JP2006344193A - Random traceability stratification average collation analysis method, x-r control chart function part program, random traceability stratification average collation cause analysis program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a program allowing distinctive monitoring of influence collation for each factor in a precision area for analyzing a cause of a chronic characteristic failure problem while continuing production in a field, wherein influence collation cannot be determined conventionally due to deviation or interaction even when a component within a standard is used. <P>SOLUTION: The production with traceability is characterized in that possible traceability of production processes serving as all of product quality factors from a material to completion of a product, components, and materials are uniformly mixed at random. Also, the record of the production are collected intentionally, and as the flow of a chart, the record is stratified while using traceability identification for one factor as an item. Differently from a conventional method, one axis is added to the X-R control chart, and a product problem characteristic mean transition and a mean transition of a factor characteristic such as a component dimension are monitored on the same graph, and collation between the character means is checked. By automatic patrol between the factor and its factor characteristic, a cause of the factor characteristic, which is most correlative with influence, is automatically determined, and notification via sound is also carried out. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an information processing and statistical processing program for cause analysis.

  In causal analysis, the birth and content (Figure 2) of traceability of production factors such as parts, materials and process conditions, and operational factors such as fuel, equipment and operating conditions must be first accurate. This traceability and identification management are also ISO-9000 guideline items, and it goes without saying that this management accuracy and management range affect quality management. With the progress of design using CAD, simulator confirmation, and high-precision measurement tools, the element accuracy has also increased, and the occurrence of defects in the process that significantly breaks the required standards has been reduced. In addition, the functions of the components have become complicated and subtle due to the multi-functional high performance of the product, and even if it is an element within the standardized standard, it is not visible due to interaction, etc., or it is judged that the influence is small at the beginning of production As a result, there are many chronic problems that may occur if they are not included in the scope of management, and the standardization of the fuel design described in the paragraph 0004 of the detailed description of the invention of Patent Document 1 is expected to distinguish between plant factors and fuel factors. However, there is a need to improve the statistical accuracy as described in the paragraph 0015 and the paragraph 0038 in the detailed description of the invention of Patent Document 2. However, even if all the element characteristics are statistically within ± σ as in Patent Document 2, there is a demand for a method for further improving accuracy because the deviation and interaction are not offset-divided.

  As a cause analysis method, there is an ATS method described in Non-Patent Document 1. What (WHAT) defect occurred and why it did not occur, where (WHERE) where the defect occurred and where it did not occur, when (WHEN) when the defect occurred and when it did not occur, how (HOW) Measures and collects factual information on four aspects of how a defect occurs and how it does not occur, searches for distinction points and change points, estimates several possible causes of the difference, and distinguishes them Confirm the existence of contradictions between the points, change points, and the four aspects of information. It is a standard form method with tables for practical use, and is a method of causal analysis by effectively utilizing the expertise that can be possessed. However, while learning this method, as described in the previous section, analysis of process defects due to unobservable factors caused by minute fluctuations of parts within the standard and the interaction of biased element characteristics can be performed using existing single-function calculations. To analyze the cause while producing on-site using graphing and application software for graphing, there are many cases that are harsh and more chronic. Although it is one method to narrow down factors (factorial candidates) from logic and make engineers who can analyze using L12 direct experiment design method, etc. resident in the process, even in the types of products, industrial fields, and specialist engineers, Even those who do not have a little technical skill and can not make logical considerations are able to easily perform statistical cause analysis together with conventional product performance characteristics and supervision in plain production management. Cause analysis program software is required.

  Japanese Laid-Open Patent Publication No. 2001-153988   Japanese Laid-Open Patent Publication No. 2005-38098   US, Kepner Trigo, 1977, ATS (Analytic, Trouble, Shooting) technique

  As described in the background section, while there are many advanced problems in production under the specialized technology of each industry, there are a lot of chronic defect problems. Although both process workers and engineers are working hard day and night, even if they are standardized, they can be viewed even if they are statistically processed by minute or interaction such as fluctuations within the specifications of parts, materials, and process components. In many cases, it is difficult to elucidate the cause, thus providing a method to prevent statistical bias of multiple factor elements and to contrast the influence of one factor.

  There are many single-function programs for creating graphs such as standard deviation, correlation calculation, and scatter plots, and using them, the cause analysis work has become a little easier, but it is still enormous under the complicated current product elements Analytical operation time is required. Therefore, these single functions are systematically linked to monitor the performance of the product performance characteristics in daily production management, change the product characteristics approaching the control limit, and cause the chronic process defects. Statistical causes that can be handled with simple personal computer operation, without distinguishing between those who are engaged in products, industrial fields, general production sites, those who have specialized skills, and those who do not have a little. Provide an analysis program.

