An automatic method and system for selecting a substitute material part for a raw material

Abstract

This invention discloses an automatic selection method and system for substitute materials of raw materials. The method includes the following steps: based on the bill of materials corresponding to each historical material selection record, automatically generating several material selection rules through machine learning, each selection rule including the selection weight corresponding to the parameter indicators of each material in the bill of materials; obtaining a verification dataset and verifying the verification pass rate of each selection rule based on the verification dataset; selecting the selection rule with the highest verification pass rate as the target selection rule, and automatically selecting substitute materials of the raw materials based on the target selection rule and the bill of materials received in real time. This invention can reduce the manual intervention time in the material selection process and improve the material selection efficiency.

Description

Technical Field

[0001] This invention relates to the field of chip manufacturing technology, and in particular to a method and system for automatically selecting substitute components for raw materials. Background Technology

[0002] The selection of alternative components for raw materials usually requires manual intervention. The component engineers rely on their experience in component selection during chip manufacturing and their understanding of the materials to select suitable components for chip manufacturing. However, the component selection process lacks objective data measurement standards. When engineers lack experience in component selection or fail to fully evaluate the currently available materials, errors in component selection can easily occur, affecting the chip manufacturing process.

[0003] Therefore, there is a need for an automatic selection method for alternative raw materials. This method should generate selection strategies for alternative raw materials based on historical material selection records using machine learning, thereby reducing manual intervention time during the material selection process and improving material selection efficiency. Summary of the Invention

[0004] To address the technical problems of error-prone and inefficient manual selection of substitutes for raw materials, this invention provides an automatic selection method and system for substitutes for raw materials. The specific technical solution is as follows:

[0005] This invention provides a method for automatically selecting substitute materials for raw materials, comprising the following steps:

[0006] Based on the bill of materials corresponding to each historical material selection record, several material selection rules are automatically generated through machine learning. Each material selection rule includes the selection weight corresponding to the parameter indicators of each material in the bill of materials.

[0007] Obtain a verification dataset and verify the pass rate of each of the material selection rules based on the verification dataset;

[0008] The material selection rule with the highest verification pass rate is selected as the target material selection rule, and the substitute material for the original material is automatically selected according to the target material selection rule and the material list received in real time.

[0009] The automatic selection method for substitutes of raw materials provided by this invention combines historical material selection records to automatically generate material selection rules, and automatically selects substitutes of raw materials based on the generated material selection rules and the material list received in real time. This reduces the error risk caused by human intervention in the selection process of substitutes of raw materials and improves the efficiency of selecting substitutes of raw materials.

[0010] In some embodiments, after automatically selecting substitute materials for the original materials according to the target material selection rules and the bill of materials received in real time, the method further includes:

[0011] The real-time received bill of materials and the results of automatically selecting alternative materials for the original materials are stored as material selection data in the historical material selection record;

[0012] After the preset number of times the automatic selection of substitute materials for raw materials is performed, the historical material selection record is sent to manual review.

[0013] When the manual review result is approved, the substitute material for the original material is automatically selected according to the current target material selection rules and the material list received in real time;

[0014] When the manual review result is not approved, the current target material selection rule is optimized according to the historical material selection record, and the substitute material for the original material is automatically selected according to the optimized material selection rule and the material list received in real time.

[0015] The automatic selection method for substitute materials of raw materials provided by the present invention optimizes the selection rules in real time based on the results of automatic selection of substitute materials of raw materials, and introduces manual review data to further improve the accuracy of automatic selection of substitute materials of raw materials.

[0016] In some implementations, the automatic generation of several material selection rules based on the bill of materials corresponding to each historical material selection record using machine learning specifically includes:

[0017] Based on the bill of materials, project type, customer name and part specifications corresponding to each historical material selection record, several primary alternative material selection rules are automatically generated through machine learning.

[0018] Based on the bill of materials corresponding to each historical material selection record, several auxiliary material selection rules are automatically generated through machine learning.

[0019] The automatic selection of substitute materials for the original materials based on the target material selection rules and the real-time received bill of materials specifically includes:

[0020] The primary alternative material is automatically selected based on the target material selection rules and the real-time received bill of materials, project type, customer name, and part specifications.

