier IDs, purchase quantities, purchase prices, and corresponding purchase opportunities are combined into a purchase plan; the purchase opportunities include the current time and the predicted time.

[0059] This application selects the supplier ID with the highest comprehensive supplier score as the first supplier, and purchases all engineering materials whose engineering material scores are not "0" from the first supplier; obtains the remaining unpurchased engineering materials, sets the engineering material scores corresponding to the purchased engineering materials to "0", and recalculates the comprehensive supplier scores of each supplier; uses the supplier ID with the highest comprehensive supplier score as the remaining supplier, and purchases all engineering materials whose engineering material scores are not "0", repeats the operation until all engineering materials in the procurement demand are purchased, and gives priority to engineering materials produced by the supplier ID with the highest comprehensive supplier score, thereby reducing the number of selected supplier IDs while ensuring the quality of the engineering materials, and saving the number of transportation times of the engineering materials.

[0060] Compared with the prior art, the beneficial effects of this application are:

[0061] 1. This application obtains procurement requirements, queries the supply chain data and market data of the corresponding engineering materials in the engineering materials cloud management platform according to the various engineering materials in each procurement project in the procurement requirements, and performs preprocessing operations; integrates the material historical prices, raw material historical prices and recording times in the supplier data after data preprocessing, and the market data with the same recording time into comprehensive material data and inputs them into the price prediction model to obtain the corresponding material forecast prices; generates the current supplier material score and forecast supplier material score corresponding to each engineering material according to the current material price and material forecast price corresponding to each supplier data and the supplier score in the supplier data; screens the supplier material score corresponding to each engineering material to obtain the supplier material score label; generates the comprehensive score of each supplier in the corresponding procurement project according to the supplier material score label and the supplier material score; generates a procurement plan according to the comprehensive supplier score, conducts a comprehensive evaluation of each supplier and each engineering material before purchasing the engineering materials, and then makes a reasonable plan for the procurement suppliers; thereby improving the efficiency of the procurement plan.

[0062] 2. This application selects the supplier ID with the highest comprehensive supplier score as the first supplier, and purchases all engineering materials whose engineering material scores are not "0" from the first supplier; obtains the remaining unpurchased engineering materials, sets the engineering material scores corresponding to the purchased engineering materials to "0", and recalculates the comprehensive supplier scores of each supplier; selects the supplier ID with the highest comprehensive supplier score as the remaining supplier, and purchases all engineering materials whose engineering material scores are not "0", repeats the operation until all engineering materials in the procurement demand are purchased, and gives priority to engineering materials produced by the supplier ID with the highest comprehensive supplier score, thereby reducing the number of selected supplier IDs while ensuring the quality of engineering materials, and saving the number of transportation times of engineering materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0064] Figure 1 This is a schematic diagram of the principle of a supply chain-based engineering material price management system for this application;

[0065] Figure 2 A method flow chart of a supply chain-based engineering material price management system for this application;

[0066] Figure 3 A procurement solution flow chart for generating procurement requirements for this application. DETAILED DESCRIPTION

[0067] The technical solution of the present application will be described clearly and completely below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0068] See also Figure 1-Figure 2 The first aspect of the present application provides a supply chain-based engineering material price management system, including:

[0069] Data acquisition module, data analysis module, human-computer interaction module and engineering material management cloud platform;

[0070] Data collection module: obtains procurement requirements through the human-computer interaction module connected to it, and queries the supply chain data and market data of the corresponding engineering materials in the engineering materials cloud management platform according to the engineering materials in each procurement project in the procurement requirements; procurement requirements include a number of required engineering materials and the corresponding procurement quantity, etc.; supply chain data includes a number of supplier data, supplier data includes supplier ID, supplier rating, material price and material planned production volume, as well as a number of material historical prices of each supplier, raw material historical prices and their corresponding recording time, market data includes recording time, market price and market supply and demand ratio, etc.; supplier rating is the average of the comprehensive ratings of the supplier's credit and product quality by each customer corresponding to the supplier; material planned production volume is the planned output of the supplier to produce the corresponding engineering materials within the specified time; pre-process the supply chain data and market data;

