JP2008165832A - Device, method and program for displaying quality variation, and recording medium recorded the program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for displaying quality variations that can surely recognize periodic information of quality, on the basis of the number of products, such as the occurrence of failures for every prescribed number of products. <P>SOLUTION: The device includes a quality data storage DB 14 for storing results of measuring each product measured with measuring equipment, in association with manufacturing order; an interval statistic calculating unit 21 for shifting interval that corresponds to the products of fixed segment numbers by fixed shift numbers to find statistic amount for each interval, and for producing a graph that indicates the statistics at equal intervals in the manufacturing order, and a display unit 25 for displaying the graph. An interval statistic calculating unit 21 performs a predetermined conversion so that the frequency distribution approximates a normal distribution with respect to the measured results of products, and calculates a confidence interval, by carrying out the inverse transformation of conversion correspondence confidence interval that corresponds to the measured results after the conversion. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a quality fluctuation display device and a quality fluctuation display method for displaying a quality fluctuation measured in a manufacturing process.

  Conventionally, in order to manage the quality of a product, a method for associating the characteristic value of the quality with work contents and production conditions has been proposed.

  For example, Patent Document 1 describes the correlation between device state data indicating the state of the manufacturing device and product data such as the yield and electrical characteristics of the processed product, based on the time processed by the manufacturing device. It is disclosed to seek.

  Patent Document 2 discloses that the production achievement rate in an injection molding machine, whether or not an abnormality has occurred, whether or not molding conditions have been changed, and quality data are displayed on a multi-screen on a common time axis.

  Further, Patent Document 3 discloses a method of displaying a history of quality data, manufacturing condition change contents, equipment abnormality contents, and countermeasures in alignment with the time axis.

Japanese Patent Application Laid-Open No. H10-228561 discloses that a correlation is analyzed by combining manufacturing process conditions and product inspection data.
JP-A-9-219347 (published August 19, 1997) Japanese Patent Laid-Open No. 2001-293761 (released on October 23, 2001) JP 2004-198148 (released July 15, 2004) JP 2004-186445 (released July 2, 2004) Motohisa Konno and Toshikatsu Hayashi, “Multivariate Data Utilization with JMP”, Kaibundo, June 2004, p29-32

  In the conventional configuration, manufacturing conditions and product quality data are managed on the basis of the time axis. However, in the case of a process in which individual products are produced sequentially with a manufacturing device, the manufacturing time of each product is due to the occurrence of abnormalities in the manufacturing process, countermeasures, fluctuations in manufacturing speed due to the environment, etc. Not evenly spaced. Similarly, the measurement time of the quality characteristic value of the product performed after a predetermined time from the manufacturing time is not equally spaced between the products manufactured sequentially.

  For this reason, when the variation of the quality characteristic value is displayed with respect to the time axis, information indicating the number of products measured within a certain time is lost.

  For example, the manufacturing apparatus has three equivalent manufacturing processing units, and these three manufacturing processing units discharge products one by one in order. Then, it is assumed that some trouble occurs in one manufacturing processing unit, and the quality characteristic value changes at a rate of one in three. That is, the quality characteristic value changes in three cycles. In such a case, in the conventional configuration, since the fluctuation of the quality characteristic value is displayed based on the time axis, and the information indicating the number of products is lost, the information of the period is also lost. End up. For this reason, it is impossible to immediately recognize that one of the manufacturing processing units has a problem.

  Furthermore, when the manufacturing device and the quality measuring device are different, the travel time from the manufacturing device to the quality measuring device occurs, so simply matching the time axis can associate the quality abnormality with the conditions of the manufacturing device. Can not.

  The present invention has been made in view of the above problems, and its purpose is to provide a quality fluctuation display device capable of reliably recognizing quality period information based on a number basis such that a defect occurs every predetermined number, A quality variation display method, a quality variation display program, and a recording medium on which the program is recorded are realized.

  In order to solve the above problems, the quality variation display device of the present invention is a quality variation display device that displays a predetermined quality variation in a plurality of products manufactured in a manufacturing facility, and is measured by a measurement facility. A quality data storage unit that stores the measurement results of each product in association with the manufacturing order, and reads out the measurement results and the manufacturing order in association with each other from the quality data storage unit, and a certain number of products that are continuous in the manufacturing order. A graph creation unit that shifts a section corresponding to a predetermined shift number, obtains a statistic of the measurement result for each section, and creates a graph showing the statistic at regular intervals in manufacturing order; and the graph creation unit, A display unit for displaying the created graph, and the statistics include an average value of measurement results of products included in each section, a median value of measurement results of products included in each section, and a product included in each section. The standard deviation or variance of the measurement results is at least one of the confidence intervals in the measurement results of the products included in each interval, and the statistic includes the confidence interval, and the graph creating means includes the product measurement results On the other hand, it is characterized in that the confidence interval is calculated by performing a predetermined conversion such that the frequency distribution approximates a normal distribution, and inversely converting the conversion corresponding confidence interval corresponding to the measurement result after the conversion.

  Further, it is preferable that the graph creating unit adds a line indicating at least one of an upper limit standard value and a lower limit standard value of the predetermined quality to the graph.

  In order to solve the above problems, the quality variation display method of the present invention is a quality variation display method of a quality variation display device for displaying a predetermined quality variation in a plurality of products manufactured in a manufacturing facility, The quality variation display device includes a quality data storage unit, a graph creation unit, and a display unit, and the quality data storage unit stores the measurement result of each product measured by the measurement facility in association with the manufacturing order. The data storage step and the graph creating means read out the measurement result and the manufacturing order in association with each other from the quality data storage unit, and the number of shifts corresponding to a certain number of products that are consecutive in the manufacturing order A graph creating step for creating a graph in which the statistics of the measurement results are obtained for each section, and the statistics are shown at regular intervals in the order of manufacture, and the graph creating means creates A display step for displaying the graph by the display unit, and the statistic includes an average value of measurement results of products included in each section, a median value of measurement results of products included in each section, and each section. At least one of the standard deviation or variance of the measurement result of the product to be measured, the confidence interval in the measurement result of the product included in each interval, and the statistic includes the confidence interval, and in the graph creation step, The graph creating means performs a predetermined conversion such that the frequency distribution approximates a normal distribution with respect to the measurement result of the product, and inversely converts the conversion corresponding confidence interval corresponding to the measurement result after the conversion, thereby the confidence interval. It is characterized by calculating.

  Each means in the quality variation display device can be executed on a computer by a quality variation display program. Furthermore, by storing the quality variation display program in a computer-readable recording medium, the quality variation display program can be executed on an arbitrary computer.

  The quality variation display device of the present invention includes a quality data storage unit that stores a measurement result of each product measured by a measurement facility in association with a manufacturing order, and the measurement result and the manufacturing order from the quality data storage unit. Read the data in correspondence with each other, shift the intervals corresponding to a certain number of products that are consecutive in the order of manufacture by a certain number of shifts, obtain the statistics of the measurement results for each interval, and show the statistics at regular intervals in the order of manufacture. And a display unit for displaying the graph created by the graph creation means, and the statistics are included in the average value of the measurement results of the products included in each section and in each section. The median of the measurement results of the products, the standard deviation or variance of the measurement results of the products included in each section, and the confidence interval in the measurement results of the products included in each section, and The metric includes the confidence interval, and the graph creating means performs a predetermined conversion such that the frequency distribution approximates a normal distribution with respect to the measurement result of the product, and a conversion corresponding confidence interval corresponding to the converted measurement result The confidence interval is calculated by inversely transforming.

  Therefore, by looking at the displayed graph, it is possible to recognize periodic variations in quality on the basis of the number, for example, a defect occurs at a rate of one for a certain number. Further, by analyzing the quality fluctuation graph, it is possible to confirm a plurality of frequency components of the periodic fluctuation according to various abnormal causes.