[0021] The auxiliary substitute material is automatically selected based on the target material selection rules and the bill of materials received in real time.

[0022] The automatic selection method for substitute materials of raw materials provided by the present invention generates primary substitute material selection rules and secondary substitute material selection rules respectively during the automatic selection process of substitute materials of raw materials, and performs automatic selection of primary substitute materials and secondary substitute materials according to the corresponding selection rules, thereby improving the efficiency of the primary substitute material selection process and the accuracy of the secondary substitute material selection process.

[0023] In some implementations, the method of automatically generating several material selection rules based on the bill of materials corresponding to each historical material selection record using machine learning further includes:

[0024] Based on the bill of materials and material selection reference type parameters corresponding to each historical material selection record, several material selection rules are automatically generated through machine learning. Each of the material selection reference type parameters corresponds to several preset material selection reference bases.

[0025] The automatic selection method for substitute materials of raw materials provided by the present invention introduces a material selection reference type parameter in the automatic selection process of substitute materials of raw materials, thereby further improving the accuracy of automatic material selection of selection rules.

[0026] In some implementations, before selecting the material selection rule with the highest verification pass rate as the target material selection rule, the method further includes:

[0027] The bill of materials corresponding to each historical material selection record is divided into a training dataset and a validation dataset.

[0028] Based on the training dataset, several material selection rules are automatically generated using machine learning.

[0029] The verification pass rate corresponding to each of the material selection rules is verified based on the verification dataset.

[0030] In some implementations, before obtaining the verification dataset and verifying the verification pass rate corresponding to each of the material selection rules based on the verification dataset, the method further includes:

[0031] Upon receiving updated bill of materials data, the training dataset and the validation dataset are optimized based on the updated bill of materials.

[0032] In some embodiments, according to another aspect of the invention, the invention also provides an automatic selection system for substitute raw materials, comprising:

[0033] The generation module is used to automatically generate several material selection rules based on the bill of materials corresponding to each historical material selection record through machine learning. Each material selection rule includes the selection weight corresponding to the parameter indicators of each material in the bill of materials.

[0034] A verification module, connected to the generation module, is used to verify the verification pass rate and material selection results corresponding to each of the material selection rules based on the verification dataset.

[0035] The material selection module, connected to the verification module, is used to select the material selection rule with the highest verification pass rate as the target material selection rule, and automatically select substitute materials for the original material according to the target material selection rule and the material list received in real time.

[0036] In some embodiments, the automatic selection system for substitute raw materials provided by the present invention further includes:

[0037] The update module is connected to the material selection module and the generation module respectively, and is used to store the material list received in real time and the result of automatically selecting the substitute material of the original material as material selection data in the historical material selection record;

[0038] The sending module, connected to the updating module, is used to send the historical material selection record to manual review after a preset number of automatic selection of substitute materials for the original material.

[0039] The optimization module is connected to both the material selection module and the sending module. When the manual review result is "passed", it automatically selects a substitute material for the original material according to the current target material selection rule and the material list received in real time. When the manual review result is "failed", it optimizes the current target material selection rule according to the historical material selection record, and automatically selects a substitute material for the original material according to the optimized material selection rule and the material list received in real time.

[0040] In some implementations, the generation module specifically includes:

[0041] The first generation unit is used to automatically generate several main material selection rules based on the bill of materials corresponding to each historical material selection record through machine learning.

[0042] The second generation unit is used to automatically generate several auxiliary material selection rules based on the bill of materials, project type, customer name and part specifications corresponding to each historical material selection record, using machine learning.

[0043] The material selection module specifically includes:

[0044] The first generation unit is used to automatically generate several main alternative material selection rules based on the bill of materials, project type, customer name and part specifications corresponding to each historical material selection record, using machine learning.

[0045] The second generation unit is used to automatically generate several auxiliary material selection rules based on the bill of materials corresponding to each historical material selection record through machine learning.

[0046] The material selection module specifically includes:

[0047] The first material selection unit is used to automatically select the main alternative material according to the target material selection rules and the material list, project type, customer name and part specifications received in real time;

[0048] The second material selection unit is used to automatically select auxiliary substitutes according to the target material selection rules and the material list received in real time.

