JP2017161991A - Quality evaluation system, quality evaluation method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a quality evaluation system for accurately detecting an abnormality.SOLUTION: The quality evaluation system for evaluating a product manufactured through a plurality of steps comprises: a data acquisition unit for acquiring, in at least one step of the plurality of steps, quality evaluation data on a product and processing data on manufacture of the product; a quality evaluation model generation unit for generating an evaluation model which evaluates quality of the product on the basis of the quality evaluation data and the processing data acquired by the data acquisition unit; and an abnormality determination unit for detecting an abnormality in the step on the basis of the model determined by the quality evaluation model generation unit.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a quality evaluation system, a quality evaluation method, and a program.

  In casting processing that uses surface treatment or heat treatment using chemical reaction, quality is ensured by monitoring whether the value of the data to be evaluated acquired in each manufacturing process is within the prescribed control values. I am trying.

  In addition, as a technique related to the discovery of occurrence of defects or the like in the manufacturing process, for example, Patent Document 1 summarizes data acquired in a plurality of manufacturing processes, and an inspection process or a production process in which a large defect rate occurs. There is a description of a production management system that displays.

Japanese Patent No. 4500906

However, when the manufacturing process is complicated or a manual chemical analysis value is required, the frequency with which the data to be evaluated can be acquired is low, and the acquired data varies. Therefore, even if the evaluation target data of a certain process is within the control value, the product may eventually become abnormal, and the causal relationship between the control value and the quality is not always clear.
Moreover, even if all of the enormous types of data can be acquired, it is not always possible to grasp the state of the manufacturing process simply by collecting the data as in the method described in Patent Document 1.

  Accordingly, an object of the present invention is to provide a quality evaluation system, a quality evaluation method, and a program that can solve the above-described problems.

According to the first aspect of the present invention, the quality evaluation system is a quality evaluation system for evaluating a product manufactured by performing a plurality of steps, wherein at least one of the plurality of steps, A data acquisition unit that acquires product quality evaluation data in the process and processing data related to the manufacture of the product, and an evaluation model that evaluates the quality related to the product based on the quality evaluation data and processing data acquired by the data acquisition unit. You may provide the quality evaluation model production | generation part to produce | generate and the abnormality determination part which detects the abnormality in the said process based on the evaluation model which the said quality evaluation model production | generation part produced | generated.
Since abnormality is detected based on the evaluation model generated based on the quality evaluation data and the processing data, the quality can be evaluated with high accuracy.

The quality evaluation model generation unit according to the second aspect of the present invention includes a factor specifying unit that extracts, as an evaluation factor, a parameter having a relatively large influence on the evaluation result indicated by the quality evaluation data from the processing data; An abnormality determination model generation unit that generates an evaluation model for evaluating quality related to the product based on data related to the evaluation factor among the acquired processing data may be provided.
Since an evaluation factor is specified and an evaluation model based on the evaluation factor is generated, a highly accurate evaluation model can be generated.

The quality evaluation model generation unit according to the third aspect of the present invention is based on the correlation between the data indicating the degree of accumulation of processing included in the processing data and the value of the evaluation factor. A deterioration degree evaluation model generation unit that generates an evaluation model for evaluating the deterioration degree may be provided.
By generating an evaluation model that evaluates the environment related to manufacturing using data indicating the cumulative degree of processing as an index, it is possible to detect the degree of deterioration of environmental factors that cause a reduction in quality. In addition, since evaluation can be performed using a value having no variation as an index, stable evaluation is possible.

The quality evaluation model generation unit according to the fourth aspect of the present invention is based on the correlation between the data indicating the degree of accumulation of processing included in the processing data and the quality evaluation data of the product. A degradation degree evaluation model generation unit that generates an evaluation model for evaluating the degree of deterioration of the image.
By generating an evaluation model that evaluates the environment related to manufacturing using data indicating the cumulative degree of processing as an index, it is possible to detect the degree of deterioration of environmental factors that cause a reduction in quality. In addition, since evaluation can be performed using a value having no variation as an index, stable evaluation is possible.

The plurality of steps in the fifth aspect of the present invention are surface treatments, and the abnormality determination unit detects an abnormality of a product based on the evaluation model generated by the abnormality determination model generation unit, and evaluates the deterioration level You may detect the adjustment timing of the process liquid used for surface treatment based on the evaluation model which the model production | generation part produced | generated.
By doing so, it is possible to detect the degree of deterioration of the processing liquid used for the surface treatment as the degree of deterioration of the environmental factor that causes the product quality to deteriorate, and the timing of adjusting the processing liquid as a countermeasure against the detected degree of deterioration. Can be detected.

The quality evaluation model generation unit according to the sixth aspect of the present invention is configured such that the product in the process is based on quality evaluation data and processing data acquired by the data acquisition unit in each of the different processes among the plurality of processes. An evaluation model for evaluating the quality of the image may be generated.
By doing in this way, abnormality of the product manufactured by a plurality of manufacturing processes can be detected before reaching the final process.

The product manufactured by performing the plurality of steps in the seventh aspect of the present invention is a cast product, and the quality evaluation model generation unit is a component of the cast product after the dissolution treatment among the plurality of steps. You may provide the 1st evaluation model production | generation part which produces | generates the evaluation model for evaluating a value.
In this way, the quality of the cast product can be evaluated at the stage after the melting of the casting material and before reaching the final process.

The quality evaluation model generation unit according to the eighth aspect of the present invention is based on a result of a quality test on a cast product manufactured by performing the plurality of steps and a component value of a material of the cast product. A data selection unit for calculating the first evaluation model generation unit, and the first evaluation model generation unit may generate an evaluation model based on the target component value calculated by the data selection unit.
Since the component value of the material can be evaluated by the component value evaluation model generated based on the target configuration of the component value, the quality of the material can be accurately evaluated.

The product manufactured by performing the plurality of steps in the ninth aspect of the present invention is a cast product, and the quality evaluation model generation unit performs processing related to the heat treatment before the heat treatment among the plurality of steps. You may provide the 2nd evaluation model production | generation part which produces | generates the evaluation model for evaluating a plan.
By doing so, it is possible to evaluate the validity of the designed heat treatment plan before actually performing the heat treatment.

  According to a tenth aspect of the present invention, a quality evaluation method is a quality evaluation method for a quality evaluation system for evaluating a product manufactured by performing a plurality of steps, and at least one of the plurality of steps. In the process, the quality evaluation data of the product in the process and the processing data related to the manufacture of the product are acquired, and based on the acquired quality evaluation data and the processing data, an evaluation model for evaluating the quality related to the product is generated, An abnormality in the process may be detected based on the generated evaluation model.

