Manufacturing inspection system, manufacturing inspection method, manufacturing inspection device, and program

Abstract

This manufacturing inspection system measures an object to be inspected, which is a product, through fluorescent fingerprint measurement in a product manufacturing process and outputs a manufacturing inspection result. The manufacturing inspection system comprises: a spectrofluorometer that measures fluorescent fingerprint data of the object to be inspected; a fluorescent fingerprint data calculation unit that analyzes the measured fluorescent fingerprint data using a prescribed calculation method; a pass criterion storage unit that stores a pass criterion for a product specification formulated from at least one of a design specification range defined by design information of the product and a manufacturing specification range defined by manufacturing information; and a pass / fail determination unit that determines pass / fail by comparing a result of analysis by the fluorescent fingerprint data calculation unit with the pass criterion. This makes it possible to nondestructively measure the object to be inspected and automatically determine whether or not a predetermined criterion is satisfied.

Description

Manufacturing inspection system, manufacturing inspection method, manufacturing inspection apparatus, and program

[0001] The present disclosure relates to a manufacturing inspection system, a manufacturing inspection method, a manufacturing inspection apparatus, and a program.

[0002] In the transition to a resource - recycling society, an increase in the usage rate of recycled materials in resin products and the like is required.

[0003] In recent years, technologies for analyzing components, states, etc. of inspection targets using machine learning and the like have been developed.

[0004] Patent Document 1 discloses an information processing apparatus that uses a resin film as an example of a target T for obtaining inductive information in machine learning, and analyzes a fluorescence fingerprint based on principal component scores generated by a machine learning model in which principal component scores 1 and 2 of the fluorescence fingerprint for the resin film, information on a UV absorber that may be contained in a predetermined resin film, and information on the heating state of this resin film are learned as teacher data, thereby analyzing the presence or absence of a UV absorber and the heating state in the target T.

[0005] International Publication No. 2022 / 092079

[0006] In the transition to a resource - recycling society, an increase in the usage rate of recycled materials in resin products and the like is required. In such a situation, it is required to prove what percentage of recycled materials are actually contained in each product.

[0007] The information processing apparatus described in Patent Document 1 analyzes various targets and outputs the results, but it is considered that an operator determines whether or not a predetermined criterion is satisfied.

[0008] An object of the present disclosure is to non - destructively measure an inspection target and automatically determine whether or not a predetermined criterion is satisfied.

[0009] The manufacturing inspection system disclosed herein measures an inspection target, which is a product, by fluorescent fingerprint measurement during the product manufacturing process and outputs a manufacturing inspection result, and comprises: a spectrofluorometer for measuring the fluorescent fingerprint data of the inspection target; a fluorescent fingerprint data calculation unit for analyzing the measured fluorescent fingerprint data using a predetermined calculation method; an acceptance criteria storage unit for storing acceptance criteria for product specifications formulated from one or more design specification ranges defined in the product design information and manufacturing specification ranges defined in the manufacturing information; and an acceptance / rejection determination unit for determining acceptance or rejection by comparing the results analyzed by the fluorescent fingerprint data calculation unit with the acceptance criteria.

[0010] According to this disclosure, the object to be inspected can be measured non-destructively, and it can be automatically determined whether or not it meets predetermined standards.

[0011] This is a configuration diagram showing the manufacturing inspection system of Example 1. This is a flowchart showing the manufacturing inspection method of Example 1. This is a configuration diagram showing the hardware of the manufacturing inspection system of Example 1. This is a graph showing the fluorescence fingerprint data of the inspection target (1). This is a graph showing the fluorescence fingerprint data of the inspection target (2). This is a table showing the principal component scores for inspection target (1) and inspection target (2). This is a graph showing the distribution of inspection target (1) and inspection target (2) with respect to the two principal components. This is a graph showing the relationship between the fluorescence fingerprint data after pretreatment and the proportion of recycled material contained in the resin material. This is a diagram showing an example of the screen of the display device of Example 1. This is a configuration diagram showing the manufacturing inspection system of Example 2. This is a flowchart showing the manufacturing inspection method of Example 2. This is a configuration diagram showing the manufacturing inspection system of Example 3. This is a flowchart showing the manufacturing inspection method of Example 3.

