JP3800244B2 - Quality control apparatus and control method therefor, quality control program, and recording medium recording the program - Google Patents

Description

  The present invention relates to a quality control device that controls a manufacturing process, a control method thereof, a quality control program, and a recording medium on which the program is recorded in order to manufacture a product having a predetermined quality.

  Conventionally, efforts have been made to improve the quality of products, and various quality control methods have been proposed.

  For example, in order to realize traceability, a method of managing an intermediate product product by attaching an identification code such as a barcode to an intermediate product or a base on which the intermediate product is placed has been proposed. Further, in order to realize traceability, a method of identifying intermediate products in the order of intermediate products input into each manufacturing process has been proposed.

  In addition, a method has been proposed in which a control factor for obtaining an optimum product characteristic is specified by completely grasping the causal relationship between the control factor and the product characteristic. In addition, a method for maintaining product characteristics by fixing all states of control factors has been proposed (see, for example, Patent Documents 2 and 3).

In addition, a method has been proposed in which indicators for each product characteristic are provided and comprehensive evaluation is performed based on these indicators. Also, a method for adjusting the control parameters in the manufacturing process when the quality of the product characteristics is determined in the final manufacturing process and determined as a defective product has been proposed (see, for example, Patent Document 1).
JP 7-141005 A (published on June 2, 1995) Japanese Patent Laid-Open No. 6-110504 (published on April 22, 1994) Japanese Patent Laid-Open No. 4-188301 (published July 6, 1992) Genichi Taguchi, Masataka Yoshizawa, “Quality Engineering Course 1 / Quality Engineering at the Development and Design Stage”, Japanese Standards Association, 1988

  However, there is a problem that the manufacturing cost increases in the method of managing by attaching an identification code. In addition, defective products are often extracted, reworked, and re-introduced in the manufacturing process. For this reason, it is not always possible to correctly identify the intermediate product by the above method of identifying the intermediate product in the order of the intermediate product.

  In addition, since there are many control factors, a large number of verification experiments are required to fully understand the causal relationship between control factors and product characteristics. Also, the more complicated the causal relationship, the more difficult it is to grasp completely. Moreover, even if all the states of the control factors are fixed, the product characteristics fluctuate due to temporal variations of factors that cannot be observed.

  Further, in the method of performing comprehensive evaluation from each product characteristic index, it is difficult to perform comprehensive evaluation because product characteristics are often in a trade-off relationship. In addition, in the method of determining whether the product characteristics are good or bad in the final manufacturing process, since defective products are continuously manufactured until the final process, useless manufacturing time and materials increase. In addition, the interval between control feedback loops becomes longer.

  The present invention has been made in view of the above problems, and an object thereof is to provide a quality control device that associates various data collected in different processes at a manufacturing site where traceability is difficult. It is in.

In order to solve the above problems, a quality control device according to the present invention is a quality control device that controls a manufacturing process in order to manufacture a product having a predetermined quality, and is an intermediate product measured by an intermediate characteristic inspector. Storage means for collecting and storing intermediate characteristic data and final characteristic data of the final product measured by the final characteristic inspector, data distribution of the final characteristic data stored by the storage means, and setting in the final characteristic inspector A standard value generating means for generating an intermediate characteristic standard value as a criterion for the intermediate characteristic inspector to eliminate defective intermediate products based on the final characteristic standard value, and the standard value generating means Standard value setting control means for controlling the intermediate characteristic inspector so as to set the standard value, and the storage means measures the intermediate characteristic data and the final characteristic data at the time of measurement or collection. And the correspondence between the intermediate characteristic data and the final characteristic data by taking into account the dead time generated between the intermediate characteristic inspector and the final characteristic inspector at the measured time or the collected time. It is further characterized by further comprising an association means for performing attachment.

  Here, the intermediate characteristic inspection device inspects the intermediate characteristic of the intermediate product being manufactured, and determines the quality of the intermediate product based on the intermediate characteristic standard value. By using the intermediate characteristic inspector, defective products are removed during the manufacturing process, so that it is possible to eliminate the time for manufacturing useless defective products and shorten the interval between control feedback loops.

  According to the above configuration, since the standard value generation means changes the intermediate characteristic standard value based on the final characteristic of the finished product, the intermediate characteristic standard value is automatically changed even if various parameters in the manufacturing process fluctuate. Can be adjusted.

  Preferably, the standard value generation means generates the intermediate characteristic standard value based on the difference between the average value of the data distribution of the final characteristic data and the final characteristic standard value. Further, the standard value generation means is configured to determine the average value of the data distribution of the intermediate characteristic data stored by the storage means and the intermediate characteristic standard for the difference between the average value of the data distribution of the final characteristic data and the final characteristic specification value. Preferably, the intermediate characteristic standard value is generated so that the ratio of the difference from the value is the ratio of the variation value of the data distribution of the intermediate characteristic data to the variation value of the data distribution of the final characteristic data.

  Here, the variation value of the data distribution indicates the degree of variation of the data distribution. For example, the standard deviation, the variance, the half width, and the like are increased.

  The quality control apparatus according to the present invention is a quality control apparatus that controls a manufacturing process in which it is difficult to grasp a causal relationship between a control factor and a product characteristic in order to manufacture a product having a predetermined quality. Storage means for collecting and storing intermediate characteristic data of the intermediate product measured by the intermediate characteristic inspector and final characteristic data of the finished product measured by the final characteristic inspector, and data of the final characteristic data stored by the storage means Based on the distribution and the final characteristic standard value set in the final characteristic inspector, the standard value generation for generating an intermediate characteristic standard value that is a criterion for the intermediate characteristic inspector to eliminate defective intermediate products Means and a standard value setting control means for controlling the intermediate characteristic inspector so as to set the standard value generated by the standard value generating means. The standard value generating means includes a data distribution of the final characteristic data. Average value X And the standard deviation σ is obtained, and dua = du × (σa) from the distance du between the average value X￣, the upper limit standard value Su of the final characteristic data, and the standard deviation σa of the data distribution of the intermediate characteristic data. / Σ) = (Su−X∥) × (σa / σ) to obtain a distance dua between the average value Xa￣ of the data distribution of the intermediate characteristic data and the upper limit specification value of the intermediate characteristic data, and the distance a first process for obtaining the upper limit standard value Sua of the intermediate characteristic data from dua and the average value Xa￣ by Sua = Xa￣ + dua = Xa￣ + (Su−X￣) × (σa / σ); And, from the distance dl between the average value X￣, the lower limit standard value Sl of the final characteristic data, and the standard deviation σa of the data distribution of the intermediate characteristic data, dla = dl × (σa / σ) = ( X￣−Sl) × (σa / σ) The distance dla between the average value Xa￣ of the data distribution of the data and the lower limit specification value of the intermediate characteristic data is obtained, and Sla = Xa￣−dla = Xa￣− from the distance dla and the average value Xa￣. It is characterized in that at least one of the second processes for obtaining the lower limit standard value Sla of the intermediate characteristic data is executed by (X￣−Sl) × (σa / σ).

  Further, it is desirable that the standard value generation means further generates the intermediate characteristic standard value based on a safety factor. In this case, the intermediate product can be efficiently eliminated based on the intermediate characteristic standard value. The safety factor may be set based on an empirical rule, but the disposal cost when the finished product is excluded based on the final characteristic specification value and the intermediate characteristic value based on the intermediate characteristic specification value. It is desirable to calculate based on the disposal cost when products are excluded.

  The storage means stores the intermediate characteristic data and the final characteristic data together with the measured time or the collected time. At the measured time or the collected time, the intermediate characteristic inspector and the final characteristic inspection are stored. It is desirable to further include association means for associating the intermediate characteristic data and the final characteristic data by considering the dead time generated between the containers. In this case, since the collected intermediate characteristic data and final characteristic data are associated with the measured time or the collected time, the association unit uses the measured time or the collected time as the intermediate characteristic inspector and final characteristic inspector. By roughly associating with the dead time generated between them, it is possible to associate the intermediate characteristic data with the final characteristic data.

  Further, the standard value generation means uses the intermediate characteristic data and the final characteristic data collected by the storage means after a predetermined time has elapsed since the standard value setting control means controlled the intermediate characteristic tester. It is desirable to generate the next standard value. In general, when changing the setting value of a device, a slight time lag occurs until the change is completed. Therefore, in the above case, since the data in the middle of the change is not used, more accurate control is possible.

The control method of the quality control apparatus according to the present invention is a quality control apparatus that controls a manufacturing process in which it is difficult to grasp the causal relationship between a control factor and a product characteristic in order to manufacture a product having a predetermined quality. The quality control device is a control method for storing and storing the intermediate characteristic data of the intermediate product measured by the intermediate characteristic inspector and the final characteristic data of the finished product measured by the final characteristic inspector. And the intermediate characteristic inspector eliminates defective intermediate products based on the data distribution of the final characteristic data stored in the storage means and the final characteristic standard value set in the final characteristic inspector. A standard value generation step for generating an intermediate characteristic standard value, which is a determination criterion of the standard value, and a standard value setting control step for controlling the intermediate characteristic tester so as to set the standard value generated in the standard value generation step The standard value generation step includes a step of obtaining an average value X￣ and standard deviation σ of the data distribution of the final characteristic data, and further, the average value X￣ and the upper limit standard value Su of the final characteristic data. From the distance du and the standard deviation σa of the data distribution of the intermediate characteristic data, dua = du × (σa / σ) = (Su−X￣) × (σa / σ). A distance dua between the average value Xa￣ and the upper limit standard value of the intermediate characteristic data is obtained, and Sua = Xa￣ + dua = Xa￣ + (Su−X￣) from the distance dua and the average value Xa￣. A first processing step for obtaining an upper limit standard value Sua of the intermediate characteristic data by x (σa / σ), and a distance dl between the average value X￣ and the lower limit standard value Sl of the final characteristic data; The above intermediate characteristic data From the standard deviation σa of the distribution, by means of dla = dl × (σa / σ) = (X∥−S1) × (σa / σ), the average value Xa￣ of the data distribution of the intermediate characteristic data and the intermediate characteristic data The distance dla from the lower limit standard value is obtained, and from the distance dla and the average value XaＳ, Sla = Xa￣−dla = Xa￣− (X∥−Sl) × (σa / σ) It includes at least one step of a second processing step for obtaining a lower limit standard value Sla of the intermediate characteristic data.