  An embodiment of the present invention will be described with reference to FIGS. 1 to 9 as follows. FIG. 2 shows a schematic configuration of a quality display system that displays the quality of a product manufactured from a certain material. As shown in FIG. 2, the quality display system 1 includes three first manufacturing equipment 3, second manufacturing equipment 4, third manufacturing equipment 6, and a first measuring equipment that measures quality characteristics of intermediate products being manufactured. 5, a second measurement facility 7 that measures the final quality characteristics of the manufactured product, and a quality variation display device that collects data from various facilities 3 to 7 and displays a variation in quality based on the collected data 10.

  In general, the product 8 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 manufacturing facility 3 and the second manufacturing facility 4 sequentially process the workpiece 2, the first measuring facility 5 then measures intermediate quality characteristics, and the third manufacturing facility 6 After the processing, the second measuring equipment 7 shall measure the quality characteristics of the product 8. The first manufacturing facility 3, the second manufacturing facility 4, the first measuring facility 5, the third manufacturing facility 6, and the second measuring facility 7 form a manufacturing line.

  The manufacturing equipment 3, 4, 6 performs a predetermined process on the plurality of workpieces 2 and sequentially manufactures a plurality of products 8. The manufacturing equipment 3, 4, 6 outputs production condition information indicating a change in production conditions, which is a condition of processing contents, to the quality variation display device 10. The production conditions include, for example, product specifications, classification of equipment such as molding dies.

  The product specification is a specification related to the product 8 to be manufactured, and is input to the first manufacturing facility 3 when the specification of the product 8 is changed. Here, the product specifications may be input to all the manufacturing facilities 3, 4, 6. However, since the product specifications are common to all the manufacturing facilities, it is only necessary to input the product specifications to the first manufacturing facility.

  In addition, the manufacturing facilities 3, 4, and 6 read a mold identification number for identifying each mold attached to the installed equipment (hereinafter referred to as a molding mold as an example). When the molding die is changed, the manufacturing equipment 3, 4, 6 outputs the die identification number read from the newly installed molding die to the quality variation display device 10. The mold identification number is represented by, for example, a bar code written on each mold. Then, the manufacturing equipment 3, 4, 6 recognizes the mold identification number corresponding to the installed mold by reading the barcode.

  The measuring facilities 5 and 7 measure the quality characteristics of the products manufactured in order by the manufacturing facilities 3, 4 and 6 in the same order, and output the measurement results to the quality variation display device 10. The quality characteristics of individual products are, for example, product dimensions, electrical characteristics, weight, and the like. Moreover, each of the measurement equipments 5 and 7 may measure a plurality of quality characteristics. For example, the measuring equipment 5 measures both the dimensions and electrical characteristics of the intermediate product.

  In general, a time lag occurs at a time when each facility 3 to 7 processes a certain work 2. That is, the time lag corresponds to the time required from the time of manufacture at each of the manufacturing facilities 3, 4, 6 to the time of measurement at each of the measuring facilities 5, 7. This time lag (required time) is hereinafter referred to as “dead time”. In the case shown in FIG. 2, there is a dead time Td1 (in this case, for example, 6 minutes) between the time when processing is started at the first manufacturing facility 3 and the time when processing is started at the second manufacturing facility 4. A dead time Td2 (in this case, for example, 3 minutes) occurs between the time when the process is started and the process is started at the second manufacturing equipment 4 and the time measured at the first measurement equipment 5, and the first measurement equipment A dead time Td3 (in this case, for example, 5 minutes) between the time measured at 5 and the time at which processing is started at the third manufacturing facility 6 occurs, and processing is started at the third manufacturing facility 6. A dead time Td4 (in this case, for example, 4 minutes) occurs between the time measured by the second measurement facility and the time measured by the second measurement facility. The dead time is determined by the installation conditions of the facilities 3 to 7.

  Next, the quality variation display device 10 will be described. As shown in FIG. 1, a quality fluctuation display device 10 includes a quality data acquisition unit (quality data acquisition unit) 11, a quality data input unit (quality data acquisition unit) 12, a quality data storage processing unit 13, a quality data storage DB. (Quality data storage unit) 14, production condition information acquisition unit (manufacturing data acquisition unit) 15, work content information input unit (manufacturing data acquisition unit) 16, manufacturing data storage processing unit 17, manufacturing data storage DB (manufacturing data storage unit) ) 18, timer 19, display parameter input unit 20, interval statistic calculation unit 21, time information addition unit 22, manufacturing data addition unit 23, causal relationship / dead time information storage DB (causal relationship information storage unit, required time storage unit) ) 24 and a display unit 25.

  The quality data acquisition unit 11 acquires the measurement result of each quality item measured from each product in the order of manufacture in the second manufacturing facility 4 from the first measurement facility 5 and the second measurement facility 7. Note that the quality item to be acquired is not necessarily one from each measurement facility. You may acquire the measured value regarding a some quality item from one measuring equipment. For example, the quality data acquisition unit 11 acquires the measurement values of the quality items 5a, 5b,... From the first measurement facility 5, and acquires the measurement values of the quality items 7a, 7b,.

  The quality data input unit 12 is for an operator to input measured values. Depending on the quality item, measurement with the human eye may be more accurate. For example, the presence or absence of an appearance defect of the product. In such a case, when the first measurement facility 5 or the second measurement facility 7 performs the measurement, an enlarged image of the product is displayed on the worker, and the worker measures the presence or absence of an appearance defect of each product. The quality data input unit 12 is used when inputting measurement results of quality items measured by such humans. In addition, the quality data input unit 12 receives the measurement result measured by the measurement facility having no communication function when a failure occurs in the communication between the measurement facility 5 or 7 and the quality variation display device 10. May be used to input.

  The quality data storage processing unit 13 performs processing for storing the measurement result and measurement time of each quality item in association with each other. Since each measuring equipment 5 and 7 measures the quality of each product in the order of manufacture, the quality data storage processing unit 13 stores the measurement results and measurement times of each quality item in association with each other, The measurement result of each quality item is stored in the order of manufacture.

  When the quality data acquisition processing unit 11 acquires the measurement result of each quality item from the first measurement facility 5 and the second measurement facility 7, the quality data storage processing unit 13 reads the acquired time from the timer 19 and measures the read time. The time is associated with the measurement result. And the quality data storage process part 13 stores the quality data which matched the measurement result and measurement time of each quality item in quality data storage DB14.

  Similarly, when the measurement result of each quality item is input to the quality data input unit 12, the quality data storage processing unit 13 reads the input time from the timer 19 and sets the read time as the measurement time. Correlate with. Then, the quality data storage processing unit 13 stores the quality data in which the measurement result is associated with the measurement time in the quality data storage DB 14.

  The quality data storage processing unit 13 stores the quality data corresponding to the measurement result related to the first measurement facility 5 and the quality data corresponding to the measurement result related to the second measurement facility 7 in the quality data storage DB 14 as separate tables. Store.

  The quality data storage DB (database) 14 stores the measurement results of the quality items of each product measured by the measuring equipment 5 and 7 in association with the manufacturing order. Specifically, the quality data storage DB (database) 14 receives the measurement results of the quality data storage unit and each quality item, the time when the measurement results were acquired (corresponding to the measurement time), or the measurement results. Quality data in which time (corresponding to measurement time) is associated is stored for each quality item.

  FIG. 3 is a diagram illustrating an example of quality data related to the first measurement facility 5 stored in the quality data storage DB 14. As shown in FIG. 3, the quality data storage DB 14 stores a measurement result and a measurement time in association with each quality item 5 a, 5 b,... Measured by the first measurement facility 5.

  When various production conditions are changed in each of the manufacturing facilities 3, 4, and 6, the production condition information acquisition unit 15 acquires production condition information indicating that fact and newly set production conditions. For example, when the molding die used in each of the manufacturing facilities 3, 4, 6 is changed, the production condition acquisition unit 15 sets the die identification number corresponding to the newly installed molding die to that effect. Acquired from manufacturing equipment 3, 4, 6.

  The work content information input unit 16 receives work content information related to the work content in each manufacturing facility. The work content includes, for example, failure repair when a failure occurs in the manufacturing facility, cleaning of tool dirt included in the manufacturing facility, adjustment of the position of the jig, and the like. The work content information is input to the work content input unit 16 by the worker at the end of the work.