[0049] In some implementations, the generation module is further configured to automatically generate several material selection rules based on the bill of materials and material selection reference type parameters corresponding to each historical material selection record through machine learning, wherein each material selection reference type parameter corresponds to several preset material selection reference bases.

[0050] This invention provides at least one of the following technical effects:

[0051] (1) Automatically generate material selection rules by combining historical material selection records, and automatically select substitute materials for the original materials based on the generated material selection rules and the material list received in real time, thereby reducing the error risk caused by manual intervention in the selection of substitute materials for the original materials and improving the selection efficiency of substitute materials for the original materials.

[0052] (2) Optimize the material selection rules in real time based on the results of automatically selecting substitute materials for raw materials, and introduce manual review data to further improve the accuracy of automatic selection of substitute materials for raw materials;

[0053] (3) During the automatic selection of substitute materials for raw materials, the main substitute material selection rules and the auxiliary substitute material selection rules are generated respectively, and the main substitute material and the auxiliary substitute material are automatically selected according to the corresponding selection rules, thereby improving the efficiency of the main substitute material selection process and the accuracy of the auxiliary substitute material selection process.

[0054] (4) Introduce material selection reference type parameters in the automatic selection process of raw material substitutes to further improve the accuracy of automatic material selection rules. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a flowchart of an automatic selection method for substitute materials of raw materials according to the present invention;

[0057] Figure 2 This is another flowchart of the automatic selection method for alternative raw materials according to the present invention;

[0058] Figure 3 This is a flowchart illustrating the automatic selection method for substitute materials of raw materials according to the present invention, which selects the main substitute material and the auxiliary substitute material respectively.

[0059] Figure 4 This is another flowchart illustrating the automatic generation of several material selection rules in the automatic selection method for alternative raw materials of the present invention.

[0060] Figure 5 This is an example diagram of an automatic selection system for substitute raw materials according to the present invention;

[0061] Figure 6 This is another example diagram of an automatic selection system for substitute raw materials according to the present invention;

[0062] Figure 7 This is an example diagram of the generation module in the automatic selection system for substitute raw materials of the present invention.

[0063] The diagram is labeled as follows: Generation Module-10, First Generation Unit-11, Second Generation Unit-12, Verification Module-20, Material Selection Module-30, First Material Selection Unit-31, Second Material Selection Unit-32, Update Module-40, Sending Module-50, and Optimization Module-60. Detailed Implementation

[0064] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0065] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or sets.

[0066] To keep the drawings concise, only the parts relevant to the invention are shown schematically in each figure, and they do not represent the actual structure of the product. Furthermore, for ease of understanding, in some figures, components with the same structure or function are shown only schematically, or only one is labeled. In this document, "a" not only means "only one," but can also mean "more than one."

[0067] It should also be further understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0068] Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0069] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the specific implementation methods of the present invention will be described below with reference to the accompanying drawings. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings and other implementation methods can be obtained based on these drawings without any creative effort.

[0070] One embodiment of the present invention, such as Figure 1 As shown, the present invention provides a method for automatically selecting substitute materials for raw materials, including the following steps:

[0071] Based on the bill of materials corresponding to each historical material selection record, the S100 automatically generates several material selection rules through machine learning.

[0072] Specifically, each material selection rule includes the selection weight corresponding to the parameter indicators of each material in the bill of materials. Since the parameter indicators of each material in the bill of materials (BOM) itself represent the corresponding material characteristics, several material selection rules can be generated directly based on the parameter indicators of each material in the BOM using supervised or unsupervised machine learning methods.

[0073] For example, for a capacitor assembly, the selection rules are composed of selection weights corresponding to parameters such as length, width, height, capacitance, and rated voltage.

[0074] Furthermore, since the output of the material selection rule cannot be predicted, a Model-Free RL combined with reinforcement learning is used. When supervised learning fails to identify the problem, the model is allowed to perform machine learning again through Temporal Difference feedback, thereby iterating to obtain a more ideal material selection rule.

[0075] S200 obtains the verification dataset and verifies the pass rate of each material selection rule based on the verification dataset.