[0071] Data analysis module: obtain supplier data from the supply chain data after data preprocessing corresponding to each engineering material; integrate the material historical price, raw material historical price and recording time in the supplier data, and market data with the same recording time into comprehensive material data and input it into the price prediction model to obtain the corresponding material prediction price;

[0072] And, generating a current supplier material score and a predicted supplier material score corresponding to each engineering material according to the current material price and the predicted material price corresponding to each supplier data and the supplier score in the supplier data; the current supplier material score represents the supplier's comprehensive evaluation of the corresponding engineering material at the current moment, and the predicted supplier material score represents the supplier's comprehensive evaluation of the corresponding engineering material at the prediction moment;

[0073] The supplier material scores corresponding to each engineering material are screened to obtain supplier material score labels; the supplier material score labels include qualified labels and unqualified labels;

[0074] Generate a comprehensive score for each supplier in the corresponding procurement project based on the supplier material score label and supplier material score; Generate a procurement plan based on the supplier comprehensive score; The procurement plan includes supplier ID, engineering materials, total purchase amount, purchase price and procurement timing, etc.;

[0075] Human-computer interaction module: used to obtain procurement requirements and display procurement plans.

[0076] This embodiment obtains procurement requirements, queries the supply chain data and market data of the corresponding engineering materials in the engineering materials cloud management platform according to each engineering material in each procurement project in the procurement requirements, and performs data preprocessing; integrates the material historical prices, raw material historical prices and recording time in the supplier data after data preprocessing, and the market data with the same recording time into material comprehensive data, and inputs them into the price prediction model to obtain the corresponding material prediction price; generates the current supplier material score and predicted supplier material score corresponding to each engineering material according to the current material price and material prediction price corresponding to each supplier data and the supplier score in the supplier data; screens the supplier material score corresponding to each engineering material to obtain the supplier material score label; generates the comprehensive score of each supplier in the corresponding procurement project according to the supplier material score label and the supplier material score; generates a procurement plan according to the supplier comprehensive score, obtains the engineering material prediction price through multivariate data, conducts a comprehensive evaluation of each supplier and each engineering material before procurement of engineering materials, and then makes reasonable planning for procurement suppliers; thereby improving the efficiency of the procurement plan.

[0077] Data preprocessing includes: data cleaning and data normalization;

[0078] Data cleaning: directly remove missing values ​​from the acquired supplier data and market data to obtain normal data;

[0079] Data normalization: Normalize the normal data using the maximum and minimum normalization method to obtain preprocessed detection data.

[0080] The price prediction model is constructed through the LSTM model, including:

[0081] Obtain the historical prices of several materials, the historical prices of raw materials and their corresponding recording time and the market data corresponding to the recording time for each supplier ID in the engineering materials required for procurement;

[0082] Performing data preprocessing on several material historical prices, raw material historical prices and their corresponding recording times and market data corresponding to the recording times to obtain preprocessed several material historical prices, raw material historical prices and their corresponding recording times and market data corresponding to the recording times;

[0083] The historical prices of materials with the same recording time and the same engineering material and supplier ID, the historical prices of raw materials and the market data with the same recording time are integrated into a time feature vector; the market data includes market prices, market supply and demand ratios, etc.;

[0084] Using the principal component analysis method to reduce the characteristics of a number of time characteristic vectors, a reduced time characteristic vector is obtained; the reduced characteristic vector is a characteristic vector obtained by the principal component analysis method after the cumulative contribution rate reaches 90%;

[0085] Integrate each time feature vector into several prediction sequences according to the order of recording time;

[0086] Divide several prediction sequences into training set, test set and validation set in chronological order; the ratio between training set, test set and validation set is 7:2:1;

[0087] Select the LSTM model as the base model;

[0088] Train the basic model with the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain the pre-trained model;

[0089] By verifying the pre-trained model on the test set, we finally obtained a price prediction model whose input is the comprehensive data of engineering materials and whose output is the corresponding predicted price of materials.