[0012] The embodiments of this disclosure will be described below with reference to the drawings.

[0013] Figure 1 is a diagram showing the configuration of the manufacturing inspection system in Example 1.

[0014] The manufacturing inspection system 100 shown in this figure comprises a fluorescent fingerprint data calculation unit 1, an acceptance criteria storage unit 2, an acceptance / rejection determination unit 3, an acceptance / rejection information storage unit 4, an output unit 5, and a rejection cause analysis unit 6. The manufacturing inspection system 100 may also be called a "manufacturing inspection device." The fluorescent fingerprint data calculation unit 1, the acceptance criteria storage unit 2, the acceptance / rejection determination unit 3, the acceptance / rejection information storage unit 4, the output unit 5, and the rejection cause analysis unit 6 may each be a computing device having memory as needed, or they may be composed of a single computing device having memory. In this case, the program that causes the computing device to perform predetermined calculations may be a single program or may be divided into multiple programs. Here, the program includes a program that causes the computer to execute a manufacturing inspection procedure, which involves measuring the product, which is the object of inspection, by fluorescent fingerprint measurement in the product manufacturing process and outputting the manufacturing inspection result.

[0015] The fluorescence fingerprint data calculation unit 1 receives fluorescence fingerprint data from the spectrofluorometer 11. It then performs predetermined calculations and transmits the calculation results to the pass / fail determination unit 3 and the failure cause analysis unit 6.

[0016] The acceptance criteria storage unit 2 receives design information and manufacturing information from an external source and transmits the acceptance criteria to the pass / fail determination unit 3. The pass / fail determination unit 3 uses the acceptance criteria to determine whether the calculation result of the fluorescent fingerprint data calculation unit 1 is acceptable or unacceptable, and transmits the pass / fail information storage unit 4 and the failure cause analysis unit 6. Note that the design information and manufacturing information acquired by the acceptance criteria storage unit 2 from an external source do not necessarily need to be both acquired, as long as the acceptance criteria can be obtained; either one alone is sufficient.

[0017] The pass / fail information storage unit 4 receives an input of the individual number of the item to be inspected (e.g., serial number) from an external source and transmits pass / fail information corresponding to the individual number to the output unit 5. The output unit 5 outputs certification data containing the individual number stored in the pass / fail information storage unit 4 and the pass / fail judgment result corresponding to that individual number. The individual number is not particularly limited as long as it identifies the item to be inspected. For example, it may be a number code, symbol, barcode, two-dimensional code, etc.

[0018] The failure cause analysis unit 6 determines the pre-calculation excitation and fluorescence wavelengths that form the basis of the failure, based on the calculation results received from the fluorescence fingerprint data calculation unit 1 and the pass / fail information received from the pass / fail determination unit 3, and transmits them to the output unit 5.

[0019] The output unit 5 outputs pass / fail information corresponding to the individual number and the pre-calculation excitation and fluorescence wavelengths that form the basis for the reason for failure to the external recording medium 12. In this case, the output unit 5 may also transmit and display this data to a display device (described later).

[0020] In this figure, the spectrofluorometer 11 is installed outside the manufacturing inspection system 100, but the manufacturing inspection system 100 may include the spectrofluorometer 11.

[0021] Figure 2 is a flowchart showing the manufacturing inspection method of this embodiment.

[0022] In this diagram, first, fluorescence fingerprint data of the object to be inspected is acquired from the spectrofluorometer 11 (step S11). Next, the fluorescence fingerprint data calculation unit 1 performs calculation processing on the fluorescence fingerprint data using a predetermined calculation method (step S12). Then, the fluorescence fingerprint data calculation unit 1 transmits the calculation result to the pass / fail determination unit 3 (step S13). The pass / fail determination unit 3 also reads the pass criteria from the pass criteria storage unit 2 (step S14).

[0023] Next, the pass / fail determination unit 3 determines whether the inspected item passes or fails (step S15).

[0024] If the inspection target passes, the pass / fail determination unit 3 transmits information indicating that it has passed to the pass / fail information storage unit 4 (step S16). The pass / fail information storage unit 4 reads the individual number of the inspection target from an external source and stores it in association with the pass / fail information (step S17). The pass / fail information may also be stored on an external recording medium 12.