  Traceability and identification management, which are the management items in the ISO-9000 guidelines, are thoroughly recorded in the quality control of parts, materials, processes, and products throughout all quality processes. In order to offset the influence of the interaction and bias between components and improve statistical accuracy, mass production in which traceability differences that are possible in the production of parts, materials, and processes are mixed randomly and equally from accumulated records Random element traceability product quality records are collected artificially, and as with the premise of the ATS method of Non-Patent Document 1, the calculation results of simulators including artificial elements are not included. Only information and actual measurement records are used for statistical cause analysis.

  To fix the influence of non-confirmable elements at the center of possible mass production fluctuations, offset fluctuations and bias effects of non-confirmable elements, improve the accuracy of stratified statistics, and highlight the contrast of confirmation element traceability effects The product quality records composed of the random element traceability described in the previous section are input to the database, and the element traceability identification of parts, materials, and processes is divided into groups (keywords), and the product inspection and operation to be analyzed are performed. Obtain the group average of the output evaluation characteristics, repeat the traceability stratification for every element, and the parts, materials, and processes that are the main factor elements corresponding to the group that has the largest inter-group variation of the characteristic average of the analysis target Is identified.

  After that, with the same traceability layer as when the main factor element was specified, the group average of each element characteristic such as part dimensions, material composition ratio, process condition parameter, etc. constituting the main factor element and the group average of the analysis target characteristic To confirm the correlation, identify the element characteristic of the cause candidate that shows the most correlation, and perform regression evaluation between the element characteristic average and the analysis target characteristic average to correct the occurrence and variation of the analysis target characteristic This leads to a review of the tolerance standards for element characteristics and the offset amount.

  Causes the computer to perform the processing of paragraphs 0008 and 0009, identifies the cause factor that is the main factor element characteristic from the configuration record with random element traceability, automatically answers, verifies and confirms it, and performs regression evaluation for the cause countermeasure Therefore, the following six functions are configured into a program: a database function unit, a secondary database function unit, a statistical comparison table function unit, a histogram function unit, an X-R control chart function unit, and a scatter diagram function unit.

  Traceability identification record attached to data recorded including product serial number, machine number and operation date, and element traceability identification record of parts, materials, fuels and processes randomly incorporated into products and operating equipment, Stores four types of records: product performance inspections, operation output evaluation result characteristics records, and component characteristics records such as component dimensions, material and fuel composition ratios, process and operating condition parameters linked to embedded element traceability records And a secondary database function unit for sorting and temporarily storing the database records for each analysis stage by sorting traceability identifications into eyes (keywords).

  Statistical comparison table function that compares and monitors the statistical standard deviation, average, maximum, and minimum of both database and secondary database records to see the specificity of secondary database records sorted from the database with the traceability of the identified elements, and the cause of the cause In order to check the consistency, a histogram function unit is provided that displays the standard deviation and the estimated defect rate of both the database and the secondary database on the monitor and indicates the frequency distribution of the problem characteristics.

  On the X-R chart, not only on the time axis, but on the traceability that you want to check, and not only the transition of one characteristic average such as product performance characteristics and operation output characteristics, but also part dimensions, materials, and fuel component ratios The average transition of element characteristics such as process and equipment operation setting condition parameters is displayed on the same graph, and the synergistic relationship with the transition of the element characteristics can be monitored according to the conventional one characteristic monitoring function. X-R control chart function unit that makes the response of the main cause element characteristic name and correlation index of the cause candidate showing the most correlation between the averages of the characteristics to be analyzed in the monitor display, message diagram, and voice, and regression The stratified level of the analysis target characteristics and element characteristics checked by the X-R control chart function unit to review and correct the tolerance standard and offset amount of the factor element characteristics from the scatter chart including the linear formula and regression line Providing a scatter plot function unit for monitor display performs regression evaluation between.

  The situation of nuclear power generation for which Patent Document 1 has been applied is not necessarily related to the power plant of Patent Document 2, and it is a reality that no matter what the actual situation is, it will not be carried as planned. It is premised on the identification of the cause even in the countermeasures of the cause, it is possible to analyze the cause and cause of the accuracy range that could not be seen before, improve the production quality with accuracy, and the chronic defect problem regardless of pharmacy medical analysis, agriculture or industry It can be used not only for countermeasures for the cause, but also for business development in new fields and development of new products through accurate factor analysis, advancement of crisis prevention due to clarification of the operating limits of individual equipment, and Performance and miniaturization can be achieved, which contributes to reducing wasteful consumption of resources.