[0090] This embodiment predicts the price of the corresponding engineering materials of each supplier within a certain period of time in the future by training the price prediction model, and the time corresponding to the predicted price is before the procurement deadline; the model is trained with a large amount of supplier data and market data to make the prediction result more in line with the actual situation and more accurate.

[0091] The current material score and predicted material score of each engineering material are generated based on the current material price and predicted material price corresponding to each supplier data and the supplier score in the supplier data, including:

[0092] Obtain the current material price DJi, material forecast price CYJi, current material planned production volume DJCi, forecast material planned production volume YJCi, total transportation cost YFi, supplier rating GPi and material demand volume CX of the corresponding engineering materials in the procurement requirements for several supplier IDs in the corresponding engineering materials;

[0093] Determine whether the material demand is greater than the current material planned production:

[0094] Yes, calculate the current supplier material score DGCPi through the formula, DGCPi=0;

[0095] No, calculate the current supplier material score DGCPi through the formula; DGCPi = α1×exp(-(DYJi+YFi) / BF)+α2×1 / (1+exp(-(DJCi-CX) / DC))+α3×GPi / max(GPi); when the material demand exceeds the planned material production, it means that the supplier cannot fully provide the corresponding engineering materials at the corresponding time, and therefore cannot meet the procurement demand; set the current supplier's supplier material score for the engineering material to 0; as the current material price increases, the current material planned production decreases, the total transportation cost increases and the supplier score decreases, the company's intention to choose products with higher current material prices will decrease. Similarly, the lower the supplier score, the company's intention to purchase products in the supplier ID will also decrease, which will lead to a decrease in the current supplier material score; conversely, the company's intention to choose products with higher current material prices will increase, that is, the current supplier material score will increase.

[0096] Determine whether the material demand is greater than the predicted material planned production volume:

[0097] Yes, the supplier material score YGCPi is predicted by the formula, YGCPi=0;

[0098] No, the predicted supplier material score YGCPi is calculated by the formula; YGCPi = α1×exp(-(CYJi+YFi) / BF)+α2×1 / (1+exp(-(YJCi-CX) / DC))+α3×GPi / max(GPi); as the predicted price of the predicted material increases, the planned production volume of the predicted material decreases, the total transportation cost increases and the supplier score decreases, the company's intention to choose products with higher predicted material prices will decrease. Similarly, the lower the supplier score, the company's purchase intention for the products in the supplier ID will also decrease, which will lead to a decrease in the predicted supplier material score; conversely, the company's intention to choose products with higher predicted material prices will increase, that is, the predicted supplier material score will increase.

[0099] Among them, i represents the number of the supplier of the corresponding engineering material; α1, α2 and α3 are weight coefficients, which are set according to experience, and α1, α2 and α3; BF is the standard cost, which is set according to experience; max(GPi) is the maximum value of the scores of several suppliers corresponding to the engineering material; DC is the unit output, which is set according to experience.

[0100] This embodiment comprehensively scores the engineering materials corresponding to the corresponding suppliers by comprehensively considering the planned output, material prices, transportation costs and supplier scores of the corresponding suppliers of engineering materials, so as to facilitate subsequent screening of suppliers.

[0101] Filter the supplier material scores corresponding to each engineering material to obtain the supplier material score labels, including:

[0102] Obtain the current supplier material score DGCPi and predicted supplier material score YGCPi corresponding to the engineering material;

[0103] Determine in turn whether each current supplier material score DGCPi is greater than the corresponding engineering material score threshold; the engineering material score threshold is set based on experience; if yes, the supplier material score label corresponding to the current supplier material score is marked as qualified; if no, the supplier material score label corresponding to the current supplier material score is marked as unqualified;

[0104] Determine in turn whether each predicted supplier material score YGCPi is greater than the corresponding predicted engineering material score threshold; if yes, mark the supplier material score label corresponding to the current supplier material score as qualified; if no, mark the supplier material score label corresponding to the current supplier material score as unqualified;

[0105] The predicted engineering material score threshold is set according to the engineering material score threshold.