  According to an eleventh aspect of the present invention, the program causes a computer of a quality evaluation system that evaluates a product manufactured by performing a plurality of steps to perform at least one of the plurality of steps in the step. Means for acquiring product quality evaluation data and process data relating to the manufacture of the product; means for generating an evaluation model for evaluating quality related to the product based on the acquired quality evaluation data and process data; and the generated evaluation model Based on the above, it may function as a means for detecting an abnormality in the process.

  According to the present invention, it is possible to create an appropriate quality evaluation model even if the amount of data necessary for product evaluation is enormous, and to detect product abnormalities based on the quality evaluation model.

It is a functional block diagram which shows an example of the quality evaluation system in 1st embodiment of this invention. It is a figure which shows an example of the manufacturing process in 1st embodiment of this invention. It is a figure which shows the outline of the quality evaluation process by the quality evaluation system in 1st embodiment of this invention. It is a figure which shows an example of the factor extraction in 1st embodiment of this invention. It is a figure which shows the example of an output of the determination result in 1st embodiment of this invention. It is a flowchart which shows an example of the production | generation process of the abnormality determination model in 1st embodiment of this invention. It is a flowchart which shows an example of the abnormality determination process in the manufacturing process in 1st embodiment of this invention. It is a figure which shows the relationship between the processing number of the product in 1st embodiment of this invention, and the generation | occurrence | production state of a rejected product. It is a figure which shows the relationship between the process number of the product in 1st embodiment of this invention, and the state of a process liquid. It is a 1st flowchart which shows an example of the calculation method of the management value which detects the replacement timing of the process liquid in 1st embodiment of this invention. It is a 2nd flowchart which shows an example of the calculation method of the management value which detects the replacement timing of the process liquid in 1st embodiment of this invention. It is a functional block diagram which shows an example of the quality evaluation system in 2nd embodiment of this invention. It is a figure which shows an example of the manufacturing process in 2nd embodiment of this invention. It is a figure which shows the outline of the quality evaluation process by the quality evaluation system in 2nd embodiment of this invention. It is a figure which shows an example of the 2nd evaluation model in 2nd embodiment of this invention. It is a flowchart which shows an example of the production | generation process of the 1st evaluation model in 2nd embodiment of this invention. It is a flowchart which shows an example of the production | generation process of the 2nd evaluation model in 2nd embodiment of this invention.

<First embodiment>
Hereinafter, a quality evaluation system according to an embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a functional block diagram showing an example of a quality evaluation system in the first embodiment of the present invention.
The quality evaluation system 10 finally determines that a product manufactured through a plurality of manufacturing processes is ineligible even if the data to be evaluated in the intermediate manufacturing process is within the control value range. To prevent it. In a casting process that requires chemical treatment-based surface treatment or heat treatment, the amount of data related to the process is enormous, and it is difficult to detect the cause of the abnormality using all these data. In addition, it is not easy to acquire the amount of data necessary for product evaluation. The quality evaluation system 10 generates an evaluation model for detecting a product abnormality using limited data obtained in the manufacturing process, and performs a quality inspection of the product based on the evaluation model. In the first embodiment, the surface treatment is described as an example.

The quality evaluation system 10 includes, for example, one or a plurality of computers (such as server terminal devices). As shown in the figure, the quality evaluation system 10 includes a data acquisition unit 11, a quality evaluation model generation unit 12, an abnormality determination unit 13, a storage unit 14, and an input / output unit 15.
The data acquisition unit 11 acquires the quality evaluation data of the product in the process and the processing data related to the manufacture of the product in at least one of the plurality of processes for the product manufactured by performing the plurality of processes. Quality evaluation data is data obtained by evaluating the quality of a product. The quality evaluation data is, for example, an inspection result obtained by performing a quality inspection on the product. The processing data is data relating to various manufacturing processes in manufacturing processes such as product materials, manufacturing methods, and manufacturing environments.
The quality evaluation model generation unit 12 generates an evaluation model for evaluating the quality related to the product based on the quality evaluation data and the processing data acquired by the data acquisition unit 11.

The abnormality determination unit 13 detects an abnormality in the manufacturing process based on the evaluation model generated by the quality evaluation model generation unit 12.
The storage unit 14 stores various information such as quality evaluation data and processing data.
The input / output unit 15 receives an input operation by the user. Further, the input / output unit 15 outputs a determination result by the abnormality determination unit 13 and the like.

The quality evaluation model generation unit 12 includes a factor specifying unit 121, an abnormality determination model generation unit 122, and a deterioration degree evaluation model generation unit 123.
The factor specifying unit 121 extracts, from the processing data, a parameter that has a relatively large influence on the evaluation result indicated by the evaluation data as an evaluation factor.
The abnormality determination model generation unit 122 generates a quality evaluation model for evaluating the quality of the product based on the evaluation factor, based on the data regarding the specified evaluation factor among the processing data.
The degradation level evaluation model generation unit 123 generates a model for evaluating the degradation level of the environment related to product manufacturing based on the correlation between the data indicating the cumulative degree of processing included in the processing data and the evaluation factor or the quality evaluation data. Generate.

FIG. 2 is a diagram showing an example of a manufacturing process in the first embodiment of the present invention.
As shown in FIG. 2, when a certain product is manufactured, a plurality of steps of processing, surface treatment, painting, and assembly are required. Each process of processing, surface treatment, painting, and assembly also includes a plurality of processes. In the present embodiment, detection of an abnormality that occurs in a surface treatment called anodizing treatment among these manufacturing steps will be described as an example. The anodizing treatment also includes a plurality of steps such as “washing”, “deoxy soy”, “water washing A”, “anodizing”, “water washing B”, “chromate treatment”, “drying”, etc. The abnormality was judged by quality inspection after the process. This quality inspection takes time (for example, several weeks) until results are obtained, and if a rejected product occurs as a result of the quality inspection, work such as inspection of all products produced after that occurs. In some cases, it may be necessary to rework a product that has already completed this manufacturing process. Even if the data to be monitored during the anodizing process is within the control value range, the result of the quality inspection sometimes fails. In order to improve productivity, appropriate quality evaluation in anodizing treatment and generation of rejected products have been desired. Therefore, in the present embodiment, a high-accuracy quality evaluation method that does not require time until recognizing the result and does not cause rework, and a method that prevents the occurrence of rejected products are provided.