  Now that PCs have become supermarkets, the spread of them has naturally become a daily tool, but on one side, there are parts that are not fully utilized, and the conventional X-R control chart that can be easily handwritten on paper Has been useful for monitoring the performance transition of product performance characteristics, etc., but the monitor graph of the X-R control chart function section according to claim 1 is a component dimension, material or fertilizer component which is an element characteristic constituting the product. By processing rate, process condition parameters, etc. on a monolith, it is possible to monitor the transition of element characteristics instead of on the same graph, and it is now possible to easily identify the element characteristics that cause abnormalities and changes in product performance characteristics Therefore, it has become possible to manage production processes efficiently without dividing the product and industrial fields.

  Since the characteristic values of the constituent elements are different for each product or device, the average control analysis according to the random traceability layer according to claim 3 is applied to the operation control of devices having various computers such as satellites, aircraft, automobiles, and liquid crystal televisions. If there is no problem in setting the parameters of all devices in the device, make random changes to the device itself, record and accumulate each output characteristic change, and read the influence and correlation of each device on the output characteristics of all devices. Of course, priority is given to the requirements, let the device itself set the parameters of all devices without difficulty, and the relationships that the designers are not aware of will be clarified. It is possible to improve the efficiency of the automatic adjustment function in the automatic assembly machine for production.

  In order to secure the data for analysis, the records in FIG. 3 (1), such as the product serial number, production date, production location, production factory, line name, production process number, etc. shown in FIG. Record of (2) of example of material traceability, record of (3) of example of product inspection and operation output evaluation result, and (4) of example of material component ratio of element characteristics, production process parameters, part dimensions, etc. Records, as well as in accordance with ISO-9000 guidelines, traceability and identification management are thoroughly implemented at the site of all quality processes, and the traceability identification process discrimination in the examples of Fig. 3 (1) and (2) is the product 3) (3) (4) The characteristic records in the example of Fig. 3 (3) and (4) can be easily traced through the network and email. Build the available environment, only to be ordered as of the only Figure 3 record of items required at the time of analysis. Computerization is not necessarily a prerequisite for recording, but if it is arranged, the time taken in can be further reduced.

  The traceability mixed production in the flow diagram 1 is an example in the table of FIG. 3 in which data is collected in order to investigate the mold effects of the traceability differences in the components, materials, and production processes that are components of the product (2) Parts A combination of element traceability matches from the accumulated data by creating a product assembly instruction table that randomly incorporates the element traceability that is included in the production record and production records that are randomly combined, such as each CV (production use model number) A record of the product is collected, and when there is a shortage, additional production is performed in the same way with existing elements.

  The example of the record storage contents of the database function (part) in the flow chart 1 is as described in the example of the paragraphs 0017 and 0018 and the table of FIG. 3, and the example of FIG. In order to analyze which part and which dimension are the main causes of the variation in the defect rate of characteristic E in characteristic (3), we have collected production traceability records. The dimension record is the function that the database function (part) has after identifying the part B that is the most variable influencing element, and the traceability items of the part B randomly incorporated in FIG. Along with the CV (model number), a dimensional inspection record for each model number separately recorded at the time of part production is copied and sorted and written manually. Intrinsic functions allow for instant writing. Figure 3 shows a sample monitor. Actually, there are many other items in (1), (2), (3), and (4), but they are thinned out and actually scrolled up and down, left and right. Can see the whole picture. In this example program, the data No. Colors are added to the table cells of the characteristic items and the statistical range is specified arbitrarily by the user. In addition, since there are many TXT types for recording in various servers, a table conversion incorporation function (part) is located in front of the database function (part), although not shown in the explanation flow diagram 1.

  As shown in FIG. 4, the secondary database function (part) in the flow diagram 1 stores product quality record data consisting of random element traceability stored in the database function (part), and the identification code (6) in the traceability item (5). This is a function for sorting the data into eyes (keywords), sorting them into sorting groups, temporarily storing the sorting result data (7) according to the analysis stage, and monitoring the data. The table (7) in FIG. 4 is sample-monitored, the number of items (rows) is thinned out, and the lower part is cut in the middle of the table. In fact, you can scroll up, down, left, and right to see the whole picture.