  The manufacturing data storage processing unit 17 stores the manufacturing data in which the production condition information and the work content information are associated with the change of the production conditions and the generation time of the work.

  When the production condition information acquisition unit 15 acquires the production condition information from each of the manufacturing facilities 3, 4, and 6, the manufacturing data storage processing unit 17 reads the acquired time from the timer 19, and uses the read time as the production condition change occurrence time. The manufacturing data associated with the production condition information is stored in the manufacturing data storage DB 18.

  Similarly, when the work content information is input to the work content information input unit 16, the manufacturing data storage processing unit 17 reads the input time from the timer 19, and uses the read time as the work occurrence time. The manufacturing data is stored in the manufacturing data storage DB 18 in association with the information.

  The manufacturing data storage DB 18 stores the manufacturing data in which the production condition information indicating the change of the production conditions in the production equipment 3, 4, 6 is associated with the occurrence time of the change of the production conditions, or the contents of the work in the production equipment Manufacturing data in which the work content information shown is associated with the time of occurrence of the work is stored for each of the manufacturing apparatuses 3, 4, and 6.

  FIG. 4 is a diagram illustrating an example of manufacturing data related to the first manufacturing apparatus 3 stored in the manufacturing data storage DB 18. As shown in FIG. 4, the manufacturing data storage DB 18 stores, as production condition information, that the molding die has been changed to “# 3” at time 09:36:08. Further, the manufacturing data storage DB 18 stores, as work content information, that the position of the jig has been adjusted at time 10:04:56 and that the tool dirt has been cleaned at time 13:11:32. is doing.

  The display parameter input unit 20 inputs various parameters of quality data to be displayed on the display unit 25.

  Quality data parameters include both quality item type, display range (display range specifying time (for example, recent 8 hours), and display range specifying product number (for example, recent product number of 10,000). Included), the number of sections in the quality variation graph to be displayed (hereinafter referred to as section width) and the number of shifts in the section, the type of statistic to be displayed in each section, and the upper and lower standard values indicating the non-defective range of measurement results Etc.

  The statistics are, for example, the average value of the measurement results of the products included in each section, the median value of the measurement results of the products included in each section, the standard deviation σ or the variance of the measurement results of the products included in each section Alternatively, 6 × σ is a confidence interval (for example, average value ± 3 × standard deviation) in the measurement result of the product included in each interval.

Here, the confidence interval is a probability function Pr (x ₁ <x <x ₂ ) related to the quality data x when 1-α indicating a probability set arbitrarily in advance is given as 1-α. This is an interval (x ₁ , x ₂ ). Here, the probability function Pr (x ₁ <x <x ₂ ) indicates the probability that the quality data x is included in the range from x ₁ to x ₂ in the frequency distribution of the quality data x.

  For example, when the reliability 1-α is 99.74% and the quality data x follows a normal distribution, it is known that the confidence interval is (average value−3 × standard deviation, average value + 3 × standard deviation). ing.

  In the present embodiment, the interval statistic calculation unit 21 assumes that the frequency distribution of quality data approximates a normal distribution, and has a confidence interval (average value−3 × standard deviation, average) with a reliability of 99.74%. Value + 3 × standard deviation).

  The section statistic calculation unit 21 shifts a section corresponding to a certain number of sections by a certain number of shifts according to the parameters input to the display parameter input unit 20, obtains a statistic for each section, and Create a quality variation graph that shows statistics at regular intervals in manufacturing order.

  In other words, the section statistic calculation unit 21 shifts a section corresponding to a certain number of section products by a certain number of shifts according to the parameters input to the display parameter input unit 20, and obtains statistics for each section. It includes a quantity calculation means (not shown) and a statistic plot means (not shown) for creating a graph in which the statistics are plotted at regular intervals in the order of manufacture.

  Specifically, the interval statistic calculation unit 21 reads quality data corresponding to the quality item type input to the display parameter input unit 20 from the quality data storage DB 14.

  Further, the interval statistic calculation unit 21 outputs the quality data included in the display range input to the display parameter input unit 20 according to the interval width and the number of shifts input to the display parameter input unit 20 in the order of manufacture. A plurality of sections having a width and shifted by the number of shifts are allocated. Further, the section statistic calculation unit 21 calculates the statistic of the measurement result of the quality data included in each section according to the type of the statistic input to the display parameter input unit 20.

  Then, the section statistic calculation unit 21 creates a quality variation graph in which the statistics calculated for each section are plotted at equal intervals in the order of measurement time (that is, in the order of manufacture). That is, the statistics of each section are plotted on the number reference axis with the number of shifts as one unit.

  At this time, the interval statistic calculation unit 21 adds lines indicating the upper limit standard value and the lower limit standard value input to the display parameter input unit 20 to the quality variation graph.

  Then, the interval statistic calculation unit 21 outputs the created quality variation graph and the measurement time corresponding to the central product (product measured in the middle of the interval) in each interval to the time information addition unit 22.

  For example, if the products corresponding to the display range are products 1 to 10000, the section width is 50, and the number of shifts is 25, the section statistic calculation unit 21 starts from 1 to 50 and 26. 75, 51 to 100,..., 9551 to 10000 are determined as products included in each section to be displayed.

  In addition, when the statistic input to the display parameter input unit 20 is an average value, the section statistic calculation unit 21 includes products included in each section (the first to 50th products in the case of the first section). The average value of the measurement results is calculated.

  The interval statistic calculation unit 21 outputs the measurement time corresponding to the central product in each interval (for example, the 25th product in the case of the first interval) to the time information adding unit 22.

  The time information adding unit 22 adds a measurement time axis to the quality variation graph created by the section statistic calculating unit 21. Thereby, the relationship between each statistic and measurement time can be confirmed.

  Here, it is preferable to make the scale intervals of the measurement time axis equal. Thereby, progress of measurement time can be recognized easily.

  The time information addition unit 22 stores in advance a correspondence table between a predetermined reference range and the scale interval of the measurement time axis, and the display range of the quality variation graph created by the reference range and the interval statistic calculation unit 21 And the scale interval of the measurement time axis to be added is determined. For example, the time information adding unit 22 has a reference range: T1 (1 hour) or less and a scale interval of 5 minutes, a reference range T1 to T2 (3 hours) and a scale interval of 20 minutes, and a reference range T2 to T3 (10 hours) and a scale. An interval of 1 hour is stored in advance. For example, when the time corresponding to the display range of the quality variation graph created by the section statistic calculation unit 21 is 8 hours, the time information addition unit 22 determines the scale interval of the measurement time axis as 1 hour.

  The time information adding unit 22 calculates position coordinates on the number reference axis of the quality variation graph corresponding to the determined time on the measurement time axis. Specifically, the time information adding unit 22 is an approximation indicating the correspondence between the position coordinates on the number reference axis and the measurement time based on the measurement time of the central product in each section output from the section statistic calculation unit 21. Find the formula. Then, the time information adding unit 22 calculates position coordinates on the number reference axis corresponding to the determined time on the measurement time axis. Then, the time information adding unit 22 generates time data in which the time on the scale of the measurement time axis is associated with the position coordinates on the number reference axis.

  FIG. 5 is a diagram illustrating an example of time data generated by the time information adding unit 22. As shown in FIG. 5, the time information adding unit 22 has a position coordinate on the number reference axis in the first measuring equipment 5 corresponding to 9:00:00 on the scale of the measurement time axis. Is generated.

  Further, the time information adding unit 22 generates a measurement time axis with a scale at intervals of a fixed time, based on the generated time data and an approximate expression indicating the correspondence between the position coordinates on the number reference axis and the measurement time. Then, the generated quality variation graph to which the measurement time axis is added is output to the manufacturing data adding unit 23.