FIG. 3 is a diagram showing an outline of the quality evaluation process by the quality evaluation system in the first embodiment of the present invention.
The quality evaluation system 10 uses environmental data, inspection data, liquid state data, and surface treatment monitoring data as input data. The environmental data is recorded in the environmental data DB 21, the inspection data is recorded in the inspection data DB 22, the liquid state data is recorded in the liquid state data DB 23, and the surface treatment monitoring data is recorded in the surface treatment monitoring data DB 24. In addition, the storage unit 14 stores the environmental data DB 21, the inspection data DB 22, the liquid state data DB 23, and the surface treatment monitoring data DB 24. The factor specifying unit 121 performs a factor specifying process for selecting a factor that causes a rejected product using these input data. The abnormality determination model generation unit 122 performs a learning process from the factor specified by the factor specifying unit 121, the environmental data, the liquid state data, and the surface treatment monitoring data, and determines an abnormal product among the anodized products. An abnormality determination model is generated. The abnormality determination model is an example of an evaluation model. The factor specifying unit 121 generates an abnormality determination model as an evaluation model. The abnormality determination unit 13 performs a determination process for determining whether the currently processed product is abnormal from the abnormality determination model, the environmental data, the liquid state data, and the surface treatment monitoring data. The abnormality determination unit 13 outputs the determination result to the input / output unit 15. The input / output unit 15 outputs the determination result to a display or the like. Further, the input / output unit 15 reads information on a treatment method corresponding to the determination result from the treatment data, and outputs the treatment method to a display or the like. Each data and processing will be described in more detail.

The environmental data is data including date and time, temperature, humidity, and replacement time of the processing solution used for the surface treatment.
The inspection data is data in which data indicating product quality is recorded. The data indicating quality is, for example, data indicating whether the quality is normal or how much the quality is degraded when the quality is not normal. In the present embodiment, in addition to the product to be actually manufactured, the test piece for quality evaluation is processed at a predetermined interval, and the result of the quality inspection (normality / deterioration degree) of the test piece is recorded. To do. For example, when the result of the quality inspection of the test piece 1 processed at time T1 is acceptable, the product that has been anodized before the next test piece quality inspection after the test piece 1 is considered normal. After that, when the result of the quality inspection of the test piece 2 processed at time T2 becomes the deterioration degree 1, the products processed after the test piece 2 are regarded as the deterioration degree 1. In the inspection data, identification information (ID) and quality inspection results are recorded in association with each test piece. For example, a degree of deterioration of 1 to 5 is set as the degree of deterioration. Further, it may be determined by setting a threshold value such that the deterioration degree is up to 4, and the deterioration degree is 5, and the deterioration degree is 5.

The liquid state data includes information such as the pH of the treatment liquid, the temperature of the treatment liquid, the content (concentration) of the components constituting the treatment liquid, and the date and time when the liquid was collected for each treatment rod used in the anodizing treatment. Yes.
The surface treatment monitoring data includes the test piece ID, treatment rod, treatment date and time, temperature during treatment, and the like.
Environmental data, liquid state data, and surface treatment monitoring data may be referred to as processing data, and inspection data may be referred to as quality evaluation data.

  For example, various sensors such as a temperature sensor are provided in a soot filled with each treatment solution used for anodizing treatment, the room where the treatment is performed, and the values measured by the various sensors each time the test piece is processed are the environmental data DB 21. The surface treatment monitoring data DB 24 is accumulated. Further, the inspector collects the treatment liquid of each bottle at a predetermined interval, analyzes the component concentration, PH, etc. of the treatment liquid, and records the analysis result in the liquid state data DB 23. The inspector records the result of the quality inspection for the test piece in the inspection data DB 22.

  The factor specifying unit 121 performs machine learning for a plurality of test pieces using environmental data, inspection data, liquid state data, and surface treatment monitoring data, and the degree of deterioration corresponding to the failure is recorded in the inspection data. Analyze the factors. As a machine learning technique, a technique such as a decision tree or Random Forest can be used. Also, for example, C5.0 can be used as the decision tree algorithm. Since the anodizing process consists of a plurality of processes (FIG. 2), the number of data items becomes enormous when all the data related to the process is acquired. For example, there is a method of acquiring all data related to processing for all products and detecting a state where an abnormality has occurred, but inspection data is not 100% inspection of products as in this embodiment, but limited by test pieces. If the number of inspection items is not sufficient, the number of inspection data with respect to the number of data items is not sufficient, and there is a possibility that appropriate detection cannot be performed. Therefore, in this embodiment, a data item (parameter) that has a relatively large influence on the inspection result (pass / deterioration degree) is specified as an evaluation factor by using a method such as a decision tree. Next, an example of main evaluation factors identified as a result of actually applying a technique such as a decision tree in anodizing processing will be described.

FIG. 4 is a diagram showing an example of factor extraction in the first embodiment of the present invention.
FIG. 4 is an example of a decision tree generated using, as teacher data, environmental data, inspection data, liquid state data, and surface treatment monitoring data recorded in association with a plurality of test pieces. Using decision trees, analyze the relationship between one or more parameters included in environmental data, liquid state data, and surface treatment monitoring data and the occurrence of rejected products, and cause of rejected products related to one or more parameters Can be extracted. Further, when a decision tree is used, a tree-like analysis result having a hierarchical structure can be obtained. For example, if the environmental data parameter A1 is analyzed as a cause of failure when the condition A2 satisfies the condition A2, if the condition B2 satisfies the condition B2 and satisfies the condition B2 An analysis result may be obtained that indicates that the probability of occurrence of a pass is increased. In this case, each of the condition A2 related to the parameter A1 and the condition B2 related to the parameter B1 can be considered as candidates for factors that cause the rejected product. FIG. 4 shows a certain hierarchy extracted from the tree-like analysis results indicating various candidates that may be the cause of the rejected product obtained by the decision tree. Illustrated in FIG. 4 is an analysis result regarding the concentration of a substance contained in the treatment liquid in any of the plurality of treatment vessels used for the surface treatment. Here, as an example, the total solid content concentration in the chromate treatment tank will be taken up. Conventionally, the total solid content concentration of chromate treated soot is included in the monitoring items for quality control, and the control value of the total solid content concentration determined in the conventional monitoring is determined to be “Xmg / l” or less. As a result, according to the analysis result by the factor specifying unit 121, the condition that the value of the total solid content concentration is “Ymg / l” or more (Y <X) is one of the factors causing the rejected product. It can be judged. As described above, when a method such as a decision tree is used, an evaluation factor (for example, total solid content concentration) that causes a failure and a condition (≧ Ymg / l) that causes the failure can be calculated. Since the identified evaluation factor is based on actual processing data and quality evaluation data, the reliability is high. In addition, by specifying the evaluation factor, the next time the abnormality determination model is generated, the amount of data that can be used for model generation is limited to the data measured with respect to the test piece, and this compensates for the disadvantage that the amount is small. it can. Note that the evaluation factor illustrated here is one of a plurality of evaluation factors obtained by the decision tree, and other evaluation factors can also be extracted.