  As shown in FIG. 5, the statistical comparison table function (part) in the flow chart 1 confirms the distribution difference between the product characteristics, output characteristics, and element characteristics of the secondary database (SortSheet) and the original database (DataSheet) numerically. In order to observe the peculiarity of the data of the sorted secondary database, it is a function to monitor the numerical value table of the standard deviation, maximum, minimum and average of all characteristics and the input field of each characteristic specification upper and lower limit . It can also be said to be a histogram creation preparation function section described in the next section. This monitor can be scrolled down to see the whole picture.

  As shown in FIG. 6, the histogram function (part) in the flow chart 1 is selectively driven by double-clicking on any of the characteristic names in the statistical comparison table of FIG. 5 described in the previous section for product characteristics, output characteristics, and element characteristics. Then, in order to confirm the peculiarity of the characteristic distribution of the secondary database sorted in advance by arbitrary traceability, the frequency distribution of the same characteristic of the original database is displayed side by side with the histograms (8) to (11). This function represents the standard deviation, maximum / minimum, average, and estimated defect rate (12), (13), and can be monitored. FIG. 6 shows a histogram of product problem characteristics E viewed by sorting production records in which the part B is assembled with the model number 2 into the secondary database. The 2nd among the types of the parts B from the 1st to the 8th It can be seen from the fact that the estimated defect rate in (13) is zero that the mold does not produce defects. In this way, confirm the consistency of the factor and cause identification results to confirm the distribution location of the production quality of the traceability of part of the factor and cause parts in the original whole population, and check the lot of parts There is a purpose to determine whether or not each traceability such as model number can be used. The flow of recorded data information coming to the histogram function (part) in FIG. 1 is not necessarily the flow before specifying the cause, so the flow arrow line in the figure is a dotted line.

  The X-R control chart function (part) of the flow diagram 1 is not limited to the time series as shown in FIG. 7 instead of the conventional X-R control chart, and the element traceability (identification) item (14) of parts, materials, and processes is used. ) Code sequence (15) is used as the transition confirmation axis, and all the element traceability items are referred to and monitored, and the change in the problem of the driving output characteristics, the average transition of the product characteristics, and the fluctuation in the example (18) are monitored. Specify the parts, materials, and process elements that correspond to the traceability identification items that appear in (14). The range of the element characteristic items that correspond to the specified elements at the location set at the bottom of the scrolled screen that cannot be seen on the screen of FIG. The element characteristic items such as part dimensions, material composition ratio, process condition parameters, etc. in that range are automatically referenced and circulated by the stratified axis with the same traceability identification items as when the largest influential elements were identified. The correlation between the element characteristic average transition (21) and the product characteristic average transition (18) is confirmed, and from among the element characteristic items, the cause candidate name (19) and the correlation index (22) which are the element characteristics that are the most correlated automatically Is displayed in the lower part of the monitor graph, and the same content is reported in the message diagram (23) and voice.

  If traceability records can be managed throughout the quality process in accordance with ISO-9000 guidelines, this example program can be used instead of the conventional X-R chart for daily production management. If the product traceability, built-in element traceability, and product performance characteristics are recorded in the DataSheet (database), rapid identification of the components, materials, and processes that cause the rapid variation and change is possible. -R control chart functional unit is possible. In addition, it is necessary to record the random traceability element configuration described in paragraph 0018 for regression evaluation for reviewing the standards with accurate accuracy, such as the component dimensions and material component ratios of the cause element characteristics and process condition parameter settings, If the accumulation is insufficient, additional production is performed according to the random assembly instruction table.

  The scatter diagram function (part) of the flow diagram 1 monitors the correlation state of both characteristics confirmed by the XR control chart function (part) as shown in the scatter diagram of FIG. 8, and the correlation R square value (24) In accordance with the higher regression line equation (26) (27) of (25), it leads to the review of the allowable standard of the cause candidate and the element characteristic of the factor and the offset amount. The above is the description of the cause analysis program embodiment.

  At present, it is natural to use a computer to control each related device of multiple output characteristics of a device in multiple ways. Construct free control of setting parameters of internal element devices. For all devices such as satellites, aircraft, and LCD TVs, output characteristics data such as all outputs, speed, and brightness, input amounts such as fuel and power on the component side added at the time, internal device parameter settings and characteristics In addition, the environmental measurement data such as outside temperature and brightness are placed on a single table in the computer, and this contrast analysis method is used. Change the parameter setting of the device to be configured several times within random and full range at the same time, try self-test to measure the influence on the multiple output characteristics of the device at the same time, select the data of similar random configuration from the accumulation, Priority is placed on requirements such as output and efficiency within the limits of the correlation between the calculated output characteristics and each parameter setting and the capability of the regression evaluation results. In addition, self-control of the parameter settings of each of the multiple factor devices in the equipment, reduces the burden on the equipment-the device, and works efficiently to obtain a life-prolonging effect. Is used mainly for repair services and new product design.