[0025] The output unit 5 outputs the individual number and pass / fail information corresponding to the individual number (step S18).

[0026] On the other hand, if the inspection target is found to be unsatisfactory in step S15, the pass / fail determination unit 3 transmits information indicating that it is unsatisfactory to the pass / fail information storage unit 4 (step S21). This information is also transmitted to the failure cause analysis unit 6. The failure cause analysis unit 6 reads the calculation result and the pre-calculation fluorescence fingerprint data from the fluorescence fingerprint data calculation unit 1 (step S22). The failure cause analysis unit 6 then estimates the excitation and fluorescence wavelengths that form the basis of the failure (step S23).

[0027] The output unit 5 outputs the excitation and fluorescence wavelengths that form the basis for the failure cause estimated by the failure cause analysis unit 6 (step S24). The output unit 5 may also transmit the excitation and fluorescence wavelength data to an external recording medium 12 for storage.

[0028] Figure 3 is a configuration diagram showing the hardware of the manufacturing inspection system in this embodiment.

[0029] In this figure, the manufacturing inspection system 300 (computer system) is implemented using a computer and includes a processor 31 (Central Processing Unit (CPU), etc.), memory 32 (Random Access Memory (RAM), etc.), storage device 33 (Read Only Memory (ROM), etc.), input device 34, output device 35, and interface 36 (communication device, etc.). Externally installed input devices 321 (keyboard, touch panel, etc.) and display devices 322 (display, etc.) are connected to the interface 36 of the manufacturing inspection system 300. An externally installed spectrofluorometer 11 and external recording medium 12 are also connected to the interface 36.

[0030] The processor 31 is a unit that performs various calculations. The processor 31 performs various processes by executing programs and the like that loaded from the storage device 33 into the memory 32.

[0031] Here, the program is, for example, an application program that can be executed on an OS (Operating System) program. In this embodiment, the program is composed of multiple modules for each function, but it may also be implemented with multiple independent programs for each function.

[0032] Furthermore, the program may be installed in memory 32 from an external recording medium 12 via interface 36, for example. The communication path between interface 36 and an externally installed spectrofluorometer 11, etc., can be implemented via a network such as a LAN (Local Area Network) or the Internet, and may be wired or wireless. The external recording medium 12 may be a portable storage medium such as an SSD (Solid State Drive) using flash memory, an HDD (Hard Disk Drive), or a CD-ROM, or it may utilize cloud services.

[0033] The external recording medium 12 is configured to store design information, manufacturing information, individual identification numbers, etc., and to output them upon request from the processor 31, etc. The external recording medium 12 also receives and stores pass / fail information, the reasons for failure, etc., obtained from calculations by the processor 31, etc.

[0034] Here, we will explain the fluorescence fingerprint data that is used for pass / fail determination.

[0035] Fluorescent fingerprint data is acquired from the spectrofluorometer 11 shown in Figure 1. When excitation light is irradiated onto the resin material to be inspected, the electrons constituting the molecules of the resin material are excited. Subsequently, when the electrons in the excited state return to the ground state, fluorescence may be emitted.

[0036] For example, resin materials often contain additives such as antioxidants and flame retardants. Molecules of antioxidants and flame retardants may be aromatic compounds represented by benzene rings, or they may have π-conjugated electrons with alternating multiple and single bonds. In this case, when irradiated with excitation light, they emit fluorescence. Different excitation wavelengths result in different emission wavelengths. The fluorescence intensity also differs. Resin materials have the characteristic that the distribution of fluorescence wavelengths relative to the excitation wavelength and the fluorescence intensity differ depending on the type and content of the additives. This characteristic can be used to identify the resin material to be inspected. Then, using predetermined acceptance / rejection criteria, it can be determined whether or not the resin material is applicable to the product. In this specification, data that three-dimensionally represents the fluorescence wavelength and fluorescence intensity relative to the excitation wavelength is called "fluorescence fingerprint data." Note that "fluorescence fingerprint data" is not limited to any data that represents the fluorescence wavelength and fluorescence intensity relative to the excitation wavelength. Fluorescence fingerprint data can also be referred to as fluorescence data, fluorescence spectroscopy data, fluorescence intensity data, etc.