  Returning to FIG. 3, the learning process performed by the abnormality determination model generation unit 122 will be described. The abnormality determination model generation unit 122 is an evaluation factor specified by the factor specifying unit 121 among inspection data for a plurality of test pieces, environmental data, liquid state data, and surface treatment monitoring data recorded in association with the plurality of test pieces. Is used to generate an abnormality determination model for determining whether the product is normal or abnormal. For the generation of the abnormality determination model, a machine learning method such as one class SVM (support vector machine), an NN (neural network), or a BN (bayesian network), or a multivariate analysis method such as an MT method can be used. For example, when an abnormality determination model is generated using one class SVM, the abnormality determination model generation unit 122 learns data having a value that passes the evaluation factor specified by the factor specifying unit 121, and is in a normal state class. Is generated. For example, when the MT method is used, the abnormality determination model generation unit 122 generates a unit space from data having a value that passes the evaluation factor specified by the factor specifying unit 121. In general, generation of an abnormality determination model using a large number of data items requires a large amount of data, and when the amount of data is insufficient, it is difficult to generate a model having sufficient accuracy. Since the data items used for generating the abnormality determination model are limited to the evaluation factors extracted by the factor specifying unit 121, it is sufficient even if the data amount is limited to the processing data acquired in relation to the test piece. An abnormality determination model having accuracy can be generated. The abnormality determination model generation unit 122 records the generated abnormality determination model (the function defining the above class and unit space) in the storage unit 14.

  However, if a method such as a decision tree is used, it is possible to extract evaluation factors and determination conditions that are rejected as illustrated in FIG. Therefore, when a comparatively simple logic with high versatility for determining pass or fail is obtained as a result of the analysis by the factor specifying unit 121, the abnormality determination model generation unit 122 may use the logic as an abnormality determination model. Good.

  The abnormality determination unit 13 performs determination processing for determining whether or not the product is abnormal based on the processing data regarding the product to be evaluated and the abnormality determination model. For example, when the abnormality determination model is generated using one class SVM, the abnormality determination unit 13 compares the data that becomes the evaluation factor in the processing data to be evaluated with the class that has learned the normal state. The abnormality determination unit 13 determines that the processing data is normal when the processing data serving as the evaluation factor is included in the class, and determines that the processing data is abnormal when the processing data is an outlier. For example, when an abnormality determination model is generated using the MT method, the abnormality determination unit 13 calculates an MD (Mahalanobis Distance) value between the processing data serving as an evaluation factor and the basic space, and the MD value is a predetermined threshold value. If it is within the range, it is determined that the product related to the processing data is normal, and if the predetermined threshold is exceeded, the processing data is determined to be abnormal. If the abnormality determination unit 13 determines that the product is normal, the product related to the processing data is also normal and is likely to pass the quality inspection. If the product is determined to be abnormal, the product related to the processing data is abnormal and the quality is high. There is a high possibility of failing the inspection. The abnormality determination unit 13 outputs the determination result to the input / output unit 15, and the input / output unit 15 displays the determination result on the display. This determination process does not take several weeks like the conventional quality inspection. According to the abnormality determination unit 13, it is possible to quickly predict whether the processed product will pass or fail. Thereby, it can prevent producing the product which becomes disqualified, or generating rework. Moreover, since the abnormality determination part 13 performs determination using the abnormality determination model based on an evaluation factor, the accuracy of determination is high and it can detect the abnormality in a manufacturing process reliably.

FIG. 5 is a diagram illustrating an output example of the determination result in the first embodiment of the present invention.
FIG. 5A is a display example when the determination result by the abnormality determination unit 13 is abnormal. As a display example of the abnormality level, “high quality”, “low quality”, and the like may be displayed when the determination result by the abnormality determination unit 13 is acceptable. For example, when the abnormality determination model is generated using the MT method, the input / output unit 15 is close to the threshold even if the MD value acquired from the abnormality determination unit 13 is equal to or less than a predetermined threshold indicating pass. If the value is indicated, “low quality” is displayed, and if the value is sufficiently small, “high quality” is displayed.

  FIG. 5B is a display example of an abnormality factor when it is determined that there is an abnormality. For example, when an abnormality determination model is generated using the MT method, the abnormality determination unit 13 can extract a factor (abnormality factor) that has greatly contributed to the determination of abnormality by analyzing the SN ratio and sensitivity. . The input / output unit 15 outputs the abnormality factor analyzed by the abnormality determination unit 13 as shown in FIG. The PH of 漕 A, the processing time of 漕 A, the temperature of 漕 A, the temperature of 漕 B, and the like are evaluation factors specified by the factor specifying unit 121.

  FIG. 5C is a display example of a treatment plan for the abnormal factor. In the treatment data DB 25, for example, a treatment method effective for preventing the occurrence of the abnormality is recorded in association with the abnormality factor. The input / output unit 15 reads out the treatment method corresponding to the abnormality factor from the treatment data using the abnormality factor acquired from the abnormality determination unit 13, and outputs the treatment method as shown in FIG.

  Next, the flow of the abnormality determination model generation process and the product abnormality determination process of this embodiment will be described.

FIG. 6 is a flowchart showing an example of generation processing of the abnormality determination model in the first embodiment of the present invention.
As a premise, processing data and quality evaluation data related to a plurality of test pieces in an amount necessary for generating an abnormality determination model are accumulated in the environmental data DB 21, the inspection data DB 22, the liquid state data DB 23, and the surface treatment monitoring data DB 24. To do.
First, the user inputs an abnormality determination model generation instruction operation to the quality evaluation system 10. The input / output unit 15 receives the input of this operation, and instructs the data acquisition unit 11 to acquire data necessary for generating the abnormality determination model. Next, the data acquisition unit 11 acquires processing data of a plurality of test pieces (step S11). Specifically, the data acquisition unit 11 reads the liquid state data from the surface treatment monitoring data DB 24 using the ID of each test piece. Further, the data acquisition unit 11 reads and acquires the environmental data being processed from the environmental data DB 21 and the liquid state data being processed from the liquid state data DB 23 based on the processing date and time of each test piece. Next, the data acquisition unit 11 acquires the result of the quality test performed on each of the plurality of test pieces (step S12).