[0106] In this embodiment, when the supplier material scoring label is qualified, it means that the engineering materials corresponding to the supplier at this moment meet the procurement requirements, that is, the procurement quantity and price are met, as well as the material quality, etc.; when the supplier material scoring label is unqualified, it means that the engineering materials corresponding to the supplier at this moment do not meet the procurement requirements; and they will not be considered subsequently.

[0107] The predicted engineering material scoring threshold is set according to the engineering material scoring threshold, including:

[0108] Get the scoring threshold GCYi of each engineering material, and set it according to the expert's inspection; get the current material price DJi, the material forecast price CYJi, the current material planned production volume DJCi, the forecast material planned production volume YJCi and the forecast time interval T; the forecast time interval T is the time interval between the time corresponding to the material forecast price and the current time;

[0109] The predicted engineering material score threshold value YCYi of the predicted supplier material score corresponding to the supplier numbered i is calculated by the formula;

[0110] YCYi=(1+k×exp(-T×(|CYJi-DJi| / DJ+|YJCi-DJCi| / DC)))×GCYi; where DJ is the unit price and DC is the unit output; k is the adjustment coefficient, which is set according to experience; as the forecast time interval increases, the difference between the forecast material price and the current material price becomes larger, or the difference between the forecast material planned production volume and the current material planned production volume becomes larger, it means that the production trend of the corresponding engineering materials for the supplier numbered i will become more unstable in the future, and the standard of the corresponding forecast engineering material scoring threshold will also increase, which will lead to an increase in the forecast engineering material scoring threshold.

[0111] This embodiment uses the gap between the predicted value and the current value, that is, the stability of the production trend of the corresponding engineering material; adjusts the predicted engineering material scoring threshold corresponding to the predicted value; and sets the predicted engineering material scoring threshold for each predicted value in a targeted manner, so that the screening results of the predicted value are more accurate.

[0112] Generate a comprehensive score for each supplier in the corresponding procurement project based on the supplier material score label and supplier material score, including:

[0113] Get the current supplier material score DGCPi and predicted supplier material score YGCPi corresponding to each supplier; as well as the corresponding supplier material score label;

[0114] Generate several engineering material scores based on each current supplier material score DGCPi, predicted supplier material score YGCPi, and its corresponding supplier material score label;

[0115] The scores of each engineering material are summed up to obtain the supplier's comprehensive score.

[0116] According to each current supplier material score DGCPi, predicted supplier material score YGCPi, and its corresponding supplier material score label, several engineering material scores are generated, including:

[0117] Determine whether the supplier material score label corresponding to the current supplier material score DGCPi and the predicted supplier material score YGCPi of the same engineering material are both qualified;

[0118] If yes, the current excess evaluation value DCPi is calculated by the formula DCPi=(DGCPi-GCYi) / GCYi; the formula is to calculate the excess ratio of the current supplier material score compared to the predicted supplier material score. The more the current supplier material score exceeds the predicted supplier material score, the greater the excess ratio, and the current excess evaluation value will increase accordingly;

[0119] The predicted excess evaluation value YCPi is calculated by YCPi=(YGCPi-YCYi) / YCYi; the formula is to calculate the excess ratio of the predicted supplier material score compared to the current supplier material score. The more the predicted supplier material score exceeds the current supplier material score, the greater the excess ratio, and the predicted excess evaluation value will increase accordingly;

[0120] The current supplier material score DGCPi or the predicted supplier material score YGCPi corresponding to the larger value of the current excess evaluation and the predicted excess evaluation is used as the corresponding engineering material score of the supplier;

[0121] No, when the supplier material score label corresponding to the current supplier material score DGCPi of the same engineering material and the predicted supplier material score YGCPi are both qualified, the current supplier material score DGCPi or the predicted supplier material score YGCPi with the qualified supplier material score label will be used as the corresponding engineering material score of the supplier; otherwise, the engineering material score will be set to "0".