A control method for a quality control apparatus according to the present invention is a control method for a quality control apparatus that controls a manufacturing process in order to manufacture a product having a predetermined quality. The quality control apparatus includes an intermediate characteristic inspector. A storage means for storing the intermediate characteristic data of the measured intermediate product and the final characteristic data of the finished product measured by the final characteristic inspector together with the measured time or the collected time is provided, and the final stored by the storage means Based on the data distribution of the characteristic data and the final characteristic standard value set in the final characteristic inspector, the intermediate characteristic inspector generates an intermediate characteristic standard value that is a criterion for eliminating defective intermediate products. A standard value generation step, a standard value setting control step for controlling the intermediate characteristic inspector so as to set the standard value generated in the standard value generation step, and a measured time or An association step for associating the intermediate characteristic data with the final characteristic data by considering a dead time generated between the intermediate characteristic inspector and the final characteristic inspector at the time. It is said.

  According to the above method, since the intermediate characteristic standard value is changed based on the final characteristic of the finished product in the standard value generation step, the intermediate characteristic standard value is automatically changed even if various parameters in the manufacturing process fluctuate. Can be adjusted.

  Each means in the quality control apparatus can be executed on a computer by a quality control program. Further, the quality control program can be executed on an arbitrary computer by storing the quality control program in a computer-readable recording medium.

  As described above, in the quality control device according to the present invention, the standard value generation unit changes the intermediate characteristic standard value based on the final characteristic of the finished product, so even if various parameters in the manufacturing process fluctuate, There is an effect that the characteristic standard value can be automatically adjusted.

  Hereinafter, an embodiment of the present invention will be described with reference to FIGS. FIG. 1 shows a schematic configuration of a quality control system for controlling the quality of a product manufactured from a certain material. As shown in the figure, the quality control system 1 includes a control target device (measuring device) 3 and 4 to be controlled and an intermediate property tester (measuring device) 5 for inspecting an intermediate property that is a property of an intermediate product being manufactured. Collecting data from the final characteristic tester (measuring device) 6 for inspecting the final characteristics of the manufactured product and various devices 3 to 6, and controlling the control target devices 3 and 4 based on the collected data And a quality control device 10 that changes the standard value in the intermediate characteristic tester 5.

  In general, the product 7 is manufactured through a number of processing steps on the workpiece 2, but in the present embodiment, it is manufactured as follows in order to facilitate understanding of the present invention. That is, the first and second controlled devices 3 and 4 sequentially process the workpiece 2, and then the intermediate characteristic inspection device 5 performs intermediate characteristic inspection processing, and the other devices perform various processes. After performing the above, it is assumed that the final characteristic tester 6 performs final characteristic inspection processing.

  Each of the control target devices 3 and 4 acquires a set value of a control value from the quality control device 10 and operates based on the acquired set value. In addition, each of the control target devices 3 and 4 actually measures the control value and transmits the measured value to the quality control apparatus 10. Here, for example, if the device to be controlled is a heater (heater), the control value is a numerical value indicating the temperature, such as a temperature, a voltage corresponding to the temperature, or a resistance value.

  The intermediate characteristic tester 5 acquires a standard value from the quality control device 10 and inspects the intermediate characteristic of the workpiece 2 (intermediate product) based on the acquired standard value. The intermediate characteristic inspector 5 transmits the inspected intermediate characteristic to the quality control apparatus 10.

  The intermediate product determined by the intermediate characteristic inspector 5 to be out of specification is usually excluded immediately after the intermediate characteristic is inspected. However, if there is a possibility that the second controlled device 4 can easily be within the standard by performing the process again for the intermediate product determined to be out of the standard, the intermediate product is controlled by the second control. It can be re-entered into the processing steps performed by the target device 4. Further, when it is difficult to exclude the intermediate characteristic immediately after the inspection due to the system configuration or the device structure, it may be excluded in a later process.

  The final characteristic inspector 6 inspects the final characteristic of the finished product based on a predetermined standard value. At this time, the finished product determined to be out of the standard is discarded or repaired or disassembled and reused, and the finished product determined to be within the standard is shipped as the product 7. Further, the final characteristic inspector 6 transmits the inspected final characteristic to the quality control apparatus 10.

  Normally, a time lag occurs at a time when each device 3 to 6 processes a certain work 2. This time lag is referred to as “dead time”. In the case shown in FIG. 1, a dead time Td1 occurs between the time when the process is started at the first control target device 3 and the time when the process is started at the second control target device 4. The dead time Td2 is generated between the time when the process is started at the second controlled object device 4 and the time when the process is started at the intermediate characteristic tester 5, and the time when the process is started at the intermediate characteristic tester 5 And a dead time Td3 occurs between the time when the final characteristic tester 6 starts processing.

  The control target device, the intermediate characteristic tester, and the final characteristic tester may be one or more. Moreover, not all devices related to manufacturing are necessarily controlled devices. One device may include a plurality of devices to be controlled.

  As shown in FIG. 1, the quality control apparatus 10 includes a data storage unit (storage unit) 11, a scheduler (association unit) 12, a standard value generation unit (standard value generation unit, standard value setting control unit) 13, and an influence. It is configured to include a degree generation unit (influence degree generation unit) 14, an influence degree storage DB (database) (influence degree storage unit) 15, an estimation unit 16, and a control command unit (control change unit) 17. The control device 10 is configured by a PC (Personal Computer) based computer, for example.

  The data storage unit 11 sequentially acquires the measurement values of the control values from each of the control target devices 3 and 4, stores the acquired measurement values and the acquisition time (collected time) as control data, and from the intermediate characteristic tester 5 Intermediate characteristics are acquired sequentially, the acquired intermediate characteristics and acquisition time are stored as intermediate characteristic data, and the final characteristics are sequentially acquired from the final characteristic inspector 6, and the acquired final characteristics and acquisition time are stored as final characteristic data. To do. Further, the data storage unit 11 notifies the scheduler 12 that data acquisition has started. In addition, you may memorize | store the time which each apparatus 3-6 measured instead of acquisition time.

  The scheduler 12 associates the control data, the intermediate characteristic data, and the final characteristic data stored in the data storage unit 11 with each other that is estimated to be the same work 2, and determines the acquisition time of each data and the dead time. To use. Further, the scheduler 12 determines the timing for changing the set value of the control value for controlling the control target devices 3 and 4 and the timing for changing the standard value of the intermediate characteristic of the intermediate characteristic tester 5.

  The association may be performed in units of individual workpieces 2, or may be performed in units of a predetermined number of workpieces 2 or workpieces 2 processed in a predetermined time. Hereinafter, a series of data included in the predetermined number or predetermined time is referred to as a set.

  The standard value generation unit 13 reads the set of intermediate characteristic data and the set of final characteristic data associated with each other by the scheduler 12 from the data storage unit 11, and generates a standard value of the intermediate characteristic based on the read set. In addition, the standard value generation unit 13 transmits the generated standard value to the intermediate characteristic tester 5 based on the timing determined by the scheduler 12. The standard value generation unit 13 may generate the standard value only when the standard value of the intermediate characteristic of the intermediate characteristic tester 5 should be changed. Further, the standard value generation unit 13 generates an estimated value of the standard value of the intermediate characteristic and transmits the generated estimated value to the estimation unit 16.

  The influence generation unit 14 generates an influence degree in response to a request from the estimation unit 16. Here, the degree of influence indicates how the intermediate characteristic varies when the control value is changed. Further, the influence degree generation unit 14 transmits the generated influence degree to the estimation unit 16 and stores it in the influence degree storage DB 15. Specifically, the influence generation unit 14 reads the set of control data and the set of intermediate characteristic data associated with each other by the scheduler 12 from the data storage unit 11, and based on the read set, the average value of the control data The average value of the intermediate characteristic data and / or the amount of variation of the variance with respect to the change amount of is calculated, and the degree of influence is generated by normalization. An arbitrary index indicating the degree of variation can be used instead of the variance, and for example, a standard deviation or a half width can be used.

  The estimation unit 16 reads the set of intermediate characteristic data associated by the scheduler 12 from the data storage unit 11, and reads the average value of the intermediate characteristic and / or the variance variation amount from the read set from the influence storage DB 15. Estimate using the degree of influence. In addition, the estimation unit 16 estimates the number of defects when the distribution of the intermediate characteristic is varied with the estimated value of the amount of variation, and evaluates the sum by the number of individual defects. Then, a change amount of the set value of the control value when the number of defects is minimized is obtained and transmitted to the control command unit 17.

  The control command unit 17 changes the set value of the control value based on the change amount of the set value of the control value obtained by the estimation unit 16. Further, the control command unit 17 transmits the changed setting value as a control command to the control target devices 3 and 4 based on the timing determined by the scheduler 12. Thereby, the set value of the control value in the control target devices 3 and 4 is changed.

  The processing operation in the quality control system 1 having the above configuration will be described with reference to FIG. FIG. 2 shows an outline of the processing operation in the quality control system 1. Although not shown in the figure, when the quality control apparatus 10 starts operation, various data are sequentially acquired from each of the devices 3 to 6 and stored continuously.

  First, the scheduler 12 acquires various data from the data storage unit 11 (step S10. Hereinafter, it may be simply referred to as “S10”. The same applies to other steps), and the intermediate characteristic data and the final data are obtained. It is determined whether or not the association with the characteristic data is possible (S11).

  When the above association is possible (YES in S11), the standard generation unit 13 generates a standard value of the intermediate characteristic using the set of associated intermediate characteristic data and the set of final characteristic data. The standard value of the intermediate characteristic is corrected by transmitting it to the intermediate characteristic tester 5 (S12). On the other hand, if the association is impossible (NO in S11), the process proceeds to the next step (S13) without correcting the standard value of the intermediate characteristic.

  Next, the scheduler 12 determines whether or not the control data and the intermediate characteristic data can be associated (S13). If the association is impossible (NO in S13), the process returns to step S10 and the above processing operation is repeated.

  On the other hand, if the above association is possible (YES in S13), it is determined whether or not the difference between the measured value of the intermediate characteristic data and the estimated value of the intermediate characteristic data estimated last time is equal to or greater than a predetermined value ( S14). If the difference between the measured value and the estimated value is greater than or equal to the predetermined value (YES in S14), the impact generation unit 14 generates the impact and updates the impact stored in the impact storage DB 15. (S15). On the other hand, if the difference between the measured value and the estimated value is less than the predetermined value (NO in S14), the process proceeds to the next step (S16) without updating the influence level.