  Next, the factor specifying unit 121 executes an evaluation factor specifying process (step S13). First, the factor specifying unit 121 generates teacher data in which test data, environmental data, liquid state data, and surface treatment monitoring data relating to each test piece are associated with each test piece. Next, for example, the factor specifying unit 121 generates a decision tree using an algorithm such as C5.0 by using teacher data, and analyzes a factor causing the product to be rejected and its determination logic. The factor specifying unit 121 records the analysis result in the storage unit 14. In addition, the factor specifying unit 121 instructs the abnormality determination model generation unit 122 to generate an abnormality determination model. Next, the abnormality determination model generation unit 122 generates an abnormality determination model using the processing data (for example, the total solid component concentration) that is the evaluation factor specified by the factor specifying unit 121 (step S14). For example, the abnormality determination model generation unit 122 generates an abnormality determination model based on data in which processing data serving as a factor indicates a normal value using a one-class SVM, MT method, or the like. The abnormality determination model generation unit 122 records the generated abnormality determination model in the storage unit 14 (step S15). This completes the generation process of the abnormality determination model. When the generation process of the abnormality determination model is completed, it is possible to perform pass / fail determination of the actual product.

FIG. 7 is a flowchart showing an example of abnormality determination processing in the manufacturing process according to the first embodiment of the present invention.
First, the data acquisition part 11 acquires the process data which the factor specific | specification part 121 specified as an evaluation factor at predetermined intervals, for example (step S20). For example, if the data specified as the evaluation factor is environmental data or surface treatment monitoring data, the data acquisition unit 11 acquires a desired value from various sensors. If the data specified as the evaluation factor is liquid state data, the data acquisition unit 11 acquires a desired value from the liquid state data DB 23. The data acquisition unit 11 outputs the acquired processing data to the abnormality determination unit 13. Next, the abnormality determination unit 13 reads the abnormality determination model from the storage unit 14, and determines the abnormality of the processing data based on the processing data and the abnormality determination model acquired by the data acquisition unit 11 (step S21). Further, the abnormality determination unit 13 analyzes the abnormality factor if the abnormality factor can be analyzed based on the abnormality determination model. The abnormality determination unit 13 outputs a determination result or the like to the input / output unit 15. The input / output unit 15 displays the determination result and the like on the display (step S22). The inspector looks at the judgment result, and if “abnormal” is displayed, the inspector knows that there is a high possibility that a rejected product will be manufactured in the processing environment indicated by the current processing data, and temporarily manufactures it. It is possible to take measures such as stopping the process and performing a more detailed inspection or taking measures according to the treatment plan displayed by the input / output unit 15.

  So far, the analysis of the cause of the rejected product and the method for detecting the abnormality in the manufacturing process based on the analysis result have been described. Next, a method for preventing the occurrence of rejected products will be considered. As described above, it was possible to extract the total solid component concentration of the chromate treated soot as an abnormal factor of rejected products. For example, it may be possible that the generation of rejected products may be prevented if the chromate treatment tank treatment liquid is replaced with the total solid component concentration of Yg / ml or more as a guide. However, in general, when a chemical analysis value or the like is used as an index for judging the deterioration of the processing liquid, it is difficult to make a judgment because of large variations in values. Therefore, it is considered to determine the deterioration of the processing liquid using a stable numerical value such as the number of processing cases as an index. When the type of product to be processed changes, the shape and size of the product can be adopted as a scale value for the number of processing cases and can be used as an indicator of deterioration.

FIG. 8 is a diagram showing the relationship between the number of processed products and the occurrence status of rejected products in the first embodiment of the present invention.
FIG. 8 is a graph showing the timing at which a failure occurs as a result of the quality inspection performed on the anodized product in relation to the cumulative number of processing. The vertical axis in FIG. 8 indicates the number of processed products that have been anodized, and the horizontal axis indicates the cumulative number of processed items. One bar graph indicates the number of processed products that have been anodized for one order. In the figure, hatched portions indicate the number of cases determined to be unacceptable as a result of the quality inspection. This example shows that an abnormality is likely to occur when the cumulative number of processes exceeds “Z”.

FIG. 9 is a diagram showing the relationship between the number of processed products and the state of the processing liquid in the first embodiment of the present invention.
FIG. 9 is a graph showing the relationship between the cumulative number of processed anodized products and the total solid content concentration of the chromate treated soot. The vertical axis in FIG. 9 represents the total solid content concentration in the chromate treatment tank, and the horizontal axis represents the cumulative number of treatments. When the condition that the total solid content concentration is Yg / ml or more, which is the evaluation factor specified by the factor specifying unit 121, is superimposed on this graph, the cumulative number of processing when the total solid content concentration is Yg / ml is about Z. Indicates that there is. In other words, it indicates that there is a correlation between the tendency of the product to easily become abnormal, the cumulative number of processing cases, and the total solid content concentration. The total solid content concentration tends to vary in the measured value, but the variation does not occur as long as the cumulative number of processed cases. Therefore, if the cumulative treatment number Z is used as a guide and the chromate treatment tank treatment liquid is replaced, the treatment liquid can be stably replaced at an appropriate timing, and the occurrence of rejected products can be prevented. It is done. Based on this knowledge, a method for detecting the timing of liquid treatment replacement using stable parameters such as the number of treatments will be described. Note that data indicating the cumulative degree of processing using the processing liquid, such as the processing area, as well as the number of processing cases, can be considered as candidates for the index of liquid replacement in the anodizing process. Hereinafter, the index candidate will be described as the number of processing cases.

FIG. 10 is a first flowchart illustrating an example of a management value calculation method for detecting the replacement timing of the processing liquid in the first embodiment of the present invention.
First, the data acquisition unit 11 acquires the number of processed products (step S30). For example, the data acquisition unit 11 may acquire the number of processes input to the quality evaluation system 10 by the inspector. Alternatively, when one test piece is processed for each product 100, the number of test pieces processed by the data acquisition unit 11 after the previous processing liquid replacement may be read from the quality evaluation data DB 22 and multiplied by 100. .

  Next, the deterioration degree evaluation model generation unit 123 obtains a correlation between the evaluation factor (for example, the total solid content concentration) specified by the factor specifying unit 121 and the number of processing cases (step S31). The deterioration degree evaluation model generation unit 123 obtains a correlation between the total solid content concentration and the number of processing cases using a method such as regression analysis, for example. Next, the deterioration level evaluation model generation unit 123 calculates a management value that serves as a guide for liquid replacement timing (step S32). Specifically, the degradation level evaluation model generation unit 123 determines that the total solid content concentration is Y based on the correlation Y obtained in step S32 and the threshold value Y (g / ml) of the condition that causes the failure determination by the factor specifying unit 121. The number of processing cases (Z in FIG. 9) when (g / ml) is obtained is calculated. The calculated number of treatments is a management value that serves as an indication of the liquid treatment replacement timing. This management value is an example of an evaluation model for liquid treatment replacement timing. The degradation degree evaluation model generation unit 123 calculates a management value as an evaluation model. Next, another method for obtaining the management value will be described.