[0122] In this embodiment, the current supplier material score of engineering material 1 of supplier A is 0.9, and its corresponding supplier material score label is qualified; the predicted supplier material score of engineering material 1 of supplier A is 0.6, and its corresponding supplier material score label is unqualified; since the supplier material score label corresponding to the current supplier material score of engineering material 1 is qualified, and the supplier material score label corresponding to the predicted supplier material score of engineering material 1 is unqualified, the current supplier material score of 0.9 with a supplier material score label of qualified is used as the engineering material score of engineering material 1 in supplier A.

[0123] See also Figure 3 , generating a procurement plan based on the comprehensive rating of suppliers, including the following steps:

[0124] Step 1: Obtain the comprehensive supplier score of each supplier ID, as well as the engineering materials and each engineering material in the procurement requirements;

[0125] Step 2: Sort the supplier IDs from largest to smallest according to the supplier's comprehensive score;

[0126] Step 3: Select the supplier ID with the highest comprehensive supplier score as the first supplier, and purchase all engineering materials with a non-0 engineering material score from the first supplier; that is, purchase engineering materials corresponding to the current supplier material score or the predicted supplier material score at the corresponding time; the first supplier is the supplier ID that is preferred when selecting each engineering material in the procurement demand;

[0127] Step 4: Get the remaining unpurchased engineering materials, set the engineering material scores corresponding to the purchased engineering materials to "0", and recalculate the comprehensive supplier scores of each supplier; use the supplier ID with the highest comprehensive supplier score as the remaining supplier, and purchase all engineering materials whose engineering material scores are not "0"; the remaining supplier refers to the supplier ID selected to purchase other engineering materials after the first supplier;

[0128] Step 5: Repeat step 4 until all engineering materials in the procurement requirements have been purchased;

[0129] Step 6: Combine all engineering materials and their corresponding supplier ID, total purchase amount, purchase price and corresponding purchase time into a purchase plan; the purchase time is the current moment or the predicted moment, the predicted moment is the future moment with a predicted time interval from the current moment; the purchase price is the current material price or the predicted material price.

[0130] This embodiment selects the supplier ID with the highest comprehensive supplier score as the first supplier, and purchases all engineering materials whose engineering material scores are not "0" from the first supplier; obtains the remaining unpurchased engineering materials, sets the engineering material scores corresponding to the purchased engineering materials to "0", and recalculates the comprehensive supplier scores of each supplier; uses the supplier ID with the highest comprehensive supplier score as the remaining supplier, and purchases all engineering materials whose engineering material scores are not "0", repeats the operation until all engineering materials in the procurement demand are purchased, and gives priority to engineering materials produced by the supplier ID with the highest comprehensive supplier score, thereby reducing the number of selected supplier IDs while ensuring the quality of the engineering materials, and saving the number of transportation times of the engineering materials.

[0131] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0132] The working principle of this application is as follows: by obtaining procurement requirements, querying the supply chain data and market data of the corresponding engineering materials in the engineering materials cloud management platform according to each engineering material in each procurement project in the procurement requirements, and performing data preprocessing; integrating the material historical price, raw material historical price and recording time in the supplier data after data preprocessing, and the market data with the same recording time into material comprehensive data and inputting them into the price prediction model to obtain the corresponding material predicted price; generating the current supplier material score and predicted supplier material score corresponding to each engineering material according to the current material price and material predicted price corresponding to each supplier data and the supplier score in the supplier data; screening the supplier material score corresponding to each engineering material to obtain the supplier material score label; generating the comprehensive score of each supplier in the corresponding procurement project according to the supplier material score label and the supplier material score; generating a procurement plan according to the supplier comprehensive score, obtaining the predicted price of engineering materials through multivariate data, taking the predicted results into consideration in the procurement plan, improving the foresight of the product procurement plan, and thus improving the efficiency of the procurement plan, avoiding the problem that the existing technology only considers the status information of the current supply chain, lacks a comprehensive evaluation of each supplier and each engineering material before the procurement of engineering materials, and then reasonably plans the procurement suppliers; thus leading to the problem of low efficiency of the procurement plan.