  Next, the estimation unit 16 extracts a set of intermediate characteristic data from the data storage unit 11, acquires an influence degree from the influence degree storage DB 15, and uses the extracted intermediate characteristic data set and the acquired influence degree. Then, it is estimated how the data distribution of the intermediate characteristic data fluctuates when the set value of the control value is changed (S16), and the number of defects is estimated (S17).

  Next, the estimation unit 16 determines the estimated value (distribution variation amount) of the intermediate characteristic data so that the number of defects is minimized (S18), and also determines the change amount of the set value of the corresponding control value. . Next, the control command unit 17 determines the set value of the control value based on the determined change amount, and transmits the determined set value of the control value to the control target devices 3 and 4 to set the control value. A change instruction is issued (S19). Then, it returns to step S10 and repeats said process operation.

  According to the above configuration, since the data stored in the data storage unit 11 is associated with the measured time or the collected time, the scheduler 12 generates the measured time or the collected time between the measuring devices. By roughly associating with the dead time, the measurement data of the measuring devices can be associated with each other. Further, the dead time is used only by the scheduler 12, and even if the dead time is changed, the data stored in the data storage unit 11 is not affected. Therefore, it is possible to easily change the dead time.

  In addition, since the standard value generation unit 13 changes the standard value of the intermediate characteristic based on the final characteristic of the finished product, the standard value of the intermediate characteristic is automatically adjusted even if various parameters in the manufacturing process fluctuate. be able to.

  Further, the estimation unit 16 estimates the data distribution variation of the intermediate characteristic data when the set value of the control value is changed, estimates the number of defects to be generated, and sets the control value that minimizes the estimated number of defects. The amount of change of the set value is determined, and the set value of the control value of the control target device is changed. As a result, the intermediate product is inspected during the manufacturing, and the control of the control target device is changed so that the number of defective intermediate products is reduced. Therefore, the interval of the control feedback loop can be shortened. In addition, by estimating the variation in the data distribution, it is possible to take into account data variations, and thus it is possible to more accurately estimate the number of defects generated.

  Further, since the correspondence relationship between the variation amount of the average value of the control data and the variation amount of the data distribution of the intermediate characteristic data is known from the influence degree stored in the influence degree storage DB 15, the variation of the data distribution of the intermediate characteristic data is known. The amount can be estimated more accurately, and the number of defects generated can be estimated more accurately.

  Further, when the estimated value of the intermediate characteristic is far from the actual measurement value, the influence generation unit 14 creates the influence and stores it in the influence storage DB 15, so even if any fluctuation occurs in the manufacturing process, the influence degree Can be corrected automatically. Therefore, the fluctuation amount of the data distribution of the intermediate characteristic data can be estimated more accurately, and a product with appropriate quality can be manufactured.

  The quality control apparatus 10 can be applied to various processes. As an example of an applicable process, there is a sensor element manufacturing process described in JP-A-2002-287803. This manufacturing process includes a molding / grinding step (step 1), a firing step (step 2), a plating step (step 3), an electrode protective layer forming step (step 4), and a trap layer forming step (step 5). , An aging process (process 6), an assembly process (process 7), and a characteristic inspection process (process 8). Molding specific gravity, firing specific gravity, plating film thickness, coating film thickness, coating weight, and aging temperature / additive concentration are acquired as intermediate characteristics from Steps 1 to 6, respectively, and final characteristics are acquired from Step 8 above. In addition, an apparatus used in each process is a control target device.

  Hereafter, each block in the quality control apparatus 10 is demonstrated in detail based on FIGS.

  FIG. 3 shows a schematic configuration of the data storage unit 11. The data storage unit 11 includes a control data storage unit 20 and a control data storage DB 21, an intermediate characteristic data storage unit 22 and an intermediate characteristic data storage DB 23, a final characteristic data storage unit 24, and a final characteristic data storage DB 25. .

  The control data storage unit 20 acquires the set value of the control value instructed to the control target device by the control command unit 17, the measurement value actually measured by the control target device with respect to the set value, and the measurement value. The acquisition time is acquired as control data, and the acquired control data is stored in the control data storage DB 21. In addition, acquisition and storage of control data are performed for each of the control target devices 3 and 4. The set value of the control value may be acquired from the control target devices 3 and 4 or may be acquired from the control command unit 17.

  The intermediate characteristic data storage unit 22 acquires the two intermediate characteristics A and B inspected by the intermediate characteristic inspector 5, and stores the acquired intermediate characteristics A and B and the acquisition time in the intermediate characteristic data storage DB 23 in association with each other. The final characteristic data storage unit 24 acquires the two final characteristics A and B inspected by the final characteristic inspector 6, and associates the acquired final characteristics A and B with the acquisition time and stores them in the final characteristic data storage DB 25. .

  4A and 4B show examples of control data stored in the control data storage DB 21 for the control target devices 3 and 4, respectively. FIGS. 7C and 7D show examples of intermediate characteristic data stored in the intermediate characteristic data storage DB 23 and the final characteristic data storage DB 25, respectively.

  The serial number shown in FIG. 4 is the serial number of the workpiece 2 processed by each device. As described above, since the workpiece 2 may join, be omitted, or reintroduced during manufacture, even if the serial number of each device is the same, it does not necessarily correspond to the same workpiece 2. For example, the workpiece 2 corresponding to the manufacturing number 1 of the first controlled object is not necessarily the same as the workpiece 2 corresponding to the manufacturing number 1 of the second controlled object.

  The intermediate characteristic inspected by the intermediate characteristic inspector 5 is a characteristic of an intermediate product being manufactured, and is inspected before the final characteristic inspector 6 inspects. For this reason, the intermediate characteristic does not always have the same value as the final characteristic. In this embodiment, two types of intermediate characteristics and final characteristics are acquired, but one type may be used, or three or more types may be used.

  Next, details of the scheduler 12 will be described with reference to FIGS. 5 and 6. As described above, the scheduler 12 associates various data stored in the data storage unit 11, determines the timing for extracting the associated data, and controls the control target devices 3 and 4. The timing for changing the set value of the value and the standard value of the intermediate characteristic in the intermediate characteristic inspector 5 is determined. In addition, the scheduler 12 transmits the determined timing to the standard value generation unit 13 and the control command unit 17.

  FIG. 5 shows a schematic configuration of the scheduler 12. As illustrated, the scheduler 12 includes a setting unit 30, a control value extraction timing determination unit 31, an intermediate characteristic extraction timing determination unit 32, a final characteristic extraction timing determination unit 33, a control value change timing determination unit 34, and a standard value change timing. The configuration includes a determination unit 35.

  The setting unit 30 sets dead times Td1 to Td3 generated between devices that acquire data, margin times Tm1 and Tm2, and a corresponding set period Ts, and transmits them to the units 31 to 35. These values may be acquired from the user via an input device such as a keyboard, may be acquired from a storage device, or may be acquired from another device via a communication device.

  Here, the first margin time Tm1 refers to a period from the time when the change of the set value of the control value or the change of the standard value of the intermediate characteristic is instructed to the first acquisition time extracted from various data. The first margin time Tm1 is obtained by removing a time lag from instructing the device to change the set value or the standard value until the change is actually reflected on the work 2, or by removing or re-injecting a defect. It is set to remove the deviation.

  The second margin time Tm2 is a period from the last acquisition time extracted from the intermediate characteristic data to the time when the first control target device 3 is instructed to change the set value of the control value. The second margin time Tm2 is set in consideration of the processing time spent for obtaining the set value to be changed.

  The margin times Tm1 and Tm2 and the set period Ts may be represented by the number of workpieces 2. When the period Ts of the set is represented by the number of works 2, the number of data included in each associated set is constant, so that the accuracy of the statistics of each set can be made substantially coincident. Therefore, each set can be appropriately associated even if the dead time fluctuates, for example, when any of the devices 3 to 6 is temporarily stopped and restarted.

  In addition, the following three methods are mentioned as a method of deriving the values of the dead times Td1 to Td3 and the margin times Tm1 and Tm2. That is, firstly, there is a method derived by calculation from the design value of the quality control system, that is, the time required for processing in each device, and the design value of the moving distance and moving speed when the work moves between the devices. It is done. Secondly, there is a method in which a test operation is performed to measure the processing time and movement time of the workpiece and derive from the actual measurement values. Thirdly, there is a method in which an actual manufacturing operation is performed, and a processing time and a moving time of the workpiece are measured and derived from an actual measurement value.

  In practice, it is desirable to derive these methods by appropriately combining them. For example, the initial setting value before the quality control system is operated is derived by the first method, and after the operation, the fine adjustment is performed by the second or third method.

  The intermediate characteristic inspected by the intermediate characteristic inspector 5 is a characteristic of an intermediate product being manufactured, and is inspected before the final characteristic inspector 6 inspects. For this reason, the intermediate characteristic does not always have the same value as the final characteristic. In this embodiment, two types of intermediate characteristics and final characteristics are acquired, but one type may be used, or three or more types may be used.

  Next, details of the scheduler 12 will be described with reference to FIGS. 5 and 6. As described above, the scheduler 12 associates various data stored in the data storage unit 11, determines the timing for extracting the associated data, and controls the control target devices 3 and 4. The timing for changing the set value of the value and the standard value of the intermediate characteristic in the intermediate characteristic inspector 5 is determined. Further, the scheduler 12 transmits the determined timing to each of the blocks 13, 14, 16, and 17.

  FIG. 5 shows a schematic configuration of the scheduler 12. As illustrated, the scheduler 12 includes a setting unit 30, a control value extraction timing determination unit 31, an intermediate characteristic extraction timing determination unit 32, a final characteristic extraction timing determination unit 33, a control value change timing determination unit 34, and a standard value change timing. The configuration includes a determination unit 35.

  The setting unit 30 sets dead times Td1 to Td3 generated between devices that acquire data, margin times Tm1 and Tm2, and the associated set period Ts, and transmits them to the units 31 to 35. These values may be acquired from the user via an input device such as a keyboard, may be acquired from a storage device, or may be acquired from another device via a communication device. Further, the margin times Tm1 and Tm2 and the set period Ts may be expressed by the number of intermediate products.

  Here, the first margin time Tm1 refers to a period from the time when the change of the set value of the control value or the change of the standard value of the intermediate characteristic is instructed to the first acquisition time extracted from various data. The first margin time Tm1 is a time lag from when the change of the set value or standard value is instructed to the equipment until the change is actually reflected on the workpiece, or a deviation due to defective removal or re-input. Or is set to be removed.