  The causal relationship / dead time information storage DB 24 stores the presence / absence of a causal relationship between each manufacturing facility 3, 4 and 6 and each quality item. If the causal relationship exists, the processing start time and quality of the manufacturing facility are further stored. A dead time indicating a time lag with respect to the data measurement time is stored. It is assumed that there is a causal relationship when the processing in the manufacturing facilities 3, 4 and 6 affects the quality data, and there is no causal relationship when there is no effect. The presence / absence of the causal relationship is determined in advance based on the processing contents of the manufacturing facilities 3, 4 and 6 and the measurement items. Further, the dead time is also determined in advance at the time of designing the production line.

  FIG. 6 shows an example of storage in the causal relationship / dead time information storage DB 24. In FIG. 6, “−1” indicates the causal relationship “none”, and the numbers other than “−1” indicate the causal relationship “present” and the dead time. As shown in FIG. 6, for example, in the causal relationship / dead time information storage DB 24, the first manufacturing facility 3 and the third manufacturing facility 6 have a causal relationship “no” with respect to the measurement item 5 a in the first measurement facility 5. And the fact that the second manufacturing facility 4 is “causal” and its dead time is “3 minutes”. Further, the causal relation / dead time information storage DB 24 stores the causal relation / dead time relating to each of the measurement facilities 5 and 7 in a separate table, as shown in FIGS. 6 (a) and 6 (b). Thereby, the causal relationship and dead time of each measuring equipment 5 and 7 can be read easily.

  The manufacturing data adding unit 23 adds manufacturing data to the quality variation graph output from the time information adding unit 22 based on the information stored in the causal relationship / dead time information storage DB 24.

  The manufacturing data adding unit 23 reads the name of the manufacturing equipment having a causal relationship with the quality item included in the quality variation graph and the dead time of the manufacturing equipment from the causal relation / dead time information storage DB 24. Then, the manufacturing data adding unit 23 reads the manufacturing data corresponding to the read manufacturing equipment name from the manufacturing data storage DB 18 and adds the manufacturing data to the quality variation graph. At this time, the manufacturing data adding unit 23 generates a manufacturing time axis in which the scale is adjusted by subtracting the dead time read from the causal relationship / dead time information storage DB 24 from the scale of the measurement time axis, and the generated manufacturing time axis Is added to the quality variation graph.

  Further, the manufacturing data adding unit 23 adds production condition information and work content information on an axis having a manufacturing time axis corresponding to the occurrence time included in the manufacturing data. Then, the manufacturing data adding unit 23 causes the display unit 25 to display a quality variation graph to which the measurement time axis, the manufacturing time axis, the production condition information, and the work content information are added.

  Thereby, work content information and / or production condition information can be associated with the measurement result of the quality item without considering the dead time.

  Note that the manufacturing data adding unit 23 reads only the manufacturing data related to the change of the product specification from the manufacturing data corresponding to the first manufacturing facility 3 regardless of the causal relationship, and the manufacturing corresponding to the name of the manufacturing facility having the causal relationship. Similar to the data, the product specification change information may be added to the quality variation graph. Thereby, it is possible to recognize the time point when the product specification is changed.

  The display unit 25 displays quality data, and is constituted by, for example, a liquid crystal panel.

  Next, a processing flow and a display example in the quality variation display device 10 will be described. FIG. 7 is a flowchart showing the flow of processing in the quality fluctuation display device 10. 8 and 9 show display examples on the display unit 25. FIG.

  First, the quality data acquisition unit 11 acquires a measurement result for each quality item from the first measurement facility 5 and the second measurement facility 7. Alternatively, the measurement result of the intermediate product is input to the quality data input unit 12 by the worker. Here, it is assumed that the quality data acquisition unit 11 acquires the measurement result. Then, the quality data storage processing unit 13 reads the time when the quality data acquisition unit 11 acquires the measurement result from the timer 19 as the measurement time, and creates quality data in which the read measurement time is associated with the measurement result. . Further, the quality data storage processing unit 13 stores the created quality data in the quality data storage DB 14 (S1).

  Next, the display parameter input unit 20 receives input of various parameters relating to quality data to be displayed (S2).

  Subsequently, the section statistic calculation unit 21 reads quality data corresponding to the quality item type input to the display parameter input unit 20 from the quality data storage DB 14.

  Then, the interval statistic calculation unit 21 outputs the quality data included in the display range input to the display parameter input unit 20 according to the interval width and the number of shifts input to the display parameter input unit 20 in the order of manufacture. A plurality of sections having a width and shifted by the number of shifts are allocated.

  Further, the section statistic calculation unit 21 calculates the statistic of the measurement result of the quality data included in each section according to the type of the statistic input to the display parameter input unit 20. Then, the section statistic calculation unit 21 creates a quality variation graph in which the statistics calculated for each section are plotted at regular intervals in the order of measurement time (that is, in the order of manufacture) (S3). That is, the statistics of each section are plotted on the number reference axis with the number of shifts as one unit.

  At this time, the interval statistic calculation unit 21 adds the upper limit standard value and the lower limit standard value input to the display parameter input unit 20 to the quality variation graph. Thereby, the trend of increase / decrease of defective products can be easily recognized.

  The interval statistic calculation unit 21 outputs the created quality variation graph and the measurement time of the central product in each interval of the number reference axis of the quality variation graph to the time information adding unit 22.

  FIG. 8 is a diagram illustrating an example of a quality variation graph created by the section statistic calculation unit 21.

  The upper part of FIG. 8 shows interval statistics when quality item: 5b, display range: 8 hours, interval width: 1, shift number: 1, and statistic type: average value are input to the display parameter input unit 20. An example of the quality fluctuation graph created by the quantity calculation unit 21 is shown. As shown in the upper part of FIG. 8, the section statistic calculation unit 21 creates a number reference axis with one shift number as the section interval, and the measurement result of the quality item 5b of each product on the number reference axis. (Since there is one product included in the section width, the measurement result of the product corresponds to the average value of the section) is plotted as it is. As a result, the quality variation graph created by the interval statistic calculation unit 21 plots the measurement results of each product at a constant number (here, one) at equal intervals. This makes it possible to immediately confirm the number-based periodic fluctuations in which the quality varies.

  In the upper part of FIG. 8, there are scales that divide the illustrated scale interval into five sections, but the scales are not shown for the sake of simplicity.

  On the other hand, in the middle part of FIG. 8, the display parameter input unit 20 stores the quality item: 5b, display range: 8 hours, interval width: 50, shift number: 25, statistic type: average value / confidence interval (here, An example of the quality variation graph created by the interval statistic calculation unit 21 when the average value ± 3 × standard deviation) · 6 × standard deviation is input is shown.

  As shown in the figure, the section statistic calculation unit 21 creates a number reference axis with 50 section widths and 25 shifts according to the parameters input to the display parameter input unit 20. Further, when the product corresponding to the display range (8 hours in this case) input to the display parameter input unit 20 is the product from No. 1 to No. 550, the interval statistic calculation unit 21 is the No. 1 to No. 50, It is determined that the product is included in each section assigned No. 26 to No. 75, No. 51 to No. 100,..., No. 501 to No. 550. The section statistic calculation unit 21 calculates the average calculated from the measurement result of the quality item 5b in the first section (1st to 50th products), the second section (26th to 75th products),. Values / confidence intervals (in this case, mean value ± 3 × standard deviation) and 6 × standard deviation are plotted at regular intervals in the order in which they were manufactured (that is, in the order of the first interval, the second interval,...). To do.

  As a result, the average value / confidence interval (in this case, average value ± 3 × standard deviation) and 6 × standard deviation obtained from the products included in each interval are set at regular intervals (25 in this case) at regular intervals. Plotted. As a result, the quality variation graph can confirm the number-based periodic variation as in the upper part of FIG.

  Also, the quality fluctuation graph as shown in the middle part of FIG. 8 has a larger number of sections and shifts than the upper part of FIG. 8, so that noise fluctuations due to measurement errors and components due to short period fluctuations on the number basis are removed. can do. As a result, it is possible to easily confirm the presence or absence of quality fluctuations caused by a sudden event (for example, a mounting position shift at the time of cleaning tool dirt). In addition, an increase / decrease in the defect rate can be confirmed by including a confidence interval of mean value ± 3 × standard deviation in the quality fluctuation graph on the assumption that the quality data approximates a normal distribution.