[0037] The fluorescence obtained in this way can be detected without being affected by the black pigment, even in recycled materials that have been blackened in appearance due to the addition of black pigments.

[0038] Below, we will explain the inspection targets (1) and (2) using polypropylene resin (PP) as an example of a resin material.

[0039] Figure 4A is a graph showing the fluorescence fingerprint data of the subject (1) being examined. The horizontal axis represents the fluorescence wavelength, and the vertical axis represents the excitation wavelength. The fluorescence intensity is represented by contour lines.

[0040] In this figure, the fluorescence intensity is high in the region where the fluorescence wavelength is 250-350 nm and the excitation wavelength is 250-300 nm. This indicates the fluorescence of additives contained in the virgin material. Therefore, the material being tested (1) is identified as virgin material or a mixed material with a very high proportion of virgin material.

[0041] Note that the graph in this figure is also called an excitation-fluorescence matrix (EEM). The EEM is a three-dimensional plot of the excitation wavelength, the fluorescence wavelength generated thereby, and the intensity of that fluorescence.

[0042] Figure 4B is a graph showing the fluorescence fingerprint data of the inspection target (2).

[0043] In this figure, unlike Figure 4A, fluorescence indicating an additive is not obtained. Therefore, the inspection target (2) is determined to be a recycled material or a mixed material with a very high ratio of recycled materials.

[0044] The reason why fluorescence indicating an additive is not obtained even when the recycled material is irradiated with excitation light is thought to be that antioxidants and flame retardants, which are additives contained in the used virgin material that is the raw material of the recycled material, have become deactivated due to aging deterioration or heat treatment.

[0045] Note that in order to enable each recycled material to be discriminated by fluorescence fingerprint data, different types and ratios of additives may be added to each recycled material when manufacturing the recycled material.

[0046] Figure 5A is a table showing the principal component scores for the inspection target (1) and the inspection target (2). Here, the principal component score is obtained by performing preprocessing such as normalization, smoothing, and differentiation on the fluorescence fingerprint data and calculating the value of the principal component by a predetermined analysis. The principal component score shown in this figure is the value of the fluorescence intensity at a predetermined coordinate indicating the respective characteristics of the virgin material and the recycled material in the graphs of Figures 4A and 4B.

[0047] Figure 5B is a graph showing the distribution of the inspection target (1) and the inspection target (2) with respect to two principal components. The first principal component is taken on the horizontal axis and the second principal component is taken on the vertical axis. In the figure, the region above the boundary line indicated by the broken line represents the virgin material region, and the region below represents the recycled material region.

[0048] From the results shown in this figure, it can be seen that the inspection target (1) is a virgin material and the inspection target (2) is a recycled material.

[0049] Thus, by analyzing the principal components, it is possible to determine whether the resin material is a virgin material or a recycled material.

[0050] Figure 5C is a graph showing the relationship between the fluorescence fingerprint data after pretreatment and the proportion of recycled material contained in the resin material. The horizontal axis represents the principal component analysis values, which are the fluorescence fingerprint data after pretreatment, and the vertical axis represents the recycled material content, which is the composition ratio.

[0051] The regression equation shown in this figure can be expressed by the following linear function.

[0052] Y = b ０ +Σ(b) ｉ ×x ｉ ) In the equation, Y is the objective function, b ｉ x is the regression coefficient. ｉ b is the explanatory variable. ０ It is a constant.

[0053] In this figure, for the sake of simplification and representation as a two-dimensional graph, the value of i is fixed and the principal component analysis values ​​are treated as one type. In reality, i may be an integer greater than or equal to 2, and may be an n-th degree function such as a quadratic or cubic function. Furthermore, it may be an exponential function, a logarithmic function, or a (1 / n)th degree function such as a square root or cube root.

[0054] This figure shows that, by using a regression equation, it can be determined that the recycled material content (e.g., by mass) is 20% for subject (1) and 70% for subject (2).

[0055] In this way, the recycled material content can be estimated by using principal component analysis values, etc.

[0056] Figure 6 shows an example of the screen of the display device in this embodiment.