[0133] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A supply chain-based engineering material price management system, characterized in that: include: Data acquisition module, data analysis module, human-computer interaction module and engineering material management cloud platform; The data acquisition module: obtains the procurement requirements through the human-computer interaction module connected thereto, queries the supply chain data and market data of the corresponding engineering materials in the engineering material cloud management platform according to each engineering material in each procurement item in the procurement requirements; and performs data preprocessing on the supply chain data and market data; The data analysis module: obtains supplier data and market data from the supply chain data after data preprocessing corresponding to each engineering material; integrates the material historical price, raw material historical price and recording time in the supplier data and the market data with the same recording time into comprehensive material data and inputs it into the price prediction model to obtain the corresponding material prediction price; and, generating a current supplier material score and a predicted supplier material score corresponding to each engineering material according to the current material price and the predicted material price corresponding to each supplier data and the supplier score in the supplier data; Filter the supplier material scores corresponding to each engineering material to obtain the supplier material score labels; Generate a comprehensive score for each supplier in the corresponding procurement project based on the supplier material score label and supplier material score; Generate a procurement plan based on the comprehensive ratings of suppliers; The human-computer interaction module is used to obtain procurement requirements and display procurement plans.

2. The engineering material price management system based on supply chain according to claim 1 is characterized in that: The data preprocessing includes: data cleaning and data normalization.

3. The engineering material price management system based on supply chain according to claim 1 is characterized in that: The price prediction model is constructed through the LSTM model, including: Obtain the historical prices of several materials, the historical prices of raw materials and their corresponding recording time and the market data corresponding to the recording time for each supplier ID in the engineering materials required for procurement; Performing data preprocessing on several material historical prices, raw material historical prices and their corresponding recording times and market data corresponding to the recording times to obtain preprocessed several material historical prices, raw material historical prices and their corresponding recording times and market data corresponding to the recording times; Integrate the material historical prices, raw material historical prices and market data with the same recording time and the same engineering material and supplier ID into a time feature vector; the market data includes market prices and market supply and demand ratios; Using the principal component analysis method to reduce the characteristics of a number of time characteristic vectors, a reduced time characteristic vector is obtained; the reduced characteristic vector is a characteristic vector obtained by the principal component analysis method after the cumulative contribution rate reaches 90%; Integrate each time feature vector into several prediction sequences according to the order of recording time; Divide several prediction sequences into training set, test set and validation set in chronological order; Select the LSTM model as the base model; Train the basic model with the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally obtained a price prediction model whose input is the comprehensive data of engineering materials and whose output is the corresponding predicted price of materials.

4. The engineering material price management system based on supply chain according to claim 1 is characterized in that: The method of generating a current supplier material score and a predicted supplier material score corresponding to each engineering material according to the current material price and the predicted material price corresponding to each supplier data and the supplier score in the supplier data includes: Obtain the current material price DJi, material forecast price CYJi, current material planned production volume DJCi, forecast material planned production volume YJCi, total transportation cost YFi, supplier rating GPi and material demand volume CX of the corresponding engineering materials in the procurement requirements for several supplier IDs in the corresponding engineering materials; Determine whether the material demand is greater than the current material planned production: Yes, calculate the current supplier material score DGCPi through the formula, DGCPi=0; No, calculate the current supplier material score DGCPi through the formula; DGCPi=α1×exp(-(DYJi+YFi) / BF)+α2×1 / (1+exp(-(DJCi-CX) / DC))+α3×GPi / max(GPi); Determine whether the material demand is greater than the predicted material planned production volume: Yes, the supplier material score YGCPi is predicted by the formula, YGCPi=0; No, the predicted supplier material score YGCPi is calculated by formula; YGCPi=α1×exp(-(CYJi+YFi) / BF)+α2×1 / (1+exp(-(YJCi-CX) / DC))+α3×GPi / max(GPi); Among them, i represents the number of the supplier of the corresponding engineering material; α1, α2 and α3 are weight coefficients, α1, α2 and α3∈(0,1); BF is the standard cost, and DC is the unit output.