  The second margin time Tm2 is a period from the last acquisition time extracted from the intermediate characteristic data to the time when the first control target device 3 is instructed to change the set value of the control value. The second margin time Tm2 is set in consideration of the processing time spent for obtaining the set value to be changed.

  In addition, the following three methods are mentioned as a method of deriving the values of the dead times Td1 to Td3 and the margin times Tm1 and Tm2. That is, firstly, there is a method derived by calculation from the design value of the quality control system, that is, the time required for processing in each device, and the design value of the moving distance and moving speed when the work moves between the devices. It is done. Secondly, there is a method in which a test operation is performed to measure the processing time and movement time of the workpiece and derive from the actual measurement values. Thirdly, there is a method in which an actual manufacturing operation is performed, and a processing time and a moving time of the workpiece are measured and derived from an actual measurement value.

  In practice, it is desirable to derive these methods by appropriately combining them. For example, the initial setting value before the quality control system is operated is derived by the first method, and after the operation, the fine adjustment is performed by the second or third method.

  The control value extraction timing determination unit 31 determines a control value extraction timing that is a timing for extracting a measured value of the control value from each control data. First, the first dead time Td1, the first margin time Tm1, and the set period Ts are acquired from the setting unit 30, and the operation start time T0 for starting the operation of the scheduler 12 and the change of the set value of the control value are changed. A change instruction time Ti to be instructed (i is an integer equal to or greater than 1) is acquired from the control value change timing determination unit 34. Hereinafter, the operation start time and the change instruction time are collectively referred to as Ti (i is an integer of 0 or more).

  Next, the control value extraction timing determination unit 31 determines, as the first control value extraction timing, from time Ti + Tm1 obtained by adding the first margin time Tm1 to the operation start time or change instruction time Ti to the set period Ts. Next, the time from the time Ti + Tm1 + Td1 obtained by adding the first dead time Td1 to the start time Ti + Tm1 of the first control value extraction timing to the set period Ts is determined as the second control value extraction timing. In addition, the control value extraction timing determination unit 31 transmits the second control value extraction timing start time Ti + Tm1 + Td1 to the intermediate characteristic extraction timing determination unit 32.

  The intermediate characteristic extraction timing determination unit 32 determines an intermediate characteristic extraction timing that is a timing for extracting the intermediate characteristic from the intermediate characteristic data. First, the second dead time Td2 and the set period Ts are acquired from the setting unit 30, and the start time Ti + Tm1 + Td1 of the second control value extraction timing is acquired from the control value extraction timing determination unit 31.

  Next, the intermediate characteristic extraction timing determination unit 32 determines, as the intermediate characteristic extraction timing, from the time Ti + Tm1 + Td1 + Td2 obtained by adding the second dead time Td2 to the start time Ti + Tm1 + Td1 of the second control value extraction timing to the set period Ts. The intermediate characteristic extraction timing determination unit 32 transmits the intermediate characteristic extraction timing start time Ti + Tm1 + Td1 + Td2 to the final characteristic extraction timing determination unit 33 and transmits the intermediate characteristic extraction timing end time Ti + Tm1 + Td1 + Td2 + Ts to the control value change timing determination unit 34. To do.

  The final characteristic extraction timing determination unit 33 determines a final characteristic extraction timing that is a timing for extracting the final characteristic from the final characteristic data. First, the third dead time Td3 and the set period Ts are acquired from the setting unit 30, and the intermediate characteristic extraction timing start time Ti + Tm1 + Td1 + Td2 is acquired from the intermediate characteristic extraction timing determination unit 32.

  Next, the final characteristic extraction timing determination unit 33 determines, as the final characteristic extraction timing, from the time Ti + Tm1 + Td1 + Td2 + Td3 obtained by adding the third dead time Td3 to the start time Ti + Tm1 + Td1 + Td2 of the intermediate characteristic extraction timing to the set period Ts. Further, the final characteristic extraction timing determination unit 33 transmits the final characteristic extraction timing end time Ti + Tm1 + Td1 + Td2 + Td3 + Ts to the standard value change timing determination unit 35.

  The control value change timing determination unit 34 determines first and second control value change timings that are timings for changing the set values of the control values in the first and second control target devices 3 and 4. First, the first dead time Td1 and the second margin time Tm2 are acquired from the setting unit 30, and the end time Ti + Tm1 + Td1 + Td2 + Ts of the intermediate characteristic extraction timing is acquired from the intermediate characteristic extraction timing determination unit 32.

  Next, the control value change timing determination unit 34 determines the time Ti + Tm1 + Td1 + Td2 + Ts + Tm2 obtained by adding the second margin time Tm2 to the end time Ti + Tm1 + Td1 + Td2 + Ts of the intermediate characteristic extraction timing as the first control value change timing. The time Ti + Tm1 + 2 × Td1 + Td2 + Ts + Tm2 obtained by adding the first dead time Td1 to the timing Ti + Tm1 + Td1 + Td2 + Ts + Tm2 is determined as the second control value change timing. The control value change timing determination unit 34 transmits the first control value change timing Ti + Tm1 + Td1 + Td2 + Ts + Tm2 to the control value extraction timing determination unit 31 and the standard value change timing determination unit 35 as the next change instruction time Ti.

  The standard value change timing determination unit 35 determines a standard value change timing that is a timing for changing the standard value of the intermediate characteristic in the intermediate characteristic tester 5. First, the first and second dead times Td1 and Td2 are acquired from the setting unit 30, the end time Ti + Tm1 + Td1 + Td2 + Td3 + Ts of the final characteristic extraction timing is acquired from the final characteristic extraction timing determination unit 33, and changed from the control value change timing determination unit 34 The instruction time Ti is acquired.

  Next, the standard value change timing determination unit 35 specifies the change instruction time Ti that appears first after the end time Ti + Tm1 + Td1 + Td2 + Td3 + Ts of the final characteristic extraction timing, and the first and second dead times Td1,. The time Ti + Td1 + Td2 at which Td2 is added is determined as the standard value change timing.

  FIG. 6 illustrates the timing operation of the scheduler 12 configured as described above. In the figure, the areas to which n to (n + 3) are attached indicate a set associated with each device 3 to 6. In the illustrated example, the first dead time Td1 is 2 seconds, the second dead time Td2 is 4 seconds, and the third dead time Td3 is 20 seconds. The first and second margin times Tm1 and Tm2 are set to 0.75 seconds, and the set period Ts is set to 4.5 seconds.

  After the scheduler 12 starts operation, the control value change timing determination unit 34 determines the change instruction time Tn (= t) for the nth set. Next, the control value extraction timing determination unit 31 determines the time Tn + Tm1 (= t + 0.75) to the set period Ts (= 4.5) as the first control value extraction timing, and the time Tn + Tm1 + Td1 (= t + 2 + 0. 75) to the set period Ts is determined as the second control value extraction timing.

  Next, the intermediate characteristic extraction timing determination unit 32 determines the interval from the time Tn + Tm1 + Td1 + Td2 (= t + 6 + 0.75) to the set period Ts as the intermediate characteristic extraction timing. Since the control data can be extracted before the intermediate characteristic extraction timing, and the intermediate characteristic data can be extracted at the intermediate characteristic extraction timing, as described above, based on the extracted control data and intermediate characteristic data, the control target devices 3 and 4 are controlled. Each control value set value is obtained, and a control command for changing to the obtained set value is transmitted to each of the control target devices 3 and 4.

  The control value change timing determination unit 34 determines the transmission timing. That is, the control value change timing determination unit 34 determines the time Tn + Tm1 + Td1 + Td2 + Ts + Tm2 (= t + 12) as the first control value change timing, and determines the time Tn + Tm1 + 2 × Td1 + Td2 + Ts + Tm2 (= t + 14) as the second control value change timing. Further, the control value change timing determination unit 34 changes the time Tn + Tm1 + Td1 + Td2 + Ts + Tm2 (= t + 12), which is the first control value change timing, to the change instruction item Tn + 1 (= t + 12) for the (n + 1) th set. The extraction of data and intermediate characteristic data and the change of the set value of the control value are repeated.

  On the other hand, the final characteristic extraction timing determination unit 33 determines from the time Ti + Tm1 + Td1 + Td2 + Td3 (= t + 26 + 0.75) to the set period Ts as the final characteristic extraction timing. Intermediate characteristic data can be extracted by this final characteristic extraction timing, and final characteristic data can be extracted at this final characteristic extraction timing. Therefore, as described above, based on the extracted intermediate characteristic data and final characteristic data, the standard of intermediate characteristics can be extracted. A value is obtained, and a control command for changing to the obtained standard value is transmitted to the intermediate characteristic tester 5.

  This transmission timing is determined by the standard value change timing determination unit 35. That is, the control value change timing determination unit 35 specifies the change instruction time Ti (= t + 36) that appears first after the end time Ti + Tm1 + Td1 + Td2 + Td3 + Ts (= t + 26 + 0.75 + 4.5) of the final characteristic extraction timing, and the time Ti + Td1 + Td2 (= t + 42). ) Is determined as the standard value change timing. Thereafter, the extraction of the intermediate characteristic data and the final characteristic data and the change of the standard value are repeated.

  Some acquired data is inappropriate for the above association. FIG. 14 shows the types of inappropriately acquired data that occurs in a general production site, and the cause and contents thereof. As shown in the figure, “abnormal value”, “re-measurement”, “re-injection”, and “removed product” can be cited as inappropriate types of acquired data.

  “Abnormal value” indicates a case where the measured value is significantly different from the normal value. In general, in order to estimate a data distribution by calculating a statistic from a large number of data, it is desirable that the data has a shape of a data distribution close to a normal distribution in which the data are gathered around an average value. For this reason, when the abnormal value is included in the set as acquired data, it is difficult to appropriately estimate the data distribution of the set because the influence of the abnormal value increases.

  “Remeasurement” indicates a case where the device performs remeasurement on the same workpiece 2 due to measurement failure or the like. In this case, a plurality of measurement values exist for the same workpiece 2, and the measurement values before re-measurement are highly likely to be abnormal. Therefore, when the measurement value before remeasurement is included in the set as acquired data, it is difficult to appropriately estimate the data distribution of the set, as in the case of the abnormal value.