  Further, the lower part of FIG. 8 shows a section when quality item: 5b, display range: 8 days, section width: 500 pieces, shift number: 250 pieces, and statistic type: average are inputted to the display parameter input unit 20. An example of a graph created by the statistic calculator 21 is shown.

  In the figure, “average” (solid line in the figure) indicates an average value in each section when the section width is 500 pieces and the shift number is 250 pieces. “Average (for comparison)” (dotted line in the figure) is an average value in each section when the section width is 50 and the number of shifts is 25 (that is, the average value in the middle of FIG. 8). In order to clarify the effect of the broken line, it is added as a comparison.

  As shown in the figure, by increasing the section width and the number of shifts, the tendency of long-term quality fluctuation can be easily confirmed. Thereby, for example, it is possible to easily understand the relationship between long-term equipment fluctuations such as wear of the conveyor belt in each of the manufacturing equipments 3, 4 and 6 and quality fluctuations.

  In order to check the quality fluctuation due to an unexpected event such as the exchange of the conveyor belt, as shown in the middle part of FIG. 8, a quality fluctuation graph having a relatively short section width and the number reference axis of the number of shifts can be obtained. desirable.

  Furthermore, when exchanging consumables such as exchanging the conveyor belt, a quality fluctuation graph having a long section width and a number reference axis of the number of shifts is displayed as shown in the lower part of FIG. It is preferable to make it. Thereby, the quality fluctuation | variation according to the wear degree of a consumable can be confirmed.

  Next, the time information adding unit 22 adds a measurement time axis, which is time information, to the quality variation graph created by the section statistic calculating unit 21 (S4). Specifically, the time information addition unit 22 determines the scale interval of the measurement time axis to be added based on the display range of the graph created by the section statistic calculation unit 21 and a predetermined reference range. For example, when the display range of the graph created by the section statistic calculation unit 21 corresponds to 8 hours, the time information addition unit 22 determines the scale interval of the measurement time axis as 1 hour.

  Then, the time information adding unit 22 obtains an approximate expression indicating the correspondence between the position coordinates on the number reference axis and the measurement time based on the measurement time of the central product in each interval output from the interval statistic calculation unit 21. Then, position coordinates on the number reference axis corresponding to the scale time on the measurement time axis are calculated. The time information adding unit 22 generates time data in which the determined time interval of the measurement time axis is associated with the position coordinates on the number reference axis.

  Further, the time information adding unit 22 performs measurement having a scale with a fixed time interval based on the generated time data and an approximate expression indicating the correspondence between the position coordinates on the axis having the number reference scale and the measurement time. Generate a time axis. Then, the time information adding unit 22 adds the generated measurement time axis to the quality variation graph created by the section statistic calculating unit 21.

  Part B in FIG. 9 shows an example of the measurement time axis added by the time information adding unit 22. As shown in the drawing, the lengths of the scales of the measurement time axis added by the time information adding unit 22 are not equal intervals. For example, the length of the scale interval from time 10:00 to 11:00 in the figure is longer than the length from time 12:00 to 13:00. The magnitude of the scale interval length on the measurement time axis indicates whether or not the production line is retained.

  That is, when the length of the scale on the measurement time axis is small, the number of measurements per unit time is small, and the production line stays. On the other hand, when the length of the scale on the measurement time axis is large, the number of measurements per unit time is large, and the production line flows smoothly. Thus, the presence or absence of the production line can be easily confirmed by looking at the scale interval length of the measurement time axis.

  In parallel with the processes of S1 to S4, the process of S5 is performed. That is, the production condition information acquisition unit 15 acquires, from the first manufacturing equipment 3, the second manufacturing equipment 4, and the third manufacturing equipment 6, for example, production condition information related to changes in product specifications and molding dies. Further, the work content information input unit 16 receives work content information in each of the manufacturing facilities 3, 4, 6 from the worker.

  Then, the manufacturing data storage processing unit 17 reads the time when the production condition information acquisition unit 15 acquires the production condition information from the timer 19 and creates manufacturing data in which the read time is associated with the production condition information. Similarly, the manufacturing data storage processing unit 17 reads the time when the work content information is input to the work content information input unit 16 from the timer 19 and creates manufacturing data in which the read time is associated with the work content information. To do. The manufacturing data storage processing unit 17 stores the created manufacturing data in the manufacturing data storage DB 18.

  Subsequently, the manufacturing data adding unit 23 performs a process of adding manufacturing data to the quality variation graph to which the measurement time axis is added (S6). Specifically, the manufacturing data adding unit 23 reads out the name of the manufacturing equipment and the dead time having a causal relationship with the quality item displayed in the quality fluctuation graph from the causal / waste time information storage DB 24. For example, the manufacturing data adding unit 23 obtains “first manufacturing equipment 5” as the name of the manufacturing equipment having a causal relationship with the quality item 5b from the causal / dead time information storage DB 24 as shown in FIG. Read 9 minutes of dead time of the first manufacturing equipment 5.

  Further, the manufacturing data adding unit 23 reads manufacturing data of the manufacturing equipment corresponding to the read manufacturing equipment name from the manufacturing data storage DB 18. For example, the manufacturing data adding unit 23 reads out manufacturing data as shown in FIG. 4 as manufacturing data corresponding to the read manufacturing equipment name “first manufacturing equipment 5”.

  The manufacturing data adding unit 23 generates a manufacturing time axis having a scale obtained by subtracting the dead time read from the causal relation / dead time information storage DB 24 from the scale of the measurement time axis, and adds the generated manufacturing time axis to the quality variation graph. To do.

  Further, the manufacturing data adding unit 23 adds the manufacturing data read from the manufacturing data storage DB 18 on the manufacturing time axis. For example, when manufacturing data as shown in FIG. 4 is read out, the manufacturing data adding unit 23 changes the molding die to “# 3” at the time of occurrence 09:36:08 on the manufacturing time axis. Is added at the time of occurrence 10:04:56 and “tool dirt cleaning” is performed at the time of occurrence 13:11:32.

  The manufacturing data adding unit 23 causes the display unit 25 to display a quality variation graph to which the measurement time axis, the manufacturing time axis, the production condition information, and the work content information are added.

  Part C in FIG. 9 shows an example of the measurement time axis, the manufacturing time axis, the production condition information, and the work content information added by the manufacturing data adding unit 23. As shown in the figure, for example, the manufacturing data adding unit 23 generates a manufacturing time axis having a scale obtained by subtracting 9 minutes of dead time from the scale of the measurement time axis, and on the manufacturing time axis, as shown in FIG. Production condition information and work content information included in such manufacturing data are added.

  In addition, according to the production condition information indicating the change of the molding die and product specification, it can be recognized that the same molding die and product specification are applied to the section until the next change. Therefore, as shown in the drawing, the manufacturing data adding unit 23 performs section display in which sections before and after the change time are divided with respect to product specifications and molding dies that are production condition information. This makes it easy to understand what production conditions are applied in which section.

  Further, as shown in the figure, the manufacturing data adding unit 23 mainly expresses work content information that occurs suddenly by characters representing the work content information and an arrow indicating the time when the work content occurs. ing. This makes it easy to understand what kind of work has occurred at what point.

  Further, by displaying the production condition information and the work content information on the production time axis having a scale obtained by subtracting 9 minutes of dead time from the scale of the measurement time axis, as shown in part C in FIG. It is possible to immediately understand the relationship between the work content information and the quality variation. As a result, it is possible to easily find out which production condition changes or work contents cause the quality variation. For example, in FIG. 9, it can be determined that the work content “cleaning of tool stains” is a strong candidate as a cause of the increase in defects.

  In addition, the manufacturing data adding unit 23 adds only manufacturing data relating to the manufacturing equipment 3, 4, 6 having a causal relationship with the quality item of the quality variation graph to be displayed. For this reason, unnecessary information that is not causally related to the quality variation is not displayed, and the cause of the quality variation can be easily pursued.