[0057] This figure shows that the following are displayed on a single screen: a button to load fluorescent fingerprint data, a graph display of the resulting fluorescent fingerprint data, a setting for pass / fail criteria, a display of the pass / fail criteria, a button to output the pass / fail result, a display of the pass / fail result, and a display of candidate wavelengths that caused failure. Here, the loading of fluorescent fingerprint data can be done manually or automatically. The setting of the pass / fail criteria can also be done automatically, but the operator may be allowed to set or adjust the pass / fail range. The output of the pass / fail result can be done in batch processing or automatically.

[0058] Here, the candidate wavelengths that may cause failure can be rephrased as the excitation and fluorescence wavelengths that form the basis for the failure.

[0059] Furthermore, the items displayed on a single screen are not limited to those shown in this diagram; some of these items may be displayed.

[0060] By displaying raw data such as fluorescent fingerprint data, pass / fail criteria, pass / fail results, and reasons for failure in this manner, it is possible to support operators in making decisions regarding the pass / fail status of products.

[0061] In describing Example 2, only the differences from Example 1 will be explained, and similar configurations will not be described.

[0062] Figure 7 is a diagram showing the configuration of the manufacturing inspection system in this embodiment.

[0063] The manufacturing inspection system 700 shown in this figure does not have the failure cause analysis unit 6 shown in Figure 1. The other configurations are the same as in Figure 1.

[0064] Figure 8 is a flowchart showing the manufacturing inspection method in this embodiment.

[0065] Unlike Figure 2, in this figure, the process proceeds to steps S16 to S18 without branching in step S15.

[0066] In describing Example 3, only the differences from Example 1 will be explained, and similar configurations will be omitted from the description.

[0067] Figure 9 is a diagram showing the configuration of the manufacturing inspection system in this embodiment.

[0068] In the manufacturing inspection system 900 shown in this figure, the pass / fail information storage unit 4 is configured not to receive external input of the individual identification number of the inspection target shown in Figure 1. The other configurations are the same as in Figure 1.

[0069] Figure 10 is a flowchart showing the manufacturing inspection method in this embodiment.

[0070] In this figure, unlike in Figure 2, the process proceeds to step S108 without performing step S17 as in Figure 2 after step S16. In step S108, pass / fail information is output from the output unit 5.

[0071] 1: Fluorescent fingerprint data calculation unit, 2: Acceptance criteria storage unit, 3: Pass / fail judgment unit, 4: Pass / fail information storage unit, 5: Output unit, 6: Failure cause analysis unit, 11: Spectrofluorometer, 12: External recording medium, 31: Processor, 32: Memory, 33: Storage device, 34: Input device, 35: Output device, 36: Interface, 100, 300, 900: Manufacturing inspection system, 321: Input device, 322: Display device.

Claims

1. A manufacturing inspection system that measures an inspection target, which is a product, by fluorescent fingerprint measurement in the product manufacturing process and outputs a manufacturing inspection result, comprising: a spectrofluorometer for measuring the fluorescent fingerprint data of the inspection target; a fluorescent fingerprint data calculation unit for analyzing the measured fluorescent fingerprint data using a predetermined calculation method; an acceptance criteria storage unit for storing acceptance criteria for the product specifications formulated from one or more design specification ranges defined in the product design information and manufacturing specification ranges defined in the manufacturing information; and an acceptance / rejection determination unit for determining acceptance or rejection by comparing the results analyzed by the fluorescent fingerprint data calculation unit with the acceptance criteria.

2. The manufacturing inspection system according to claim 1, further comprising a pass / fail information storage unit that stores the pass / fail judgment result of the pass / fail judgment unit and the individual number of the inspection target obtained from an external source in association with each other.

3. The manufacturing inspection system according to claim 2, further comprising an output unit that outputs certification data including the individual number stored in the pass / fail information storage unit and the pass / fail judgment result corresponding to the individual number.

4. If the pass / fail determination unit determines that the product is unsatisfactory, the output unit displays the excitation and fluorescence wavelengths that form the basis for the cause of failure before analysis using the calculation method, as described in claim 3.

5. The manufacturing inspection system according to claim 1, wherein the manufacturing inspection result includes the component to be inspected and the proportion of the component.

6. The manufacturing inspection system according to claim 5, wherein the object to be inspected is a resin material, and the components include recycled material and virgin material.