[0076] Specifically, the validation dataset can be pre-set manually and adjusted at any time, or a number of data points can be selected from historical material selection records as the training dataset, and the remaining data as the validation dataset, to generate and validate the material selection rules.

[0077] The S300 selects the material selection rule with the highest verification pass rate as the target material selection rule, and automatically selects the substitute material for the original material based on the target material selection rule and the bill of materials received in real time.

[0078] Specifically, the material selection rules can be automatically adjusted or manually adjusted according to different chip requirements. In different material selection rules, the weight of the same indicator parameter is different, and the corresponding verification pass rate will also be different. For example, taking a capacitor component used in the production process of certain preset chip models as an example, considering the indicator parameters such as the length, width, height, capacitance, and rated power of the capacitor component, the material selection rules can set the indicator parameters to be considered and set their values ​​to 1 within a preset range, while setting the values ​​of indicator parameters not considered to 0. These material selection rules are verified by a verification dataset.

[0079] For example, as shown in Table 1 below, Rules 1 to 4 are different material selection rules. Under the corresponding material selection rules, different amounts of replacement materials are obtained, and the pass rates are also different.

[0080]

[0081] Table 1

[0082] The automatic selection method for substitute materials of raw materials provided in this embodiment combines historical material selection records to automatically generate material selection rules, and automatically selects substitute materials of raw materials based on the generated material selection rules and the material list received in real time. This reduces the error risk caused by manual intervention in the selection process of substitute materials of raw materials and improves the selection efficiency of substitute materials of raw materials.

[0083] In one embodiment, such as Figure 2 As shown, after step S300 automatically selects the substitute material for the original material based on the material selection rule with the highest verification pass rate and the real-time received bill of materials, it also includes:

[0084] The S400 stores the real-time received bill of materials and the results of automatically selecting alternative materials as material selection data in the historical material selection record.

[0085] For example, the bill of materials received in real time and the results of automatically selecting alternative materials for the original materials in Rules 1 and Rules 2 in Table 1 can be stored as material selection data in the historical material selection record.

[0086] After the S500 performs the preset number of automatic selections of substitute materials for raw materials, it sends the historical material selection records to manual review.

[0087] For example, in Table 1, after executing the material selection rules Rule 1 to Rule 4, 1333, 1455, 2133 and 39423 alternative materials were found respectively, and these material selection records need to be sent to manual review.

[0088] During the manual review process, it was found that Rule 1 and Rule 2 had the highest verification pass rate of 98%, followed by Rule 3 at 92%, while Rule 4 had an extremely low pass rate of 5%. Rule 4 identified a large number of potential substitutes, but its accuracy was too low after comparison with the verification dataset, rendering it unsuitable. Rule 1 and Rule 2 achieved similar results, but Rule 1 was relatively more rigorous, thus identifying fewer substitutes. After review, Rule 2 might be evaluated as superior to Rule 1, and Rule 2 might be chosen as the target material selection rule.

[0089] When the S610 receives a manual review result indicating that the material has passed, it automatically selects a substitute material for the original material based on the current target material selection rules and the real-time received bill of materials.

[0090] For example, when the target material selection rule Rule2 receives a manual review result indicating that it has passed, it can automatically select alternative materials for the original material based on Rule2 and the bill of materials received in real time.

[0091] When the S620 receives a manual review result indicating failure, it optimizes the current target material selection rules based on historical material selection records, and automatically selects alternative materials for the original material based on the optimized material selection rules and the real-time received bill of materials.

[0092] For example, Rule2 can be used as the target material selection rule. If the manual review result is not passed, Rule2 can be optimized based on historical material selection records.

[0093] The automatic selection method for substitute materials of raw materials provided in this embodiment optimizes the selection rules in real time based on the results of automatically selecting substitute materials of raw materials, and introduces manual review data to further improve the accuracy of automatic selection of substitute materials of raw materials.

[0094] In one embodiment, such as Figure 3 As shown, step S100 automatically generates several material selection rules based on the bill of materials corresponding to each historical material selection record using machine learning, specifically including:

[0095] Based on the bill of materials, project type, customer name, and part specifications corresponding to each historical material selection record, S110 automatically generates several primary alternative material selection rules through machine learning.