FIG. 11 is a second flowchart illustrating an example of a management value calculation method for detecting the replacement timing of the processing liquid in the first embodiment of the present invention.
First, as in step S30 of FIG. 10, the data acquisition unit 11 acquires the number of processed products (step S33). Next, the data acquisition unit 11 acquires product quality evaluation data (step S34). For example, when the quality inspection is performed on each product, the data acquisition unit 11 acquires the result of the quality inspection input to the quality evaluation system 10 by the inspector. Alternatively, when using the test piece inspection result, the test result of the test piece processed by the data acquisition unit 11 after the previous processing liquid replacement timing (environmental data DB 21) is read from the quality evaluation data DB 22, and the result is used as the test piece. You may use as quality evaluation data of the product processed between. Next, the deterioration degree evaluation model generation unit 123 obtains a correlation between the quality evaluation data and the number of processing cases (step S35). For example, the deterioration degree evaluation model generation unit 123 analyzes the relationship between the deterioration degree indicated by each quality evaluation data and the number of processing cases using a decision tree or the like. Next, the deterioration degree evaluation model generation unit 123 calculates a management value (step S36). For example, the deterioration level evaluation model generation unit 123 extracts a condition that causes a deterioration level (deterioration level = 5) to a degree of failure. The extracted condition is a management value. The degradation degree evaluation model generation unit 123 calculates a management value as an evaluation model.

  According to the present embodiment, even when the processing data related to the product to be evaluated is enormous and the acquisition frequency of the data is low, or even when there is a variation in the acquired data, it is possible to detect abnormality. Necessary evaluation factors can be extracted to generate an abnormality determination model (quality evaluation model). Furthermore, when the evaluation factor has a variation, it is possible to determine the degradation state of the manufacturing environment based on the correlation between the evaluation factor and a stable index (such as the number of treatments) having no variation such as the number of treatments. . Further, for example, by using the number of processing cases as an index, it is possible to establish a long-term prospect in association with the production plan. In the above description, the processing liquid replacement timing is described as an example of the processing liquid adjustment timing, but the processing liquid is not limited to this example. For example, the above method may be applied to obtain a control value of adjustment timing for performing adjustment such as injecting some substance into the processing liquid.

<Second embodiment>
Hereinafter, a quality evaluation system according to two embodiments of the present invention will be described with reference to FIGS.
In the second embodiment, a casting process will be described as an example. Similarly to the first embodiment, in the case of a cast product, the number of data items of processing data related to the manufacturing process is enormous, and it is difficult to perform quality control by acquiring all the enormous data. Moreover, as will be described with reference to FIG. 13, the casting process is performed through a plurality of steps. Conventionally, material component value inspection was performed in the middle of the process, but even if it passed the inspection without any problems, the final quality inspection may be rejected, resulting in a rejected product. It was difficult to understand the factors that make it.

FIG. 12 is a functional block diagram showing an example of the quality evaluation system in the second embodiment of the present invention.
Of the configurations according to the second embodiment of the present invention, the same components as those constituting the quality evaluation system 10 according to the first embodiment of the present invention are denoted by the same reference numerals, and description thereof is omitted. The quality evaluation system 10a according to the second embodiment includes a quality evaluation model generation unit 12a instead of the quality evaluation model generation unit 12 in the configuration of the first embodiment. Further, the quality evaluation model generation unit 12a includes a data selection unit 124, a first evaluation model generation unit 125, and a second evaluation model generation unit 126.

The quality evaluation model generation unit 12a is based on at least one of the quality evaluation data and the processing data acquired by the data acquisition unit 11 in a different process among a plurality of processes for a product manufactured by performing a plurality of processes. Thus, an evaluation model for evaluating the quality of the product in the process is generated.
The data selection unit 124 calculates an ideal component value composition ratio of a material used in a first evaluation model described later.
The 1st evaluation model production | generation part 125 produces | generates the 1st evaluation model for evaluating the component value of the material of a casting product after a melt | dissolution process among the manufacturing processes of a casting product.
The 2nd evaluation model production | generation part 126 produces | generates the 2nd evaluation model for evaluating the process plan which concerns on the heat processing before heat processing among the manufacturing processes of a cast product.

FIG. 13 is a diagram showing an example of a manufacturing process in the second embodiment of the present invention.
As FIG. 13 shows, when manufacturing a cast product, the process of melt | dissolution, casting, heat processing, and a process is required. Moreover, as a quality inspection, in this embodiment, a component value inspection and an inspection result determination are performed after dissolution. In addition, a heat treatment plan is prepared and a heat treatment plan is determined before the heat treatment. In addition, a final mechanical test is performed after the heat treatment. In the past, component value inspection and mechanical testing were carried out, but even if the component inspection value after dissolution is within the control value, there may be cases where the final mechanical test is rejected, which may cause the quality to be affected. I couldn't figure it out. In the present embodiment, a method such as machine learning is used to grasp at which stage of the process the cause is, and the risk that the machine test is rejected is reduced. More specifically, since the quality of the cast product is determined by the material component value and heat treatment at the time of melting, each of the component value inspection after melting and the heat treatment plan before heat treatment is based on normal processing data. The generated evaluation model (first evaluation model, second evaluation model) is generated, and quality is determined based on the evaluation model.

FIG. 14 is a diagram showing an outline of the quality evaluation process by the quality evaluation system in the second embodiment of the present invention.
The quality evaluation system 10a uses component value data, inspection data, and heat treatment data as input data. The component value is recorded in the component value data DB 26, the inspection data is recorded in the inspection data DB 27, and the heat treatment data is recorded in the heat treatment data DB 28, respectively. Further, the storage unit 14 stores the component value data DB 26, the inspection data DB 27, and the heat treatment data DB 28. The data selection unit 124 calculates an ideal component configuration in the material of the cast product using the component value data and the inspection data. The first evaluation model generation unit 125 generates a first evaluation model that models the ideal component configuration calculated by the data selection unit 124. The second evaluation model generation unit 126 generates a second evaluation model for determining whether the heat treatment plan is appropriate using the component value data, the inspection data, and the heat treatment data. The abnormality determination unit 13 determines whether the component composition of the material after dissolution is within a normal range from the first evaluation model and the component value data of the material to be evaluated. In addition, the abnormality determination unit 13 determines whether or not the prepared heat treatment plan is appropriate from the second evaluation model, the component value data of the material to be evaluated, and the heat treatment plan. The input / output unit 15 outputs the determination result to a display or the like.