7. A manufacturing inspection method for measuring an inspection target, which is a product, by fluorescent fingerprint measurement in the product manufacturing process and outputting a manufacturing inspection result, wherein a fluorescent fingerprint data calculation unit analyzes the fluorescent fingerprint data of the inspection target measured by a spectrofluorometer using a predetermined calculation method; an acceptance criterion storage unit stores acceptance criteria for the product specifications formulated from one or more design specification ranges defined in the product design information and manufacturing specification ranges defined in the manufacturing information; and a pass / fail determination unit compares the results analyzed by the fluorescent fingerprint data calculation unit with the acceptance criteria to determine whether the product is pass or fail.

8. The manufacturing inspection method according to claim 7, wherein the pass / fail information storage unit stores the pass / fail judgment result of the pass / fail judgment unit and the individual number of the inspection target obtained from an external source in association with each other.

9. The manufacturing inspection method according to claim 8, wherein the output unit outputs certification data including the individual number stored in the pass / fail information storage unit and the pass / fail judgment result corresponding to the individual number.

10. The manufacturing inspection method according to claim 9, wherein if the pass / fail determination unit determines that the product is unsatisfactory, the output unit displays the excitation and fluorescence wavelengths that form the basis for the cause of failure before analysis using the calculation method.

11. The manufacturing inspection method according to claim 7, wherein the manufacturing inspection result includes the component to be inspected and the proportion of the component.

12. The manufacturing inspection method according to claim 11, wherein the object to be inspected is a resin material, and the components include recycled material and virgin material.

13. A manufacturing inspection apparatus that measures an inspection target, which is a product, by fluorescence fingerprint measurement in the manufacturing process of a product and outputs a manufacturing inspection result, comprising: a fluorescence fingerprint data calculation unit that analyzes the fluorescence fingerprint data of the inspection target measured by a spectrofluorometer using a predetermined calculation method; an acceptance criteria storage unit that stores acceptance criteria for the specifications of the product, which are formulated from one or more design specification ranges defined in the product's design information and manufacturing specification ranges defined in the manufacturing information; and an acceptance / rejection determination unit that determines acceptance or rejection by comparing the results analyzed by the fluorescence fingerprint data calculation unit with the acceptance criteria.

14. The manufacturing inspection apparatus according to claim 13, further comprising a pass / fail information storage unit that stores in association the pass / fail judgment result of the pass / fail judgment unit with the individual number of the inspection target obtained from an external source.

15. The manufacturing inspection apparatus according to claim 14, further comprising an output unit that outputs certification data including the individual number stored in the pass / fail information storage unit and the pass / fail judgment result corresponding to the individual number.

16. The manufacturing inspection apparatus according to claim 15, wherein, if the pass / fail determination unit determines that the product is unsuccessful, the output unit displays the excitation and fluorescence wavelengths that form the basis for the cause of failure before analysis using the calculation method.

17. A program for causing a computer to execute the following manufacturing inspection procedure, which involves measuring an object to be inspected, which is the product, by fluorescent fingerprint measurement in the product manufacturing process and outputting a manufacturing inspection result, wherein the procedure includes: a fluorescent fingerprint data calculation unit analyzing the fluorescent fingerprint data of the object to be inspected measured by a spectrofluorometer using a predetermined calculation method; a pass / fail storage unit storing pass / fail criteria for the product specifications formulated from one or more design specification ranges defined in the product design information and manufacturing specification ranges defined in the manufacturing information; and a pass / fail determination unit comparing the results analyzed by the fluorescent fingerprint data calculation unit with the pass / fail criteria to determine whether the product is pass or fail.

18. The program according to claim 17, further comprising the pass / fail information storage unit storing the pass / fail judgment result of the pass / fail judgment unit in association with the individual number of the inspected object obtained from an external source.

19. The program according to claim 18, wherein the procedure further includes the output unit outputting certification data including the individual number stored in the pass / fail information storage unit and the pass / fail judgment result corresponding to the individual number.

20. The program according to claim 19, wherein, if the pass / fail determination unit determines that the device is unsuccessful, the output unit further includes displaying the excitation and fluorescence wavelengths that form the basis for the cause of failure before analysis using the calculation method.

Images