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

  For example, in the present embodiment, three manufacturing facilities 3, 4, 6 and two measuring facilities 5, 7 are used.

  Further, the manufacturing data adding unit 23 adds a manufacturing time axis. However, the manufacturing data adding unit 23 does not add the manufacturing time axis, adjusts the generation time by adding the dead time to the generation time of the manufacturing data, and adjusts the generation time axis corresponding to the adjusted generation time. Production condition information or work content information may be added to the coordinate position. That is, the manufacturing data adding unit 23 adjusts the generation time of the manufacturing data based on the dead time stored in the causal relation / dead time information storage DB 24, and based on the adjusted generation time and the measurement time axis, Production condition information or work content information corresponding to the time of occurrence may be added to the graph. Thus, by looking at the quality variation graph, the relationship between the production condition information or work content information and the quality variation can be confirmed without considering the dead time.

  In the present embodiment, the quality data storage processing unit 13 and the manufacturing data storage processing unit 17 set the measurement time, the production condition change, and the work occurrence time by the same timer 19. As a result, the correspondence between the measurement time included in the quality data and the occurrence time included in the manufacturing data becomes accurate. However, the timer 19 exists outside the quality variation display device 10, and each measuring device 5 or 7 reads the measurement time of each product from the timer 19, and outputs the measurement result and the measurement time to the quality variation display device 10. Also good. Similarly, when there is a change in production conditions, each manufacturing facility 3, 4, 6 reads the occurrence time from the timer 19, and outputs the production condition information and the occurrence time to the quality variation display device 10. Also good.

  In the present embodiment, the time information adding unit 22 calculates an approximate expression indicating the correspondence between the position coordinates on the number reference axis and the measurement time based on the measurement time of the central product in each section. The information adding unit 22 may obtain the approximate expression based on the average (average measurement time) of the measurement times of the first product in each section or the products included in each section. The measurement time representing each section may be calculated based on the same reference in each section.

  In addition, the average of the measurement time of the product included in each section shall be calculated by the section statistics calculation part 21.

  In the above embodiment, the interval statistic calculation unit 21 assumes that the frequency distribution of the quality data approximates a normal distribution, and the confidence interval (average value−3 × standard deviation) corresponding to the reliability of 99.74%. , Average value + 3 × standard deviation) has been described.

  However, the reliability may be a value arbitrarily set in advance, for example, 99.5%.

Further, when the frequency distribution of the quality data x does not approximate a normal distribution, the mean value x _ave of said quality data _x, confidence intervals calculated from the standard deviation x _sigma (e.g., for a confidence level 99.74% , The confidence interval (x _ave −3 × x _σ , x _ave + 3 × x _σ ) does not correspond to a desired reliability (for example, 99.74%), which will be described below with reference to FIG. Explained.

The upper part of FIG. 10 shows the frequency distribution of the quality data x of these 50 products. As shown in the figure, the frequency distribution of the quality data x has a shape in which the tail is widened toward the higher value of x and does not approximate the normal distribution. In such a case, even if the interval statistic calculation unit 21 calculates the confidence interval using the average value x _ave and the standard deviation x _σ of the quality data x, the calculated confidence interval corresponds to the desired reliability. There is no confidence interval.

For example, it is assumed that (x _ave −3 × x _σ , x _ave + 3 × x _σ ) is calculated as the confidence interval corresponding to the reliability of 99.74%. In the upper part of FIG. 10, x _ave and x _ave ± 3 × x _σ are indicated by alternate long and short dash lines. As shown in the upper part of FIG. 10, since the frequency distribution of the quality data x has a shape in which the tail is expanded on the right side, it can be seen that x _ave is shifted to the right side from the peak position. In addition, the number having a larger value than x _ave + 3 × x _σ is large, and the confidence interval (x _ave −3 × x _σ , x _ave + 3 × x _σ ) does not correspond to the desired reliability of 99.74%. I understand.

  This is because the quality data x does not approximate a normal distribution. The fact that the confidence interval (average value−3 × standard deviation, average value + 3 × standard deviation) corresponds to the reliability of 99.74% is based on the assumption that the frequency distribution approximates a normal distribution.

  Therefore, the interval statistic calculation unit 21 uniformly calculates (average value−3 × standard deviation, average value + 3 × standard deviation) as a reliability interval corresponding to the reliability of 99.74% for all quality data. In this case, for the quality data that does not approximate the normal distribution, a confidence interval greatly deviating from the original is displayed. As a result, for example, the manufacturing manager may make a wrong determination that the confidence interval is close to the lower limit standard value and that there is room for the upper limit standard value.

  Therefore, the interval statistic calculation unit 21 performs conversion of quality data (variable conversion) so as to approximate a normal distribution, calculates an average value and various statistics for the converted quality data, and calculates the calculated average value. In addition, it is preferable to obtain an average value of quality data and various statistics by performing inverse transformation on the statistics.

  Here, power conversion will be described as an example of the variable conversion (see page 30 of Non-Patent Document 1). The power conversion is used for the purpose of converting a symmetric and distorted frequency distribution into a symmetric frequency distribution.

  Below, the calculation procedure of the average value and various statistics which used the power transformation in the area statistics calculation part 21 is demonstrated.

The quality fluctuation display device 10 includes a conversion method storage unit that stores a value of an exponent part in power conversion determined in advance based on a distribution trend and physical properties of each quality data. The conversion method storage unit stores the quality data item and the value of the exponent part suitable for the quality data in association with each other. Here, as an example, it is assumed that the conversion method storage unit stores “½” as the value of the exponent part for the quality data x. This indicates that the variable y = f (x) = x ^1/2 after power transformation approximates a normal distribution.

First, the interval statistic calculation unit 21 performs power conversion on the quality data x that does not approximate the normal distribution according to the exponent value (1/2 in this case) read from the conversion method storage unit. The frequency distribution of the subsequent variable y = f (x) = x ^1/2 is obtained.

  The lower part of FIG. 10 is a graph showing the frequency distribution of the variable y after conversion. As shown in the figure, the frequency distribution of the variable y approximates a symmetric normal distribution.

Next, the interval statistic calculation unit 21 calculates an average value y _ave for the variable y and various statistics (standard deviation y _σ , confidence interval (y ₁ , y ₂ ), etc.) from the frequency distribution of the variable y.

Here, since the frequency distribution of the variable y follows a normal distribution, the interval statistic calculation unit 21 sets y ₁ = y _ave −3 × for the confidence interval (y ₁ , y ₂ ) for the reliability 99.74%, for example. y _σ , y ₂ = y _ave + 3 × y _σ can be easily calculated.

As shown in the lower part of FIG. 10, the confidence interval (y ₁ , y ₂ ) calculated by the interval statistic calculation unit 21 substantially corresponds to the reliability of 99.74%.

Thereafter, the interval statistic calculation unit 21 sets the average value and the statistic calculated with respect to the variable y as the average value and the statistic of the quality data x, obtained by performing an inverse transformation f ⁻¹ . For example, the interval statistic calculation unit 21 sets f ⁻¹ (y _ave ) = y _ave ² as the average value of the quality data x. The interval statistic calculation unit 21 sets (f ⁻¹ (y ₁ ), f ⁻¹ (y ₂ )), that is, (y ₁ ² , y ₂ ² ), as the confidence interval of the quality data x.

  In the upper part of FIG. 10, the broken lines indicate the average value calculated by the interval statistic calculation unit 21 by inverse transformation and the upper and lower limits of the confidence interval. As shown in the drawing, the average value calculated by the interval statistic calculation unit 21 by the inverse transformation is located near the peak of the frequency distribution of the quality data x. Furthermore, the confidence interval calculated by the interval statistic calculation unit 21 using the inverse transformation is wide on the right side (the side where the value of the quality data x is large) and the left side (the side where the value of the quality data x is small). It can be seen that the width is narrower and substantially corresponds to the desired reliability of 99.74%.