The component value data includes the components included in the material and the component ratio. The component value data is represented by information such as C (carbon) a% and Mn (manganese) b%.
In the inspection data, data indicating the test result of the cast product is recorded. The inspection data includes, for example, product ID and test result (pass, fail) information.
The heat treatment data includes information related to a treatment method performed in the heat treatment step. For example, it includes information such as the number of heat treatments, heat treatment conditions (quenching, annealing, etc.), temperature zone, cooling conditions (water cooling, air cooling, etc.), and a subcontractor when requesting heat treatment to the outside.
The component value data and the heat treatment data are called processing data, and the inspection data is called quality evaluation data.

  The data selection unit 124 uses a machine learning method such as a decision tree to determine how many components are included by using the inspection result for each casting product included in the past inspection data and the composition ratio of the material components of the casting product as teacher data. If so, it will be analyzed whether it will pass the mechanical test. As a result of the analysis, the data selection unit 124 can obtain conditions that pass the mechanical test for the component composition of the material. The data selection unit 124 calculates component configuration information (target component value) indicating the configuration of components in an ideal material, for example, C is X1% to X2%, Mn is X3% to X4%,.

  The first evaluation model generation unit 125 uses the component configuration information calculated by the data selection unit 124 to generate a first evaluation model for determining whether the material after dissolution is normal or abnormal. For generating the first evaluation model, a one-class SVM, MT method, or the like can be used. For example, when the first evaluation model is generated using one class SVM, the first evaluation model generation unit 125 learns the component configuration information calculated by the data selection unit 124 and generates an ideal state class. For example, when the MT method is used, the first evaluation model generation unit 125 generates a unit space from the component configuration information calculated by the data selection unit 124. Although the material component value of the material used in the cast product is determined by JIS standard, the material component value actually differs depending on each material manufacturer. Furthermore, since the material dissolves at a level of several thousand tons, the value varies. Therefore, using the machine learning method (1 class SVM) and the MT method, a model is generated based on an ideal component configuration analyzed from component values of a plurality of materials that have passed the machine test in the past. It was made possible to grasp the possibility of failure in advance according to the deviation from.

  The abnormality determination unit 13 performs a determination process for determining whether or not the component value data is abnormal based on the component value data (process data) of the material after dissolution that is an evaluation target and the first evaluation model. For example, when the first evaluation model is generated using one class SVM, the abnormality determination unit 13 compares the component value data with the learned ideal state class (first evaluation model). The abnormality determination unit 13 determines that the component value data is normal when the component value data is included in the class, and determines that the component value data is abnormal when the component value data is an outlier. For example, when the first evaluation model is generated using the MT method, the abnormality determination unit 13 calculates the MD value of the component value data and the basic space, and if the MD value is within a predetermined threshold, The component value data is determined to be normal, and if it exceeds a predetermined threshold, it is determined to be abnormal. When the component value data is normal, the material having the component value data is considered normal. When the component value data is abnormal, the material having the component value data is considered abnormal. In addition to simply determining whether it is normal or abnormal, it is possible to evaluate how far the target state is deviated according to the magnitude of the MD value. Therefore, it is possible to quantitatively evaluate the variation of the component value data. The input / output unit 15 displays the determination result by the abnormality determination unit 13 on the display.

Next, quality inspection for the heat treatment plan will be described.
The second evaluation model generation unit 126 uses, as teacher data, the inspection result for each cast product included in the past inspection data, the component value of the material of the cast product, and the processing method of the heat treatment performed on the cast product. By using a machine learning method such as a decision tree, a condition indicating whether a heat treatment that passes a mechanical test by analyzing a material having any component value is analyzed. Based on this analysis, the second evaluation model generation unit 126 generates a second evaluation model that can determine whether or not the condition planned for the heat treatment is set to be rejected. The user can grasp whether or not the setting is such that the heat treatment plan prepared before the heat treatment is rejected.

  The abnormality determination unit 13 performs the heat treatment planned for the material based on the component value data of the material after melting that is the evaluation target and the heat treatment plan (processing data) designed for the material and the second evaluation model. When it performs, the determination process which determines whether the result of the mechanical test of the casting product passes or fails is performed.

FIG. 15 is a diagram showing an example of the second evaluation model in the second embodiment of the present invention.
FIG. 15 is an example of a decision tree (second evaluation model) generated by the second evaluation model generation unit 126. The abnormality determination unit 13 obtains the probability of pass / fail of the input processing data according to each condition of the decision tree. For example, if the material contains more than A% C, it is subsequently determined whether the Mn ratio is B% or less, and if C is A% or less, the temperature zone is determined. By following the conditions of the decision tree in this way, the probability of pass / fail based on past results can be obtained.
Based on the second evaluation model illustrated in FIG. 15, the abnormality determination unit 13 determines the probability of passing the mechanical test and the probability of failing when the heat treatment is performed according to the planned plan. The input / output unit 15 displays the determination result by the abnormality determination unit 13 on the display.

Next, the flow of the first evaluation model generation process and the second evaluation model generation process will be described.
FIG. 16 is a flowchart illustrating an example of a first evaluation model generation process according to the second embodiment of the present invention.
First, the data acquisition unit 11 acquires quality evaluation data and component value data (step S40). Specifically, the data acquisition unit 11 acquires information on the result (pass / fail) of a mechanical test of a cast product manufactured in the past from the quality evaluation data DB 27. Moreover, the data acquisition part 11 acquires the component value data of the material corresponding to each casting product of the acquired quality evaluation data from component value data DB26. Next, the data selection unit 124 performs a data selection process (step S41). Specifically, first, the data selection unit 124 generates teacher data in which the result of the mechanical test related to the same product is associated with the component value data. Next, the data selection unit 124 calculates a component value condition (component configuration information) for the mechanical test to pass using a technique such as a decision tree for the teacher data. The data selection unit 124 outputs the calculated component configuration information to the first evaluation model generation unit 125. Next, the first evaluation model generation unit 125 generates a first evaluation model (step S42). For example, the first evaluation model generation unit 125 generates a first evaluation model using a one-class SVM, MT method, or the like. The first evaluation model generation unit 125 records the generated first evaluation model in the storage unit 14 (step S43). When the generation process of the first evaluation model is completed, the abnormality determination unit 13 can perform abnormality determination on the dissolved material.