  “Re-input” indicates, for example, a case where the work 2 determined to be out of specification by the intermediate characteristic inspector 5 is re-input into a processing process performed by the second control target device 4. In this case, with respect to the workpiece 2 that has been re-introduced, the acquired data of the second controlled device 4 and the acquired data of the intermediate characteristic tester 5 and the final characteristic tester 6 in the subsequent process of the second controlled device 4 However, there is no acquisition data of the first control target device 3 in the previous process of the second control target device 4. For this reason, regarding the sets associated with the devices 3 to 6, the work 2 corresponding to the acquired data included in each set relates to the second control target device 4, the intermediate characteristic tester 5, and the final characteristic tester 6. The set differs from the set related to the first control target device 3, and it is difficult to accurately compare the statistics calculated from the sets.

  The “removed product” indicates, for example, a case where the workpiece 2 determined by the intermediate characteristic inspector 5 to be out of standard is excluded immediately after the inspection. In this case, regarding the workpiece 2 to be excluded, the acquired data of the intermediate characteristic inspector 5 and the acquired data of the first and second control target devices 3 and 4 in the previous process of the intermediate characteristic inspector 5 exist. The acquired data of the final characteristic tester 6 in the subsequent process of the intermediate characteristic tester 5 does not exist. For this reason, regarding the sets associated with the devices 3 to 6, the work 2 corresponding to the acquired data included in each set is the set related to the first and second control target devices 3 and 4 and the intermediate characteristic inspector 5. Therefore, it is difficult to compare the statistics calculated from each set with high accuracy.

  Thus, in order to perform statistical processing with high accuracy, it is desirable to remove inappropriately acquired data from the set. On the other hand, whether or not the acquired data should be removed depends on the type of inappropriate acquired data and whether the set to be associated is a pre-process or a post-process.

  Therefore, in the present embodiment, each of the devices 3 to 6 adds the inappropriate type information indicating the type of inappropriate data to the acquired data, transmits it to the quality control apparatus 10, and stores it in the data storage unit 11. Yes. Moreover, the quality control apparatus 10 stores the table shown in FIG. 15 in a storage unit (not shown). This table shows whether or not the acquired data is invalid data that should be removed or valid data that does not need to be removed when the type of inappropriate acquired data is associated with a set of previous or subsequent processes. Is stored.

  Then, the scheduler refers to the table shown in FIG. 15 and the inappropriate type information stored in the data storage unit 11, removes invalid data from the acquired data, and associates the set of acquired data with each other. It is carried out. As a result, inappropriate acquisition data can be removed depending on the type and whether it is associated with the previous process or the subsequent process, so that statistical processing can be performed with high accuracy. If the acquired data does not correspond to inappropriate data, it is processed as valid data. Further, when the acquired data corresponds to a plurality of types of inappropriate data, if at least one of the acquired data corresponds to invalid data, the acquired data is processed as invalid data.

  Next, details of the standard value generator 13 will be described with reference to FIG. The standard value generation unit 13 extracts a set of intermediate characteristic data from the intermediate characteristic data storage DB 23 based on the intermediate characteristic extraction timing determined by the scheduler 12, and determines the final value based on the final characteristic extraction timing determined by the scheduler 12. A set of final characteristic data is extracted from the characteristic data storage DB 25, and standard values of intermediate characteristics are generated based on the extracted set. In addition, the standard value generation unit 13 transmits the generated standard value to the intermediate characteristic tester 5 based on the standard value change timing determined by the scheduler 12.

  FIGS. 7A and 7B are diagrams illustrating the operation of generating standard values. First, as shown in FIG. 5A, an average value X￣ and standard deviation σ relating to the distribution of the extracted final characteristic A are obtained. Next, in order to examine the positional relationship of the standard value with respect to the distribution of the final characteristic A, the distance du · dl between the average value X￣ of the distribution of the final characteristic A and the upper and lower standard values Su · Sl of the final characteristic A. Is obtained from the equation du = Su−X￣ and the equation dl = X￣−Sl. Further, as shown in FIG. 5B, an average value Xa￣ and a standard deviation σa relating to the distribution of the extracted intermediate characteristic A are obtained.

Next, based on the ratio σa / σ of the standard deviation of the intermediate characteristic A to the standard deviation of the final characteristic A and the following equation, the distance du · dl in the final characteristic A is the distance dua corresponding to the intermediate characteristic A. -Convert to dla.
dua = du × (σa / σ) = (Su−X￣) × (σa / σ),
dla = dl × (σa / σ) = (X） −S1) × (σa / σ).

Then, using the converted distance dua · dla and the average value X￣a relating to the distribution of the intermediate characteristic A, the upper limit and lower limit standard values Sua · Sla of the intermediate characteristic A are obtained by the following equations.
Sua = Xa￣ + dua = Xa￣ + (Su−X￣) × (σa / σ),
Sla = Xa￣-dla = Xa￣- (X￣-Sl) × (σa / σ).

  Each standard value is generated by repeating the above-described standard value generation operation for intermediate characteristics other than the intermediate characteristic A such as the intermediate characteristic B. The standard value can be generated in the same manner when the standard value is only one of the upper limit and the lower limit.

  By the way, ideally, if all the defective products are eliminated by the intermediate characteristic inspection, the defective products should not be excluded by the final characteristic inspection of the same inspection item. However, in reality, this is not always the case because there are disturbance factors and measurement errors.

  In addition, if the disposal cost is higher when the defective product is eliminated in the final characteristic inspection than the disposal cost when the defective product is eliminated in the intermediate characteristic inspection, it is better to eliminate it in the intermediate characteristic inspection if it is the same inspection item. Overall manufacturing costs are reduced.

  Therefore, it is desirable that the standard value of the intermediate characteristic is stricter than the standard value of the final characteristic. Specifically, the standard value generation unit 13 desirably adjusts the standard value generated based on the above-described formula based on the safety coefficient.

  Conventionally, the safety factor has been empirically set to 3 to 4, but recently, the safety factor has been calculated based on the tolerance design proposed in quality engineering (for example, the second in Non-Patent Document 1). See chapter and chapter 3). The tolerance design is a method of converting the quality into a loss by a loss function and calculating an optimum tolerance that minimizes the loss.

First, the safety factor for calculating the optimum tolerance is defined by the following equation.
φ = (A0 / A) ^1/2 .
Here, φ is a safety factor, A is a disposal cost when a defective product is eliminated in the intermediate inspection, and A0 is a disposal cost when a defective product is eliminated in the final inspection.

Then, the upper limit and lower limit standard values Sua · Sla of the intermediate characteristic A are adjusted as in the following equation.
Sua = Xa￣ + dua / φ = Xa￣ + (Su−X￣) × (σa / σ) / φ,
Sla = Xa￣−dla / φ = Xa￣− (X￣−Sl) × (σa / σ) / φ.

In the present embodiment, since the intermediate characteristic is the desired desired characteristic as it is closer to the target value, the above equation is obtained. When the intermediate characteristic is a desired small characteristic that is closer to 0, the standard value of the intermediate characteristic is only the upper limit standard value Sua in the above formula. In addition, when the intermediate characteristic is a desirable desired characteristic as it is larger, the standard value of the intermediate characteristic is only the lower limit standard value Sla, and is adjusted according to the following equation.
Sla = Xa￣−dla × φ = Xa￣− (X￣−Sl) × (σa / σ) × φ.

  Next, details of the influence generation unit 14 will be described with reference to FIG. As described above, the influence degree generation unit 14 generates an influence degree indicating how the intermediate characteristic fluctuates when the control value is changed based on a request from the estimation part 16, and the generated influence degree. Is sent to the estimation unit 16 and stored in the influence storage DB 15.

  First, the influence level generation unit 14 sets an initial value of the influence degree in advance and stores it in the influence degree storage DB 15 before the scheduler 12 operates. The initial value of the degree of influence is calculated in advance by calculating the amount of change in the average value of the control factor (control value) and the amount of change in the average value and variation (variance / standard deviation) of the intermediate characteristics. Can be obtained.

  FIG. 8 shows the processing operation of the influence generation unit 14 after the operation of the scheduler 12. As shown in the figure, first, based on the previous and current first control value extraction timings determined by the scheduler 12, control data (hereinafter, referred to as “control target 1”) from the control data storage DB 21. Two sets of “first control data” are extracted. Similarly, based on the previous and current second control value extraction timings determined by the scheduler 12, the control data (hereinafter referred to as “second”) from the control data storage DB 21 regarding the second control target device 4 (control target 2). Two sets of “control data” are extracted. Similarly, two sets of intermediate characteristics are respectively extracted from the intermediate characteristic data storage DB 23 based on the previous and current intermediate characteristic extraction timings determined by the scheduler 12 (S30).

  Next, the average value of the set of intermediate characteristics extracted this time is calculated (S31), and the difference between the calculated average value and the estimated value of the average value of the intermediate characteristics acquired from the estimation unit 16 is greater than the predetermined value D. It is determined whether it is larger (S32). If it is larger (YES in S32), the following processing operation is performed. If it is not larger (NO in S32), the processing operation of the influence generation unit 14 is terminated.

Next, regarding the average value of the control data, the current change amount with respect to the previous time is calculated (S33). That is, an average value regarding a set of first control data (measured values) extracted based on the previous first control value extraction timing is X (n−1), and based on the current first control value extraction timing. Assuming that the average value for the set of first control data (measured values) extracted in this way is X (n), the current change amount with respect to the previous time is expressed by the following equation with respect to the average value of the first control data.
(Change amount of average value of first control data) = X (n) −X (n−1).

Next, regarding the average value and the variance of the intermediate characteristics, the current fluctuation amount with respect to the previous time is calculated (S34). That is, the average value and variance regarding the set of intermediate characteristics A extracted based on the previous intermediate characteristic extraction timing are Y (n−1) and Z (n−1), respectively, and based on the current intermediate characteristic extraction timing. Assuming that the average value and variance for the set of intermediate characteristics A extracted in this way are Y (n) and Z (n), respectively, with respect to the average value and the variance of the intermediate characteristics A, Is done.
(Fluctuation amount of average value of intermediate characteristic A) = Y (n) −Y (n−1),
(Fluctuation amount of dispersion of intermediate characteristic A) = Z (n) −Z (n−1).

Next, the average value of the intermediate characteristic and the variation amount of the dispersion are normalized by the change amount of the average value of the first control data, thereby generating the average value of the intermediate characteristic and the influence of the dispersion, respectively (S35). ). That is, the influence degree k1 of the intermediate characteristic A on the average value and the influence degree k2 of the intermediate characteristic A on the dispersion are expressed by the following equations.
k1 = (Y (n) -Y (n-1)) / (X (n) -X (n-1)),
k2 = (Z (n) -Z (n-1)) / (X (n) -X (n-1)).