[0096] Based on the bill of materials corresponding to each historical material selection record, S120 automatically generates several auxiliary material selection rules through machine learning.

[0097] For example, in the process of making primary alternative material selection rules, some companies and types of chip products need to use high-specification resistors and capacitors to avoid noise generation. Therefore, it is necessary to merge the bill of materials, project type, customer name and part specifications to generate data tags, and iteratively generate auxiliary material selection rules based on the merged data tags, so that the primary alternative material selection rules generated by machine learning iteration are more accurate.

[0098] S200 obtains the verification dataset and verifies the pass rate of each material selection rule based on the verification dataset.

[0099] Specifically, step S200 also includes the following steps: dividing the bill of materials corresponding to each historical material selection record into a training dataset and a validation dataset; automatically generating several material selection rules based on the training dataset using machine learning; and validating the validation pass rate of each material selection rule based on the validation dataset.

[0100] Step S300 automatically selects substitute materials for the original materials based on the material selection rules with the highest verification pass rate and the real-time received bill of materials, specifically including:

[0101] The S310 automatically selects the primary alternative material based on the target material selection rules and the real-time received bill of materials, project type, customer name, and part specifications.

[0102] The S320 automatically selects auxiliary substitutes based on the target material selection rules and the real-time received bill of materials.

[0103] The automatic selection method for substitute materials of raw materials provided in this embodiment generates primary substitute material selection rules and secondary substitute material selection rules respectively during the automatic selection process of substitute materials of raw materials, and automatically selects primary substitute materials and secondary substitute materials according to the corresponding selection rules, thereby improving the efficiency of the primary substitute material selection process and the accuracy of the secondary substitute material selection process.

[0104] In one embodiment, such as Figure 4 As shown, step S100 automatically generates several material selection rules based on the bill of materials corresponding to each historical material selection record using machine learning. Specifically, it also includes:

[0105] S130 automatically generates several material selection rules based on the bill of materials and material selection reference type parameters corresponding to each historical material selection record through machine learning.

[0106] Specifically, each material selection reference type parameter corresponds to several preset material selection reference bases.

[0107] The automatic selection method for substitute materials of raw materials provided in this embodiment introduces a material selection reference type parameter in the automatic selection process of substitute materials of raw materials, thereby further improving the accuracy of automatic material selection of selection rules.

[0108] In one embodiment, such as Figure 5 As shown, according to another aspect of the present invention, the present invention also provides an automatic selection system for substitute raw materials, including a generation module 10, a verification module 20 and a material selection module 30.

[0109] The generation module 10 is used to automatically generate several material selection rules based on the bill of materials corresponding to each historical material selection record through machine learning.

[0110] Specifically, each material selection rule includes the selection weight corresponding to the parameter indicators of each material in the bill of materials. Since the parameter indicators of each material in the bill of materials (BOM) itself represent the corresponding material characteristics, several material selection rules can be generated directly based on the parameter indicators of each material in the BOM using supervised or unsupervised machine learning methods.

[0111] Furthermore, since the output of the material selection rule cannot be predicted, a Model-Free RL combined with reinforcement learning is used. If supervised learning fails to identify the problem, the model is allowed to perform machine learning again through Temporal Difference feedback, thereby iterating to obtain a more ideal material selection rule.

[0112] The verification module 20 is connected to the generation module 10 and is used to verify the verification pass rate corresponding to each material selection rule based on the verification dataset.

[0113] Specifically, the validation dataset can be pre-set manually and adjusted at any time, or a number of data points can be selected from historical material selection records as the training dataset, and the remaining data as the validation dataset, to generate and validate the material selection rules.

[0114] The material selection module 30 is connected to the verification module 20 and is used to automatically select substitute materials for the original materials based on the material selection rules with the highest verification pass rate and the material list received in real time.

[0115] Specifically, the selection rules can be automatically adjusted or manually adjusted according to different needs, and the verification pass rate and selection results will be different for different selection rules.

[0116] The automatic material substitution selection system provided in this embodiment combines historical material selection records to automatically generate material selection rules, and automatically selects material substitutions for raw materials based on the generated material selection rules and the real-time received material list. This reduces the risk of errors caused by manual intervention in the material substitution selection process and improves the efficiency of material substitution selection.