5. The engineering material price management system based on supply chain according to claim 1 is characterized in that: The supplier material scores corresponding to each engineering material are screened to obtain supplier material score labels, including: Obtain the current supplier material score DGCPi and predicted supplier material score YGCPi corresponding to the engineering material; Determine in turn whether each current supplier material score DGCPi is greater than the corresponding engineering material score threshold; if yes, mark the supplier material score label corresponding to the current supplier material score as qualified; if no, mark the supplier material score label corresponding to the current supplier material score as unqualified; Determine in turn whether each predicted supplier material score YGCPi is greater than the corresponding predicted engineering material score threshold; if yes, mark the supplier material score label corresponding to the current supplier material score as qualified; if no, mark the supplier material score label corresponding to the current supplier material score as unqualified; The predicted engineering material scoring threshold is set according to the engineering material scoring threshold.

6. The engineering material price management system based on supply chain according to claim 5 is characterized in that: The predicted engineering material scoring threshold is set according to the engineering material scoring threshold, including: Obtain the scoring threshold GCYi of each engineering material; obtain the current material price DJi, the material forecast price CYJi, the current material planned production volume DJCi, the forecast material planned production volume YJCi and the forecast time interval T; The predicted engineering material score threshold YCYi of the corresponding predicted supplier material score of the supplier numbered i is calculated by the formula YCYi=(1+k×exp(-T×(|CYJi-DJi| / DJ+|YJCi-DJCi| / DC)))×GCYi; where DJ is the unit price, DC is the unit output, and k is the adjustment coefficient.

7. The engineering material price management system based on supply chain according to claim 1 is characterized in that: The comprehensive score of each supplier in the corresponding procurement project is generated based on the supplier material score label and the supplier material score, including: Get the current supplier material score DGCPi and predicted supplier material score YGCPi corresponding to each supplier; as well as the corresponding supplier material score label; Generate several engineering material scores based on each current supplier material score DGCPi, predicted supplier material score YGCPi, and its corresponding supplier material score label; The scores of each engineering material are summed up to obtain the supplier's comprehensive score.

8. The engineering material price management system based on supply chain according to claim 7 is characterized in that: Based on the current supplier material score DGCPi and the predicted supplier material score YGCPi; And its corresponding supplier material score tag generates several engineering material scores, including: Determine whether the supplier material score label corresponding to the current supplier material score DGCPi and the predicted supplier material score YGCPi of the same engineering material are both qualified; If yes, the current excess evaluation value DCPi is calculated by the formula DCPi=(DGCPi-GCYi) / GCYi; the predicted excess evaluation value YCPi is calculated by YCPi=(YGCPi-YCYi) / YCYi; the current supplier material score DGCPi or the predicted supplier material score YGCPi corresponding to the larger value of the current excess evaluation and the predicted excess evaluation is used as the corresponding engineering material score of the supplier; If not, when the supplier material score label corresponding to the current supplier material score DGCPi and the predicted supplier material score YGCPi of the same engineering material are both qualified, the current supplier material score DGCPi or the predicted supplier material score YGCPi with the qualified supplier material score label will be used as the corresponding engineering material score of the supplier; otherwise, the engineering material score will be set to "0".

9. The engineering material price management system based on supply chain according to claim 1 is characterized in that: Generating a procurement plan based on the comprehensive rating of suppliers includes the following steps: Step 1: Obtain the comprehensive supplier score of each supplier ID, as well as the engineering materials and each engineering material in the procurement requirements; Step 2: Sort the supplier IDs from largest to smallest according to the supplier's comprehensive score; Step 3: Select the supplier ID with the highest comprehensive score as the first supplier, and purchase all engineering materials with a score not equal to "0" from the first supplier; Step 4: Get the remaining unpurchased engineering materials, set the engineering material scores of the purchased engineering materials to "0", and recalculate the comprehensive supplier scores of each supplier; select the supplier ID with the highest comprehensive supplier score as the remaining supplier, and purchase all engineering materials whose engineering material scores are not "0"; Step 5: Repeat step 4 until all engineering materials in the procurement requirements have been purchased; Step 6: All engineering materials and their corresponding supplier IDs, purchase quantities, purchase prices, and corresponding purchase opportunities are combined into a purchase plan; the purchase opportunities include the current time and the predicted time.

Images