  Using the above procedure, the influence degree of the second control data and the influence degree of the other intermediate characteristic B are also obtained. That is, actually, the average value of the intermediate characteristic A and the degree of influence on the variance by the first control data, the average value of the intermediate characteristic A by the second control data and the degree of influence on the variance, and the first control The average value of the intermediate characteristic B by data and the degree of influence on the variance and the average value of intermediate characteristic B by the second control data and the degree of influence by the variance are generated.

  And while producing | generating the influence degree to the estimation part 16, the influence degree memorize | stored in influence degree memory | storage DB15 is updated (S36), and the operation | movement of the influence degree production | generation part 14 is complete | finished after that.

  Next, the detail of the estimation part 16 is demonstrated based on FIGS. 9-13. FIG. 9 shows a schematic configuration of the estimation unit 16. As illustrated, the estimation unit 16 includes an intermediate characteristic estimation unit (intermediate characteristic estimation unit) 40, a defect estimation unit (defective number estimation unit) 41, and an estimated value evaluation unit (change amount determination unit) 42. .

  The intermediate characteristic estimator 40 estimates the variation amount of the average value or standard deviation (variance) of the intermediate characteristic when the set value of the control value is changed, using the influence degree read from the influence degree storage DB 15. The defect estimation unit 41 estimates the number of defects when the distribution of the intermediate characteristic is varied by the variation amount estimated by the intermediate characteristic estimation unit 40, and evaluates the sum by the total number of individual defects. The estimated value evaluation unit 42 obtains the amount of change in the set value of the control value when the number of defects is minimized and sends it to the control command unit 17.

  FIG. 10 shows the processing operation of the estimation unit 16 configured as described above. First, the intermediate characteristic estimation unit 40 extracts a set of intermediate characteristic data from the intermediate characteristic data storage DB 23 of the data storage unit 11 based on the intermediate characteristic extraction timing determined by the scheduler 12 (S40).

  Next, the intermediate characteristic estimation unit 40 generates a change pattern that is a combination of change amounts of the set values of the control values related to the control target devices 3 and 4 (S41). FIG. 11 shows an example of the change pattern in a table format. In the present embodiment, since the control target is two control target devices 3 and 4, the change pattern is represented by a two-dimensional table as shown in FIG. Note that the number of change amounts of the set value of the control value can be determined empirically from a change width that is likely to have an effect and an allowable amount of the change width, but may be determined in any manner.

  In the present embodiment, as shown in FIG. 11, the number of change amounts in the control target 1 (control target device 3) and the control target 2 (control target device 4) is set to five. Further, the change width of the control object 1 is 0.003, and the change width of the control object 2 is 0.005. In the present embodiment, as shown in FIG. 11, the change amount includes 0.000, that is, “do not change”, but it is not always necessary to include it.

  Next, the estimated value table shown in FIG. 11 is completed by obtaining the average value of the intermediate characteristic or the estimated value of the standard deviation when the set value of the control value is changed for each change pattern. Specifically, first, the latest influence degree is acquired from the influence degree storage DB 15 (S42), and the acquired influence degree is multiplied by the change width (change amount) of the set value of the control value. As a result, the average value of the intermediate characteristics or the variation amount of the dispersion when the set value of the control value is changed is obtained.

  Note that only one of the average value and the variance of the intermediate characteristics may be used, or both of them may be used. Also, standard deviation may be used instead of variance as an indication of variation. In the following, the average value and standard deviation of intermediate characteristics will be used.

  Next, the fluctuation amount is added to the average value or variance of the extracted current intermediate characteristics. Regarding the variance, after this, the square root is taken and converted to the standard deviation. By these calculations, the average value of the intermediate characteristic or the estimated value of the standard deviation is obtained for each combination of the change amounts of the set values of the control values to be controlled, and the estimated value table shown in FIG. 11 is completed (S43). In FIG. 11, the estimated value Xa, i, j of the average value is shown.

Here, the change width of the set value of the control value is α, the average value of the intermediate characteristic A is X￣, the standard deviation of the intermediate characteristic A is σ, the influence of the average value of the intermediate characteristic A is k1, and the variance of the intermediate characteristic A If the influence degree of k2 is k2, the average value of the intermediate characteristic A and the estimated value Xa ^ · σ ^ of the intermediate characteristic A when the set value of the control value is changed are as follows.
Xa ^ = k1 × α + X￣,
σa ^ = (k2 × α + σ ² ) ^1/2 .

  Next, the defect estimation unit 41 uses the estimated value obtained by the intermediate characteristic estimation unit 40 to create a distribution of the changed intermediate characteristic for each change pattern, and estimates the number of intermediate characteristic defects. Specifically, the standard standard (standard value) is moved by the amount corresponding to the change in the created intermediate characteristic distribution (S44), and the elements (individual intermediate characteristics) included in the extracted set are judged to be acceptable (S45). .

The standard-standard movement of the intermediate characteristic A is expressed by the following equation as described above with respect to the standard value creating unit 13.
Sua = Xa￣ + dua = Xa￣ + (Su−X￣) × (σa / σ),
Sla = Xa￣-dla = Xa￣- (X￣-Sl) × (σa / σ).
As described above, Xa¯ = Xa ^ = k1 × α + X¯, σa = σa ^ = (k2 × α + σ 2) 1/2 so the upper specification limit and the lower limit standard value after the movement becomes the following equation (S44) .
Sua = k1 × α + X ＋+ (Su−X￣) × (k2 × α + σ ² ) ^1/2 / σ,
Sla = k1 × α + X∥− (X∥−S1) × (k2 × α + σ ² ) ^1/2 / σ.

  Using these standard values Sua and Sla, the quality of the intermediate characteristic A of the set extracted this time is determined, and an NG flag is added to the data determined to be defective (S45). By performing the above processing in the same manner for each intermediate characteristic, a pass / fail list of the intermediate characteristic is created. FIG. 12 shows an example of the pass / fail list. In the drawing, “NG” in the “good / bad” column indicates that the NG flag is assigned to each intermediate characteristic. In the illustrated case, there are three types of intermediate characteristics (A to C).

  Next, the estimated value evaluation unit 42 sums up the number of individual intermediate characteristic defects obtained by the defect estimation unit 41 (S46), and selects a change pattern that minimizes the total number of defects (S47). It should be noted here that, when the NG flag is generated with a plurality of intermediate characteristics, it is necessary to avoid counting twice when summing up the number of defects.

  Therefore, the priority order of intermediate characteristics related to pass / fail is determined in advance, and only the NG flag with the higher priority order is counted. Note that any method may be used to determine the priority order. For example, in the case of the pass / fail list shown in FIG. At this time, if the priority order of the intermediate characteristics is A> B> C, the total number of defects = the number of defects A + the number of defects B + the number of defects C = 2 + 1 + 0 = 3.

  It is desirable that the NG flag be represented by bit data that is 1 (ON) when it is given and 0 (OFF) when it is not given. In this case, the sum of the number of defects can be obtained by taking a logical sum (OR) for the NG flags of the intermediate characteristics A to C for each serial number and counting this.

  The estimated value evaluation unit 42 performs the above-described sum of the number of defects (total number of defects) for each change pattern, thereby creating a total defect number table corresponding to the estimated value table. FIG. 13 shows an example of the total defect number table, which corresponds to the estimated value table shown in FIG. The estimated value evaluation unit 42 refers to the total defect number table and selects a change pattern that minimizes the total defect number. In the case of FIG. 13, the amount of change in the set value of the control value of the control target 1 (control target device 3) is 0.003, and the amount of change in the set value of the control value of the control target 2 (control target device 4). Since the number of combinations with 0.010 is the smallest total number of defects 15, this change pattern is selected.

  Then, the estimated value evaluation unit 42 sends the selected change pattern (a combination of change amounts) to the control command unit 17 (S48). Further, the estimated value of the intermediate characteristic corresponding to the selected change pattern is sent to the influence generation unit 14. Thereafter, the operation of the estimation unit 16 is terminated.

  In addition, as a result of selecting the change pattern that minimizes the total number of defects, the change amount of either or both of the control objects 1 and 2 may be 0.000, that is, no change is necessary. In this case, transmission of the change amount to the control command unit 17 may be omitted for the control target.

The control command unit 17 acquires a change pattern selected by the estimated value evaluation unit 42, that is, a combination for each control target regarding the change amount of the set value of the control value, and based on the acquired change pattern, each control target device 3. 4 is changed, and the changed setting value is transmitted to the control target devices 3 and 4 as control commands. For example, the set value of the control value may be changed according to the following equation.
(Set value of current control value) = (Set value of previous control value) + (Change amount of set value of control value that minimizes the number of defects).

  The present invention is not limited to the above-described embodiments, and various modifications can be made within the scope shown in the claims. That is, embodiments obtained by combining technical means appropriately modified within the scope of the claims are also included in the technical scope of the present invention.

  For example, it is considered that there is a slight time lag between the time when the change value of the control value and the standard value of the intermediate characteristic are instructed and the time when each of the devices 3 to 5 actually changes. Further, it is considered that there is a slight time lag between the time when each of the devices 3 to 6 performs measurement and the time when the quality control device 10 acquires various measurement values. For this reason, it is desirable that the scheduler 12 adjusts these time lags to associate data and determine timing.

  Moreover, in this embodiment, although the data storage part 11 has memorize | stored acquisition time, you may receive and memorize | store the measurement time which is the time which each apparatus 3-6 measured from each apparatus 3. FIG. In this case, the scheduler 12 performs association using the measurement time.

  In the present embodiment, the two control target devices 3 and 4 and one intermediate characteristic tester 5 are used. However, the control target device can be one, the control target device and The number of intermediate characteristic testers can be further increased.

  As described above, the quality control device according to the present invention is a quality control device that controls a manufacturing process in order to manufacture a product having a predetermined quality, and is measured by a plurality of measuring devices provided in the manufacturing process. Storage means that collects measured data, stores the collected measurement data together with the measured time or the collected time, and considers the dead time that occurs between the measuring devices at the measured time or the collected time Thus, an association means for associating the measurement data of the measuring device is provided.

  The intermediate product measured by a certain device is measured by the next device at the time when the waste time has passed since the measurement of the intermediate product. Thus, the time measured by each measuring device or the time collected from each measuring device can be roughly correlated by the dead time.