[0117] In one embodiment, such as Figure 6As shown, the automatic selection system for substitute raw materials provided by the present invention also includes an update module 40, a sending module 50, and an optimization module 60.

[0118] The update module 40 is connected to the material selection module 30 and the generation module 10 respectively, and is used to store the real-time received material list and the results of automatically selecting the substitute materials of the original materials as material selection data in the historical material selection record.

[0119] The sending module 50 is connected to the updating module 40 and is used to send the historical material selection record to manual review after the preset number of times the substitute material for the original material is automatically selected.

[0120] The optimization module 60 is connected to the material selection module 30 and the sending module 50 respectively. When the manual review result is passed, it automatically selects the substitute material for the original material according to the current material selection rules and the material list received in real time. When the manual review result is failed, it optimizes the current material selection rules according to the historical material selection records, and automatically selects the substitute material for the original material according to the optimized material selection rules and the material list received in real time.

[0121] The automatic selection system for substitute materials provided in this embodiment optimizes the selection rules in real time based on the results of automatically selecting substitute materials for raw materials, and introduces manual review data to further improve the accuracy of automatic selection of substitute materials for raw materials.

[0122] In one embodiment, such as Figure 7 As shown, the generation module 10 specifically includes a first generation unit 11 and a second generation unit 12.

[0123] The first generation unit 11 is used to automatically generate several primary alternative material selection rules based on the bill of materials, project type, customer name and part specifications corresponding to each historical material selection record through machine learning. The second generation unit 12 is used to automatically generate several secondary alternative material selection rules based on the bill of materials corresponding to each historical material selection record through machine learning.

[0124] For example, in the process of formulating auxiliary material selection rules, some companies and types of chip products need to use high-specification resistors and capacitors to avoid noise generation. Therefore, it is necessary to merge the bill of materials, project type, customer name and part specifications to generate data tags, and iteratively generate the main alternative material selection rules based on the merged data tags, so that the main alternative material selection rules generated by machine learning iteration are more accurate.

[0125] The material selection module 30 specifically includes a first material selection unit 31 and a second material selection unit 32.

[0126] The first material selection unit 31 is used to automatically select the main substitute material of the original material according to the main substitute material selection rule with the highest verification pass rate and the material list, project type, customer name and part specifications received in real time. The second material selection unit 32 is used to automatically select the auxiliary substitute material of the original material according to the auxiliary substitute material selection rule with the highest verification pass rate and the material list received in real time.

[0127] The automatic selection system for substitute materials provided in this embodiment generates primary substitute material selection rules and secondary substitute material selection rules during the automatic selection process of substitute materials for raw materials. It then automatically selects primary and secondary substitute materials according to the corresponding selection rules, thereby improving the efficiency of the primary substitute material selection process and the accuracy of the secondary substitute material selection process.

[0128] In one embodiment, the generation module 10 is further configured to automatically generate several material selection rules based on the bill of materials and material selection reference type parameters corresponding to each historical material selection record through machine learning, wherein each material selection reference type parameter corresponds to several preset material selection reference bases.

[0129] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0130] Those skilled in the art will recognize that the units and steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0131] In the embodiments provided in this application, it should be understood that the disclosed method and system for automatically selecting substitute materials for raw materials can be implemented in other ways. For example, the system embodiment of the method for automatically selecting substitute materials for raw materials described above is merely illustrative. For instance, the division of modules or units is merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the communication connections shown or discussed may be through interfaces, communication connections of devices or units, or integrated circuits, and may be electrical, mechanical, or other forms.