  FIG. 11 shows a graph in which the quality data x of the 50 products shown in FIG. In FIG. 11, the alternate long and short dash line indicates an average value and a confidence interval (corresponding to a reliability of 99.74%) obtained from the frequency distribution of the quality data x when it is assumed that the frequency distribution of the quality data x approximates a normal distribution. To do). On the other hand, as described above, the broken line indicates that the quality data x is once converted to the variable y, and the average value obtained from the frequency distribution of the variable y according to the normal distribution and the limit of the confidence interval are inversely converted. And the upper and lower limits of the confidence interval.

  As also shown in FIG. 11, the number deviating from the confidence interval when calculated using variable transformation and inverse transformation is smaller than the number deviating from the confidence interval when the quality data x is assumed to approximate a normal distribution. I understand that. This is because the probability included in the confidence interval calculated using variable transformation and inverse transformation is closer to the desired reliability.

  In this way, by displaying the quality fluctuation graph in which the average value and the statistic calculated using the variable transformation and the inverse transformation are displayed for the product having the section width of 50, the manufacturing manager can It becomes possible to judge the fluctuation tendency more accurately.

  Here, power conversion has been described with respect to variable conversion and inverse conversion, but the conversion method is not limited to this. Based on the physical properties of each quality data, a conversion method is determined in advance so that the frequency distribution of the converted variables is closer to a normal distribution, and the conversion method is stored in association with the quality data item. Store it in the department. Then, the interval statistic calculation unit 21 may read the conversion method corresponding to the quality data from the conversion method storage unit, and obtain the average value and various statistics according to the read conversion method.

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

  That is, the quality variation display device 10 is a storage device such as a CPU that executes instructions of a control program that realizes 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 a tape system such as a magnetic tape and a cassette tape, a disk system including a magnetic disk such as a flexible disk / hard disk and an optical disk such as a CD-ROM / MO / MD / DVD / CD-R, and an IC card. 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 variation display device 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.

  As described above, in order to solve the above-described problem, the quality variation display device of the present invention is a quality variation display device that displays a predetermined quality variation in a plurality of products manufactured in a manufacturing facility, and is measured. A quality data storage unit that stores the measurement results of each product measured by the equipment in association with the manufacturing order, and shifts the intervals corresponding to a certain number of products that are consecutive in the manufacturing order by a certain number of shifts. A graph creation unit that obtains a statistic and creates a graph showing the statistic at regular intervals in the manufacturing order, and a display unit that displays the graph created by the graph creation unit are provided.

  In order to solve the above problems, the quality fluctuation display method of the present invention is a quality fluctuation display method of a quality fluctuation display device that displays a predetermined quality fluctuation in a plurality of products manufactured in a manufacturing facility. , A quality data storage step for storing the measurement results of each product measured by the measuring equipment in association with the manufacturing order, and shifting a section corresponding to a certain number of products consecutive in the manufacturing order by a certain number of shifts. The method includes a graph creation step of obtaining a statistic for each and creating a graph showing the statistic at regular intervals in the order of manufacture, and a display step of displaying the graph created by the graph creation means.

  According to the above configuration or method, the intervals corresponding to a certain number of products that are consecutive in the order of manufacture are shifted by a certain number of shifts, the statistics are obtained for each interval, and the statistics are shown at regular intervals in the order of manufacture. Create a graph. That is, the statistics corresponding to each section are shown at regular intervals. Also, a certain section and its adjacent section are shifted by a certain number of shifts. Therefore, the interval between the statistics shown corresponds to a certain number of shifts. As a result, the statistics are in a state of being plotted on an axis with a constant number of equal intervals. As a result, by viewing the graph displayed on the display unit, it is possible to recognize periodic fluctuations in quality on the basis of the number, for example, a defect occurs at a rate of one for a certain number.

  In addition, by analyzing the displayed graph, it is possible to confirm a plurality of frequency components of the above-described periodic fluctuation according to various causes of abnormality.

  Here, the statistic is, for example, the measurement result of the product number itself when the number of sections is one, and the average value or median value of these measurement results when there are a plurality of section numbers. is there.

  In addition, when the number of sections is one and the number of shifts is one, the graph created by the graph creating means shows the measurement results of the predetermined quality of each product at regular intervals in the manufacturing order. It becomes a graph.

  Further, when the number of sections is plural and the statistic is an average value of the measurement results corresponding to the products included in the sections, the graph created by the graph creating means is the predetermined number of each product arranged in the order of manufacture. It shows the moving average of quality.

  The number of shifts refers to the manufacturing order of the product manufactured first among the products included in a certain section, and the manufacturing order of the product manufactured first among the products included in the section adjacent to the section. This is equivalent to the difference.

  Furthermore, in addition to the above-described configuration, the quality variation display device of the present invention is such that the measurement equipment measures the predetermined quality in the order of manufacture, and the quality data storage unit displays the measurement result as a measurement time. The measurement time axis is created based on the measurement time stored in the quality data storage unit, and the measurement time axis is added to the graph created by the graph creation means. It is characterized by comprising time information adding means.

  According to said structure, while being able to confirm the statistic corresponding to the said predetermined quality at a fixed number interval, it can confirm visually the measurement time progress of each product. For example, it is possible to specify the measurement time when a large quality variation has occurred on the graph.

  Furthermore, in addition to the above-described configuration, in the quality variation display device of the present invention, it is preferable that the time information adding unit adds a scale of a certain time interval to the measurement time axis.

  The interval between the scales is, for example, 1 hour or 10 minutes and is not limited. Moreover, the display form of the scale is, for example, a line intersecting the axis or a line toward the inside of the graph, and is not limited.

  According to said structure, the magnitude | size of the scale space | interval length of a measurement time axis will represent the presence or absence of a retention in a manufacturing line. That is, when the scale interval length of the measurement time axis is small, the number of measurements per unit time is small, and the production line is stagnating. On the other hand, when the scale interval length on the measurement time axis is large, the number of measurements per unit time is large, indicating that the production line is flowing smoothly. Thus, the presence or absence of the production line can be easily confirmed by looking at the scale interval length of the measurement time axis.

  Further, in addition to the above-described configuration, the quality variation display device of the present invention is the manufacturing data in which the production condition information indicating the change of the production condition in the manufacturing facility and the occurrence time of the change of the production condition are associated, or A manufacturing data storage unit that stores manufacturing data in which work content information indicating the content of work in the manufacturing facility and an occurrence time of the work are associated with each other, and a time point at which measurement is performed from the time point at which the manufacturing facility manufactures. A required time storage unit that stores in advance the required time until the adjustment time obtained by adding the required time stored in the required time storage unit to the generation time of the manufacturing data, and the adjustment time and the measurement time axis And manufacturing data adding means for adding production condition information or work content information to the graph.

  Here, the time required from the time of manufacturing at the manufacturing facility to the time of measuring at the measuring facility is, for example, the time when a certain product starts to be manufactured at the manufacturing facility and the time when the measurement starts at the measuring facility. The time lag (time difference) is shown.

  According to the above configuration, the production data adding means obtains an adjustment time obtained by adding the required time to the generation time of the production data, and based on the adjustment time and the measurement time axis, production condition information or work content Information is added to the graph. Therefore, by looking at the graph, it is possible to easily understand the correspondence relationship between the production condition information or the work content information and the statistics without considering the required time. Thereby, for example, it is possible to easily recognize which production condition changes or work contents cause the statistical amount to vary greatly.

  Further, in addition to the above-described configuration, the quality variation display device of the present invention is a causal relationship for storing causal relationship information for specifying a plurality of manufacturing facilities and specifying the manufacturing facility having the predetermined quality and causal relationship. An information storage unit, wherein the manufacturing data adding means is based on the causal relationship information stored in the causal relationship information storage unit, and the production condition information or work content corresponding to the manufacturing equipment having the predetermined quality and causal relationship Preferably only information is added to the graph.

  According to said structure, a manufacturing data addition means adds only the production condition information or work content information corresponding to the manufacturing equipment which has a predetermined quality and causal relationship to a graph. Thereby, since unnecessary information having no causal relationship with quality is not displayed, it is possible to easily pursue the cause of fluctuation of the statistic shown in the graph.