  Therefore, according to the above configuration, since the collected measurement data is associated with the measured time or the collected time, the association unit wastes the measured time or the collected time between the measuring devices. By roughly associating with time, it is possible to associate the measurement data of the measuring devices with each other.

  Further, the dead time is used only by the association unit, and even if the dead time is changed, there is no influence on the data stored in the storage unit. Therefore, it is possible to easily change the dead time.

  Note that the association unit may associate a predetermined number of measurement data of the plurality of measurement devices. In this case, in each measuring device, a set including a predetermined number of measurement data is formed, and the sets in a plurality of measuring devices are associated with each other. Thereby, since the number of measurement data included in each associated set is constant, the accuracy of the statistics of each set can be substantially matched. In addition, since the sets are formed based on the number of measurements, not the measurement period of the measuring device, for example, even when a certain process in the manufacturing process is stopped for a long time and restarted, it is possible to appropriately associate each set. it can.

  In addition, when the measurement data to be associated includes inappropriate measurement data, the association means includes the inappropriate measurement data and measurement data measured by another measurement device, It is desirable to stop the association with the measurement data corresponding to the inappropriate measurement data.

  Here, inadequate measurement data is an abnormal value when the measured value is significantly different from the normal value, or when the equipment re-measures the same intermediate product due to measurement failure, etc. Examples include measurement data when an intermediate product determined to be out of specification is re-introduced into an upstream process, and when an intermediate product determined to be out of specification in an intermediate inspection is excluded immediately after the inspection.

  These inappropriate measurement data cause a decrease in accuracy in performing statistical processing. Therefore, according to the above configuration, inappropriate measurement data is not used for the association, so that it is possible to prevent the accuracy of statistical processing from being lowered.

  In the case of re-introducing the intermediate product, the re-introduced intermediate product has the process to be re-introduced and the measurement data of the measuring instrument downstream thereof, but the upstream side of the process to be re-introduced There is no measurement data of the measuring instrument in Therefore, when an intermediate product is re-introduced in a certain process in the manufacturing process, the associating means includes measurement data relating to the intermediate product of the measuring instrument on the upstream side of the process, the process, and the process. It is desirable to cancel the association of the measurement device on the downstream side with the measurement data related to the intermediate product.

  In addition, when an intermediate product is excluded immediately after the inspection, there is a process to be removed and measurement data of a measuring instrument on the upstream side of the intermediate product to be excluded, but the downstream side of the process to be removed. There is no measurement data of the measuring instrument in Therefore, when an intermediate product is removed in a certain process in the manufacturing process, the associating means includes the measurement data of the measuring device on the downstream side of the process, the measurement data on the upstream side of the process, It is desirable to cancel the association with the measurement data related to the intermediate product of the measuring device.

  A quality control device according to the present invention is a quality control device for controlling a control target device provided in a manufacturing process in order to manufacture a product having a predetermined quality, and is a control target value for the control target device. Control value set value as control value, control data that is a measurement value of the control value measured by the control target device with respect to the set value of the control value, and intermediate characteristic data of the intermediate product measured by the intermediate characteristic tester Storage means for collecting and storing the intermediate characteristic estimation means for estimating fluctuations in the data distribution of the intermediate characteristic data when the set value of the control value stored in the storage means is changed, and the intermediate characteristic estimation means The number of defects estimating means for estimating the number of defects based on the variation in the data distribution of the estimated intermediate characteristic data and the intermediate characteristic standard value, and the above-described control that minimizes the number of defects estimated by the number of defects estimating means Value A change amount determination unit that determines a change amount of a constant value, and a control change unit that changes a set value of the control value of the control target device based on the change amount determined by the change amount determination unit Yes.

  Here, as the quantity indicating the data distribution, various values such as an average value and the above-described variation value can be used.

  According to said structure, the fluctuation | variation of the data distribution of intermediate characteristic data at the time of changing the setting value of a control value is estimated, and the defect number which generate | occur | produces is estimated. Then, the amount of change in the set value of the control value is determined so that the estimated number of defects is minimized, and the set value of the control value of the control target device is changed.

  As a result, the intermediate product is inspected during the manufacturing, and the control of the control target device is changed so that the number of defective intermediate products is reduced. Therefore, the interval of the control feedback loop can be shortened. In addition, by estimating the variation in the data distribution, it is possible to take into account data variations, and thus it is possible to more accurately estimate the number of defects generated.

  In addition, the storage apparatus further includes an influence degree storage unit that stores an influence degree indicating a degree of variation in the data distribution of the intermediate characteristic data with respect to the change amount of the average value of the control data. The variation amount of the data distribution of the intermediate characteristic data is estimated based on the change amount of the set value of the control value stored by the means and the influence degree stored by the influence degree storage means. It is desirable to determine the amount of change in the set value of the control value so that the number of defects estimated by the defect number estimation means is minimized. In this case, since the correspondence relationship between the change amount of the average value of the control data and the fluctuation amount of the data distribution of the intermediate characteristic data is known, the fluctuation amount of the data distribution of the intermediate characteristic data can be estimated more accurately and is generated. The number of defects can be estimated more accurately.

  Further, the data distribution of the intermediate characteristic data corresponding to the change amount of the set value of the control value determined by the change amount determination means, and the intermediate characteristic data collected by the storage means after the control change means changes control. It is desirable to further include an influence degree creating means for creating the influence degree and storing the influence degree in the influence degree storing means when a predetermined difference or more occurs between the distribution and the distribution. In this case, even if some variation occurs in the manufacturing process, the degree of influence can be automatically corrected, so that the amount of variation in the data distribution of the intermediate characteristic data can be estimated more accurately, and a product with appropriate quality can be manufactured. .

  The storage means further collects and stores the final characteristic data of the finished product measured by the final characteristic tester, and stores the data distribution of the final characteristic data stored by the storage means and the final characteristic tester. Based on the set final characteristic standard value, the apparatus further includes a standard value generating unit that generates an intermediate characteristic standard value to be set in the intermediate characteristic inspector. The defect number estimating unit includes the standard value generating unit. It is desirable to use the generated intermediate characteristic standard value. In this case, even if some variation occurs in the manufacturing process, the intermediate characteristic standard value can be automatically corrected. Therefore, it is possible to estimate the number of defects generated more accurately and manufacture a product with appropriate quality.

  The storage means stores the control data and the intermediate characteristic data together with the measured time or the collected time. At the measured time or the collected time, the control target device and the intermediate characteristic tester are connected to each other. It is desirable to further include association means for associating the control data and the intermediate characteristic data by taking into account the dead time generated between the control data and the intermediate characteristic data. In this case, since the collected control data and intermediate characteristic data are associated with the measured time or the collected time, the associating means determines the measured time or the collected time between the control target device and the intermediate characteristic inspector. By roughly associating with the dead time generated between the control data and the intermediate characteristic data, it is possible to associate with each other.

  The intermediate characteristic estimation means estimates the data distribution of the next intermediate characteristic data using the intermediate characteristic data collected by the storage means after a predetermined time has elapsed since the control change means changed control. It is desirable to do. In this case, since data during the change of control is not used, more accurate control is possible.

  A control method for a quality control apparatus according to the present invention is a control method for a quality control apparatus that controls a manufacturing process in order to manufacture a product having a predetermined quality, and the quality control apparatus is provided in the manufacturing process. Storage means for collecting measurement data measured by a plurality of measurement devices and storing the collected measurement data together with the measured time or the collected time. An association step of associating the measurement data of the measuring devices with each other by considering the dead time generated between the measurement devices.

  According to the above method, since the collected measurement data is associated with the measured time or the collected time, the association unit determines the measured time or the collected time according to the dead time generated between the measuring devices. By roughly associating, the measurement data of the measuring devices can be associated with each other. In addition, the dead time is used only in the association step, and changing the dead time does not affect the data stored in the storage unit. Therefore, it is possible to easily change the dead time.

  A control method for a quality control apparatus according to the present invention is a control method for a quality control apparatus for controlling a control target device provided in a manufacturing process in order to manufacture a product having a predetermined quality. Is a control value setting value set as a control target value for the control target device, control data that is a measurement value of the control value measured by the control target device with respect to the control value setting value, and an intermediate A storage means for collecting and storing intermediate characteristic data of the intermediate product measured by the characteristic inspector, and fluctuation in data distribution of the intermediate characteristic data when the set value of the control value stored by the storage means is changed An intermediate characteristic estimating step for estimating the number of defects based on the fluctuation of the data distribution of the intermediate characteristic data estimated in the intermediate characteristic estimating step and the intermediate characteristic standard value; Based on the change amount determination step for determining the change amount of the set value of the control value so that the number of failures estimated in the defect number estimation step is minimized, and the change amount determined in the change amount determination step, A control change step of changing a set value of the control value of the control target device.

  According to the above method, the fluctuation of the data distribution of the intermediate characteristic data when the set value of the control value is changed is estimated, and the number of defects generated is estimated. Then, the amount of change in the set value of the control value is determined so that the estimated number of defects is minimized, and the set value of the control value of the control target device is changed. As a result, the intermediate product is inspected during the manufacturing, and the control of the control target device is changed so that the number of defective intermediate products is reduced. Therefore, the interval of the control feedback loop can be shortened. In addition, by estimating the variation in the data distribution, it is possible to take into account data variations, and thus it is possible to more accurately estimate the number of defects generated.

  Each block of the quality control apparatus 10 may be configured by hardware logic, or may be realized by software using a CPU as follows.

  That is, the quality control apparatus 10 includes a storage device (such as a CPU that executes instructions of a control program for realizing each function, a ROM that stores the program, a RAM that expands the program, a memory that stores the program and various data, and the like. Recording medium). An object of the present invention is a recording medium on which a program code (execution format program, intermediate code program, source program) of a control program of the quality control apparatus 10 which is software for realizing the functions described above is recorded so as to be readable by a computer. This can also be achieved by supplying the quality control apparatus 10 and reading and executing the program code recorded on the recording medium by the computer (or CPU or MPU).

  Examples of the recording medium include tape systems such as magnetic tape and cassette tape, disk systems including magnetic disks such as flexible disks / hard disks, and optical disks such as CD-ROM / MO / MD / DVD / CD-R, and IC cards. A card system such as an optical card (including a memory card) or a semiconductor memory system such as a mask ROM / EPROM / EEPROM / flash ROM can be used.