FIG. 17 is a flowchart showing an example of the second evaluation model generation process in the second embodiment of the present invention.
First, the data acquisition unit 11 acquires quality evaluation data, component value data, and heat treatment data (step S50). Specifically, the data acquisition unit 11 acquires information on the result (pass / fail) of a mechanical test of a cast product manufactured in the past from the quality evaluation data DB 27. Moreover, the data acquisition part 11 acquires the component value data of the material corresponding to the casting product contained in the acquired quality evaluation data from component value data DB26. Further, the data acquisition unit 11 acquires information (heat treatment data) related to the heat treatment performed on the cast product included in the obtained quality evaluation data from the heat treatment data DB 28. Next, the second evaluation model generation unit 126 calculates a second evaluation model (step S51). Specifically, the second evaluation model generation unit 126 generates teacher data in which the result of the mechanical test for the same product, the component value data, and the heat treatment data are associated with each other. The second evaluation model generation unit 126 generates a second evaluation model by a technique such as a decision tree using the teacher data. The second evaluation model generation unit 126 records the generated second evaluation model in the storage unit 14 (step S52). When the generation process of the second evaluation model is completed, the abnormality determination unit 13 can determine the heat treatment plan.

  According to this embodiment, even if the processing data related to the product to be evaluated is enormous, an evaluation model for determining conditions for passing is generated by using a machine learning method for past manufacturing information. be able to. Moreover, since the possibility of a pass and a failure can be determined in two steps among a series of manufacturing steps including a plurality of steps, the quality can be more reliably evaluated.

  Each process in the quality evaluation system described above is stored in a computer-readable recording medium in the form of a program, and the program is read out and executed by the computer of the quality evaluation system. Is called. Here, the computer-readable recording medium means a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Alternatively, the computer program may be distributed to the computer via a communication line, and the computer that has received the distribution may execute the program.

The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, what is called a difference file (difference program) may be sufficient.
Moreover, the quality evaluation system may be comprised by one computer, and may be comprised by the some computer connected so that communication was possible.

  In addition, it is possible to appropriately replace the components in the above-described embodiments with known components without departing from the spirit of the present invention. The technical scope of the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention.

DESCRIPTION OF SYMBOLS 10, 10a ... Quality evaluation system 11 ... Data acquisition part 12, 12a ... Quality evaluation model generation part 121 ... Factor identification part 122 ... Abnormality determination model generation part 123 ... Degradation degree evaluation Model generation unit 124 ... Data selection unit 125 ... First evaluation model generation unit 126 ... Second evaluation model generation unit 13 ... Abnormality determination unit 14 ... Storage unit 15 ... Input / output unit 21 ... Environmental data DB
22 ... Inspection data DB
23 ... Liquid state data DB
24 ... Surface treatment monitoring data DB
25 ... treatment data DB
26 ... Component value data DB
27 ... Inspection data DB
28 ... Heat treatment data DB

Claims (11)

A quality evaluation system for evaluating a product manufactured by performing a plurality of processes,
In at least one of the plurality of processes, a data acquisition unit that acquires quality evaluation data of the product in the process and processing data related to manufacture of the product;
Based on the quality evaluation data and processing data acquired by the data acquisition unit, a quality evaluation model generation unit that generates an evaluation model for evaluating the quality related to the product;
Based on the evaluation model generated by the quality evaluation model generation unit, an abnormality determination unit that detects an abnormality in the process,
Quality evaluation system with The quality evaluation model generation unit
A factor specifying unit that extracts, as an evaluation factor, a parameter that has a relatively large influence on the evaluation result indicated by the quality evaluation data from the processing data;
An abnormality determination model generation unit that generates an evaluation model for evaluating quality related to the product based on data related to the evaluation factor among the acquired processing data;
The quality evaluation system according to claim 1, comprising: The quality evaluation model generation unit
A degradation level evaluation model generation unit that generates an evaluation model for evaluating the degradation level of the environment related to the manufacture of the product based on the correlation between the data indicating the cumulative degree of processing included in the processing data and the value of the evaluation factor ,
The quality evaluation system according to claim 2, comprising: The quality evaluation model generation unit
Deterioration degree evaluation model generation for generating an evaluation model for evaluating the deterioration degree of the environment related to the manufacture of the product based on the correlation between the data indicating the degree of accumulation of the processing included in the processing data and the quality evaluation data of the product Part,
The quality evaluation system according to claim 2, comprising: Each of the plurality of steps is a surface treatment,
The abnormality determination unit detects an abnormality of the product based on the evaluation model generated by the abnormality determination model generation unit, and adjusts the treatment liquid used for the surface treatment based on the evaluation model generated by the degradation degree evaluation model generation unit Detect timing,
The quality evaluation system according to claim 3 or claim 4. The quality evaluation model generation unit
In each of the different processes among the plurality of processes, based on the quality evaluation data and processing data acquired by the data acquisition unit, an evaluation model for evaluating the quality related to the product in the process is generated.
The quality evaluation system according to claim 1. The product manufactured by performing the plurality of steps is a cast product,
The quality evaluation model generation unit
A first evaluation model generating unit for generating an evaluation model for evaluating a component value of the cast product after the dissolution treatment among the plurality of steps;
A quality evaluation system according to claim 6. The quality evaluation model generation unit
A data selection unit that calculates a target component value based on a result of a quality test for a cast product manufactured by performing the plurality of steps and a component value of a material of the cast product,
Further comprising
The quality evaluation system according to claim 7, wherein the first evaluation model generation unit generates an evaluation model based on a target component value calculated by the data selection unit. The product manufactured by performing the plurality of steps is a cast product,
The quality evaluation model generation unit
A second evaluation model generation unit for generating an evaluation model for evaluating a treatment plan related to the heat treatment before the heat treatment among the plurality of steps;
The quality evaluation system according to any one of claims 6 to 8, further comprising: A quality evaluation method of a quality evaluation system for evaluating a product manufactured by performing a plurality of processes,
In at least one of the plurality of steps, product quality evaluation data in the step and processing data relating to the manufacture of the product are acquired,
Based on the acquired quality evaluation data and processing data, generate an evaluation model for evaluating the quality of the product,
Detecting an abnormality in the process based on the generated evaluation model;
Quality evaluation method. A computer of a quality evaluation system that evaluates a product manufactured by a plurality of processes.
Means for acquiring, in at least one of the plurality of steps, quality evaluation data of a product in the step and processing data relating to manufacture of the product;
Means for generating an evaluation model for evaluating the quality of the product based on the acquired quality evaluation data and processing data;
Means for detecting an abnormality in the process based on the generated evaluation model;
Program to function as.

Images