  Further, in addition to the above-described configuration, the quality variation display device of the present invention includes a parameter input unit that receives the number of sections and the number of shifts, and outputs the input number of sections and the number of shifts to the graph creating unit. It is preferable to provide.

  According to the above configuration, the user can confirm the graph of the desired number of sections and the number of shifts by inputting the number of sections and the number of shifts to the parameter input means.

  The frequency component of the periodic variation in quality on the basis of the number varies depending on various causes of abnormality. When it is desired to confirm the periodic fluctuation of the long frequency component, the periodic fluctuation of the short frequency component acts as noise. In such a case, by increasing the number of sections and the number of shifts, short frequency components can be removed, and periodic fluctuations of long frequency components can be easily confirmed.

  In addition, when the number of sections is one and the number of shifts is one, the influence of measurement error or the like may increase. In such a case, by setting the number of sections and the number of shifts to a value larger than 1, the influence of measurement error and the like can be removed.

  Further, in addition to the above-described configuration, the quality variation display device of the present invention is characterized in that the statistical amount includes an average value of measurement results of products included in each section, a median value of measurement results of products included in each section, each It is at least one of the standard deviation or variance of the measurement results of the products included in the sections and the confidence interval in the measurement results of the products included in each section.

Here, the confidence interval is an interval (x ₁ , x ₂ ) when the probability function Pr (x ₁ <x <x ₂ ) related to the measurement result x is a reliability indicating a probability set arbitrarily in advance. is there.

  According to the above configuration, the average value of the measurement results of the products included in each section, the median value of the measurement results of the products included in each section, the standard deviation or variance of the measurement results of the products included in each section, each section The fluctuation of the confidence interval in the measurement results of the products included in the product can be confirmed.

  Further, in addition to the above configuration, the quality variation display device of the present invention is such that the statistical quantity includes the confidence interval, and the graph creating means approximates a frequency distribution to a normal distribution with respect to a product measurement result. The confidence interval is calculated by performing such a predetermined conversion and inversely transforming the conversion corresponding confidence interval corresponding to the measurement result after the conversion.

  According to the above configuration, even if the frequency distribution of the product measurement result is, for example, an asymmetrical shape and does not approximate the normal distribution, the frequency distribution approximates the normal distribution with respect to the product measurement result. The confidence interval is calculated by performing such a predetermined conversion, and inversely converting the conversion corresponding confidence interval corresponding to the measurement result after conversion. Thereby, the calculated confidence interval corresponds to a desired reliability more accurately. As a result, the product defect rate can be accurately determined by comparing the calculated confidence interval and the upper and lower limit specification values.

  Furthermore, in addition to the above-described configuration, the quality fluctuation display device of the present invention is characterized in that the quality fluctuation graph creating means includes a line indicating at least one of an upper limit standard value and a lower limit standard value of the predetermined quality in the quality fluctuation graph. It is preferable to add.

  According to said structure, the presence or absence of generation | occurrence | production of the defective product exceeding the upper limit specification value or the lower limit specification value can be confirmed immediately.

  The quality variation display device according to the present invention can be applied to a quality display system that displays a variation in quality of products sequentially manufactured in a manufacturing process.

It is a block diagram which shows the structure of the quality variation display apparatus which is embodiment of this invention. It is a block diagram which shows the whole structure of the quality display system containing the said quality variation display apparatus and various facilities. It is a figure which shows the memory | storage example of quality data storage DB of the said quality fluctuation display apparatus. It is a figure which shows one memory | storage example of manufacturing data storage DB of the said quality fluctuation display apparatus. It is a figure which shows an example of the time data which the time information addition part of the said quality fluctuation display apparatus produces | generates. The example of a causal relationship and dead time information storage DB of the said quality fluctuation display apparatus is shown, (a) respond | corresponds to a 1st measurement installation, (b) respond | corresponds to a 2nd measurement installation. It is a flowchart which shows the flow of a process of the said quality fluctuation display apparatus. It is a figure which shows an example of the quality variation graph which the said quality variation display apparatus displays. It is a figure which shows an example of the quality variation graph which the said quality variation display apparatus displays. It is a figure which shows the calculation method of a confidence interval when the frequency distribution of quality data does not approximate normal distribution. It is a graph which shows the relationship between quality data and a confidence interval.

Explanation of symbols

DESCRIPTION OF SYMBOLS 3 1st manufacturing equipment 4 2nd manufacturing equipment 5 1st measuring equipment 6 3rd manufacturing equipment 7 2nd measuring equipment 8 Product 10 Quality fluctuation display apparatus 11 Quality data acquisition part 12 Quality data input part 13 Quality data storage process part 14 Quality Data storage DB (quality data storage unit)
15 Production Condition Information Acquisition Unit 16 Work Content Information Input Unit 17 Manufacturing Data Storage Processing Unit 18 Manufacturing Data Storage DB (Manufacturing Data Storage Unit)
20 Display parameter input section (parameter input means)
21 Section statistic calculation part (graph creation means)
22 Time information addition part (Time information addition means)
23 Manufacturing data adding section (Manufacturing data adding means)
24 Causal relationship / dead time information storage DB (time required storage unit, causal relationship information storage unit)
25 Display section

Claims (5)

A quality fluctuation display device that displays a predetermined quality fluctuation in a plurality of products manufactured in a manufacturing facility,
A quality data storage unit that stores the measurement results of each product measured by the measurement facility in association with the manufacturing order;
The measurement results and the manufacturing order are read in association with each other from the quality data storage unit, the intervals corresponding to a certain number of products that are consecutive in the manufacturing order are shifted by a certain number of shifts, and the statistics of the measurement results for each interval. A graph creating means for obtaining a quantity and creating a graph showing the statistics at regular intervals in the order of manufacture;
A display unit for displaying the graph created by the graph creating means;
The statistics are the average value of the measurement results of products included in each section, the median value of the measurement results of products included in each section, the standard deviation or variance of the measurement results of products included in each section, and included in each section At least one of the confidence intervals in the measurement result of the product, and the statistic includes the confidence interval,
The graph creating means performs a predetermined conversion such that the frequency distribution approximates a normal distribution with respect to the measurement result of the product, and inversely converts the conversion corresponding confidence interval corresponding to the measurement result after the conversion, thereby performing the reliability. A quality fluctuation display device characterized by calculating a section.   2. The quality fluctuation display device according to claim 1, wherein the graph creating unit adds a line indicating at least one of an upper limit standard value and a lower limit standard value of the predetermined quality to the graph. A quality fluctuation display method of a quality fluctuation display device for displaying a predetermined quality fluctuation in a plurality of products manufactured at a manufacturing facility,
The quality variation display device includes a quality data storage unit, a graph creation unit, and a display unit,
A quality data storage step in which the quality data storage unit stores the measurement result of each product measured by the measurement equipment in association with the manufacturing order;
The graph creating means reads the measurement result and the manufacturing order in association with each other from the quality data storage unit, shifts a section corresponding to a certain number of sections in the order of manufacturing by a certain number of shifts, and A graph creation step for obtaining a statistic of the measurement result and creating a graph showing the statistic at regular intervals in the manufacturing order;
A display step in which the display unit displays the graph created by the graph creating means,
The statistics are the average value of the measurement results of products included in each section, the median value of the measurement results of products included in each section, the standard deviation or variance of the measurement results of products included in each section, and included in each section At least one of the confidence intervals in the measurement result of the product, and the statistic includes the confidence interval,
In the graph creation step, the graph creation means performs a predetermined conversion such that the frequency distribution approximates a normal distribution to the product measurement result, and reverses the conversion corresponding confidence interval corresponding to the measurement result after conversion. A quality variation display method, wherein the confidence interval is calculated by conversion.   A quality fluctuation display program for operating the quality fluctuation display device according to claim 1 or 2, wherein the computer functions as each of the means.   A computer-readable recording medium on which the quality variation display program according to claim 4 is recorded.

Images