  Further, the quality control apparatus 10 may be configured to be connectable to a communication network, and the program code may be supplied via the communication network. The communication network is not particularly limited. For example, the Internet, intranet, extranet, LAN, ISDN, VAN, CATV communication network, virtual private network, telephone line network, mobile communication network, satellite communication. A net or the like is available. Also, the transmission medium constituting the communication network is not particularly limited. For example, even in the case of wired such as IEEE 1394, USB, power line carrier, cable TV line, telephone line, ADSL line, etc., infrared rays such as IrDA and remote control, Bluetooth ( (Registered trademark), 802.11 wireless, HDR, mobile phone network, satellite line, terrestrial digital network, and the like can also be used. The present invention can also be realized in the form of a carrier wave or a data signal sequence in which the program code is embodied by electronic transmission.

  The quality control device according to the present invention can be applied to devices that maintain various states by controlling various processes such as control of home appliances in addition to manufacturing processes.

It is a block diagram which shows schematic structure of the quality control system containing the quality control apparatus which is embodiment of this invention. It is a flowchart which shows the processing operation of the said quality control apparatus. It is a functional block diagram which shows schematic structure of the data storage part in the said quality control apparatus. It is a figure which shows an example of the data memorize | stored in the said data storage part in a table | surface form, The same figure (a)-(d) is respectively 1st control data, 2nd control data, intermediate characteristic data, and last. An example of characteristic data is shown. It is a functional block diagram which shows schematic structure of the scheduler in the said quality control apparatus. It is a timing chart which illustrates the timing operation of the scheduler. (A) is a graph showing the data distribution of the final characteristic data, and (b) is a graph showing the data distribution of the intermediate characteristic data. The standard value is changed based on the data distribution of the final characteristic data. It shows how it is done. It is a flowchart which shows the processing operation of the influence generation part in the said quality control apparatus. It is a functional block diagram which shows schematic structure of the estimation part in the said quality control apparatus. It is a flowchart which shows the processing operation of the said estimation part. It is a figure which shows an example of the estimated value table of the intermediate characteristic which the intermediate characteristic estimation part in the said estimation part produces in a table format. It is a figure which shows an example of the quality list | wrist of the intermediate characteristic which the defect estimation part in the said estimation part produces in a table format. It is a figure which shows an example of the total defect number table which the estimated value evaluation part in the said estimation part produces in a table format. It is a figure which shows the correspondence of the kind of the improper acquisition data which arises at a common production site, its cause example, and the content in a table format. A table showing the correspondence between information indicating whether the acquired data is invalid or valid and the reason for it in the form of a table when the inappropriate type of acquired data is associated with a set of previous or subsequent processes. is there.

Explanation of symbols

1 Quality control system 3.4 Controlled equipment (measuring equipment)
5 Intermediate characteristic tester (measuring equipment)
6 Final characteristic tester (measuring equipment)
10 Quality Control Device 11 Data Storage Unit (Storage Unit)
12 Scheduler (association means)
13 Standard value generator (standard value generator, standard value setting controller)
14 Influence Level Generating Unit (Influence Level Generating Unit)
15 Influence storage DB (impact storage means)
17 Control command section (control change means)
40 Intermediate characteristic estimation unit (intermediate characteristic estimation means)
41 Number of defects estimation unit (number of defects estimation means)
42 Estimated value evaluation unit (change amount determining means)

Claims (11)

  A quality control device for controlling a manufacturing process in order to manufacture a product having a predetermined quality,
  A storage means for collecting and storing intermediate characteristic data of the intermediate product measured by the intermediate characteristic inspector and final characteristic data of the finished product measured by the final characteristic inspector;
  Based on the data distribution of the final characteristic data stored in the storage means and the final characteristic standard value set in the final characteristic inspector, the intermediate characteristic inspector is a criterion for eliminating defective intermediate products. A standard value generating means for generating a certain intermediate characteristic standard value;
  A standard value setting control means for controlling the intermediate characteristic tester so as to set the standard value generated by the standard value generating means,
  The storage means stores the intermediate characteristic data and the final characteristic data together with a measured time or a collected time,
  Associating means for associating the intermediate characteristic data and the final characteristic data by considering the dead time generated between the intermediate characteristic inspector and the final characteristic inspector at the measured time or the collected time. The quality control apparatus further comprising:   The quality according to claim 1, wherein the standard value generation means generates the intermediate characteristic standard value based on a difference between an average value of a data distribution of the final characteristic data and the final characteristic standard value. Control device.   The standard value generation means includes an average value of the data distribution of the intermediate characteristic data stored by the storage means and the intermediate characteristic specification value for a difference between the average value of the data distribution of the final characteristic data and the final characteristic specification value. The intermediate characteristic standard value is generated such that a ratio of the difference between the intermediate characteristic data and the final characteristic data is a ratio of a variation value of the data distribution of the intermediate characteristic data to a variation value of the data distribution of the final characteristic data. 2. The quality control apparatus according to 2. A quality control device for controlling a manufacturing process in which it is difficult to grasp the causal relationship between a control factor and a product characteristic in order to manufacture a product having a predetermined quality,
A storage means for collecting and storing intermediate characteristic data of the intermediate product measured by the intermediate characteristic inspector and final characteristic data of the finished product measured by the final characteristic inspector;
Based on the data distribution of the final characteristic data stored in the storage means and the final characteristic standard value set in the final characteristic inspector, the intermediate characteristic inspector is a criterion for eliminating defective intermediate products. A standard value generating means for generating a certain intermediate characteristic standard value;
A standard value setting control means for controlling the intermediate characteristic tester so as to set the standard value generated by the standard value generating means,
The standard value generation means obtains an average value X￣ and standard deviation σ of the data distribution of the final characteristic data,
A distance du between the average value X￣ and the upper limit standard value Su of the final characteristic data;
From the standard deviation σa of the data distribution of the intermediate characteristic data,
dua = du × (σa / σ) = (Su−X￣) × (σa / σ)
A distance dua between the average value Xa￣ of the data distribution of the intermediate characteristic data and the upper limit standard value of the intermediate characteristic data is obtained,
From the distance dua and the average value Xa￣
Sua = Xa￣ + dua = Xa￣ + (Su−X￣) × (σa / σ)
A first process for obtaining an upper limit standard value Sua of the intermediate characteristic data;
and,
A distance dl between the average value X￣ and the lower limit standard value Sl of the final characteristic data;
From the standard deviation σa of the data distribution of the intermediate characteristic data,
dla = dl × (σa / σ) = (X￣−S1) × (σa / σ)
A distance dla between an average value Xa￣ of the data distribution of the intermediate characteristic data and a lower limit standard value of the intermediate characteristic data is obtained,
From the distance dla and the average value Xa￣,
Sla = Xa￣−dla = Xa￣− (X∥−Sl) × (σa / σ)
A quality control apparatus characterized by executing at least one of second processes for obtaining a lower limit standard value Sla of the intermediate characteristic data.   The quality control device according to claim 1 or 4, wherein the standard value generating means further generates the intermediate characteristic standard value based on a safety factor.   The safety factor is calculated based on a disposal cost when the finished product is excluded based on the final characteristic standard value and a disposal cost when the intermediate product is excluded based on the intermediate characteristic standard value. The quality control apparatus according to claim 5.   The standard value generation means uses the intermediate characteristic data and the final characteristic data collected by the storage means after a predetermined time has elapsed since the standard value setting control means controlled the intermediate characteristic inspector. The quality control apparatus according to claim 1, wherein the standard value is generated.   In order to manufacture a product having a predetermined quality, a control method of a quality control device that controls a manufacturing process in which it is difficult to grasp a causal relationship between a control factor and a product characteristic,
  The quality control device includes storage means for collecting and storing intermediate characteristic data of the intermediate product measured by the intermediate characteristic inspector and final characteristic data of the finished product measured by the final characteristic inspector.
  Based on the data distribution of the final characteristic data stored in the storage means and the final characteristic standard value set in the final characteristic inspector, the intermediate characteristic inspector is a criterion for eliminating defective intermediate products. A standard value generation step for generating a certain intermediate characteristic standard value;
  A standard value setting control step for controlling the intermediate characteristic tester so as to set the standard value generated in the standard value generation step,
  The standard value generation step includes a step of obtaining an average value X￣ and standard deviation σ of the data distribution of the final characteristic data, and
  A distance du between the average value X￣ and the upper limit standard value Su of the final characteristic data;
  From the standard deviation σa of the data distribution of the intermediate characteristic data,
    dua = du × (σa / σ) = (Su−X￣) × (σa / σ)
  A distance dua between the average value Xa￣ of the data distribution of the intermediate characteristic data and the upper limit standard value of the intermediate characteristic data is obtained,
  From the distance dua and the average value Xa￣
    Sua = Xa￣ + dua = Xa￣ + (Su−X￣) × (σa / σ)
  A first processing step for obtaining an upper limit standard value Sua of the intermediate characteristic data;
and,
  A distance dl between the average value X￣ and the lower limit standard value Sl of the final characteristic data;
  From the standard deviation σa of the data distribution of the intermediate characteristic data,
    dla = dl × (σa / σ) = (X￣−S1) × (σa / σ)
  A distance dla between an average value Xa￣ of the data distribution of the intermediate characteristic data and a lower limit standard value of the intermediate characteristic data is obtained,
  From the distance dla and the average value Xa￣,
    Sla = Xa￣−dla = Xa￣− (X∥−Sl) × (σa / σ)
  A quality control method comprising: at least one step of a second processing step for obtaining a lower limit standard value Sla of the intermediate characteristic data.   A control method of a quality control device for controlling a manufacturing process in order to manufacture a product having a predetermined quality,
  The quality control device includes storage means for storing intermediate characteristic data of the intermediate product measured by the intermediate characteristic inspector and final characteristic data of the finished product measured by the final characteristic inspector together with the measured time or the collected time. Has
  Based on the data distribution of the final characteristic data stored in the storage means and the final characteristic standard value set in the final characteristic inspector, the intermediate characteristic inspector is a criterion for eliminating defective intermediate products. A standard value generation step for generating a certain intermediate characteristic standard value;
  A standard value setting control step for controlling the intermediate characteristic inspector so as to set the standard value generated in the standard value generation step;
  An association step for associating the intermediate characteristic data with the final characteristic data by considering the dead time generated between the intermediate characteristic inspector and the final characteristic inspector at the time measured or collected. And a quality control method comprising:   A quality control program for operating the quality control apparatus according to any one of claims 1 to 7, wherein the quality control program causes a computer to function as each means.   A computer-readable recording medium on which the quality control program according to claim 10 is recorded.

Images