[0132] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0133] Furthermore, the functional units in the various embodiments of this application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0134] It should be noted that the above description is only a preferred embodiment of the present invention. It should be pointed out that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for automatically selecting a substitute material piece for a raw material, characterized by, Including the following steps: The bill of materials corresponding to each historical material selection record is divided into a training dataset and a validation dataset. Based on the bill of materials corresponding to each historical material selection record in the training dataset, several primary alternative material selection rules and several secondary alternative material selection rules are automatically generated through machine learning. The primary alternative material selection rules are generated based on the bill of materials, project type, customer name, and part specifications. The secondary alternative material selection rules are generated based on the bill of materials. Each primary alternative material selection rule or secondary alternative material selection rule includes the selection weight corresponding to the parameter indicators of each material in the bill of materials. Obtain the verification dataset, and verify the verification pass rate corresponding to each of the main substitute material selection rules and each of the auxiliary substitute material selection rules based on the verification dataset; The primary and secondary substitute material selection rules with the highest verification pass rates are selected as target material selection rules. Substitute materials for the original materials are automatically selected based on these target material selection rules and the real-time received bill of materials. This includes: automatically selecting primary substitute materials based on the target material selection rules, the real-time received bill of materials, the project type, the customer name, and the part specifications; and automatically selecting secondary substitute materials based on the target material selection rules and the real-time received bill of materials. The real-time received bill of materials and the results of automatically selecting alternative materials for the original materials are stored as material selection data in the historical material selection record; After the preset number of times the automatic selection of substitute materials for raw materials is performed, the historical material selection record is sent to manual review. When the manual review result is approved, the substitute material for the original material is automatically selected according to the current target material selection rules and the material list received in real time; When the manual review result is not approved, the current target material selection rule is optimized according to the historical material selection record, and the substitute material for the original material is automatically selected according to the optimized material selection rule and the material list received in real time.

2. The method for automatically selecting substitute materials for raw materials according to claim 1, characterized in that, The method of automatically generating several material selection rules based on the bill of materials corresponding to each historical material selection record using machine learning also includes: Based on the bill of materials and material selection reference type parameters corresponding to each historical material selection record, several material selection rules are automatically generated through machine learning. Each of the material selection reference type parameters corresponds to several preset material selection reference bases.

3. The method for automatically selecting substitute materials for raw materials according to claim 1, characterized in that, Before obtaining the verification dataset and verifying the pass rate of each material selection rule based on the verification dataset, the method further includes: Upon receiving updated bill of materials data, the training dataset and the validation dataset are optimized based on the updated bill of materials.

4. An automatic selection system for substitute raw materials, characterized in that, include: The generation module is used to divide the bill of materials corresponding to each historical material selection record into a training dataset and a validation dataset based on the bill of materials corresponding to each historical material selection record. Based on the bill of materials corresponding to each historical material selection record in the training dataset, several primary alternative material selection rules and several secondary alternative material selection rules are automatically generated through machine learning. The primary alternative material selection rules are generated based on the bill of materials, project type, customer name, and part specifications. The secondary alternative material selection rules are generated based on the bill of materials. Each primary alternative material selection rule or secondary alternative material selection rule includes the selection weight corresponding to the parameter indicators of each material in the bill of materials. The verification module, connected to the generation module, is used to obtain a verification dataset and verify the verification pass rate corresponding to each of the main alternative material selection rules and each of the auxiliary alternative material selection rules based on the verification dataset. The material selection module, connected to the verification module, is used to select the primary substitute material selection rule and the secondary substitute material selection rule with the highest verification pass rate as the target material selection rule, and automatically select substitute parts for the original material according to the target material selection rule and the bill of materials received in real time, including: automatically selecting the primary substitute material according to the target material selection rule and the bill of materials received in real time, the project type, the customer name and the part specifications; and automatically selecting the secondary substitute material according to the target material selection rule and the bill of materials received in real time. The update module is connected to the material selection module and the generation module respectively, and is used to store the material list received in real time and the result of automatically selecting the substitute material of the original material as material selection data in the historical material selection record; The sending module, connected to the updating module, is used to send the historical material selection record to manual review after a preset number of automatic selection of substitute materials for the original material. The optimization module is connected to both the material selection module and the sending module. When the manual review result is "passed", it automatically selects a substitute material for the original material according to the current target material selection rule and the material list received in real time. When the manual review result is "failed", it optimizes the current target material selection rule according to the historical material selection record, and automatically selects a substitute material for the original material according to the optimized material selection rule and the material list received in real time.

5. The automatic selection system for substitute raw materials according to claim 4, characterized in that, The generation module is also used to automatically generate several material selection rules based on the bill of materials and material selection reference type parameters corresponding to each historical material selection record through machine learning. Each material selection reference type parameter corresponds to several preset material selection reference bases.

Images