  There are various automatic adjustment assembly machines, but for example, in optical ball axis alignment adhesive fixing assembly of multi-axis alignment optical lens and lens frame, the actual adjustment time is averaged starting from the average space coordinates of the adjustment completion record There is a method of starting the movement in the shortest direction, but the size and shape of each lens and lens frame are different, so there are many cases of stray, and there are not many cases where the merit of automating the corners is poor. Using this random-traceability stratified average contrast analysis method, the measurement space coordinate point that is a random combination of each axis so that the moving average of each axis coordinate becomes the center of the axis movable range of the automatic machine, and the shortest connecting the space point For each pair of lens and lens frame, the effect of the optical characteristics is measured for each spatial point, and the adjustment range is determined according to the regression evaluation with those axis coordinates. It prevented reliably vagus in a way that squeezing the at less haste, perform efficient adjustment.

It is a random traceability stratum average control analysis method and cause analysis flow chart. It is a flowchart explaining the relationship between management of traceability, identification, and a production process. It is a database function part monitor screen. It is a secondary database function part monitor screen. It is a statistics comparison table function part monitor screen. It is a histogram function part monitor screen. It is a XR control chart function part monitor screen. It is a scatter diagram function part monitor screen.

Explanation of symbols

1 Production traceability items (Example: Product production date and sample number)
2 Element traceability items incorporated into each product (example: model numbers of parts B and C)
3. Product characteristic items (Example: Product characteristics D and E)
4 Element characteristic items (example: each dimension of part B)
5 Setting field for sort key item (example: part B. CV (model number))
6 Sort keyword setting field (Example: Part B. CV model code)
7 Sort result (Example: Part B. Result of categorizing statistical range records by model number of CV)
8 Histogram between maximum and minimum of database (example: characteristic E histogram)
9 Histogram between specified numerical values such as database standards 10 Histogram between maximum and minimum of secondary database (example: characteristic E histogram of product group in which part B is assembled with model number 2)
11 Histogram between specified numerical values such as secondary database standards 12 Estimated failure rate of database confirmation characteristics (example: characteristic E) 13 Estimated failure rate of secondary database confirmation characteristics (example: characteristic E) 14 Next to graph Axis item setting field (example: part B.CV)
15 Graph horizontal axis label (Example: Part number code of component B for element traceability identification)
16 Analysis target characteristic item setting field (example: characteristic E)
17 Analysis target characteristic average value axis (example: characteristic E average value axis)
18 Analysis target characteristic average transition (example: characteristic E average transition)
19 Correlation confirmation element characteristic item column (example: part B dimension 5.2 ± 0.03)
20 Correlation confirmation element characteristic average value axis (example: average value axis of part B dimension 5.2 ± 0.03)
21 Correlation confirmation element characteristic average transition (example: average transition of component B size 5.2 ± 0.03)
22 Correlation index (Example: Correlation index of product characteristic E and part B size 6.2 ± 0.03)
23 Maximum correlation element characteristic (cause candidate) automatic search result report message diagram 24 Correlation index R-squared value 25 observed in linear correlation 26 Correlation index R-squared value observed in polynomial correlation 26 Linear correlation regression line equation 27 Polynomial correlation regression curve equation

To fix the influence of non-confirmable elements at the center of possible mass production fluctuations, offset fluctuations and bias effects of non-confirmable elements, improve the accuracy of stratified statistics, and highlight the contrast of confirmation element traceability effects The product quality records composed of the random element traceability described in the previous section are input to the database, and the element traceability identification of parts, materials, and processes is divided into groups (keywords), and the product inspection and operation to be analyzed are performed. Obtain the group average of the output evaluation characteristics, repeat the traceability stratification for every element, and identify the parts, materials, and processes that are the main factor elements corresponding to the group with the most variation of the characteristic average of the problem between groups To do.

After that, with the same traceability layer as when the main factor element was specified, the group average of each element characteristic such as part dimensions, material composition ratio, process condition parameter, etc. constituting the main factor element and the group average of the analysis target characteristic To confirm the correlation, identify the element characteristic of the cause candidate that shows the most correlation, and perform regression evaluation between the element characteristic average and the characteristic average of the problem, and correct the occurrence and variation of the analysis target characteristic This leads to a review of the tolerance standards for element characteristics and the offset amount.

Causes the computer to perform processing between 0007 and 0009 , identifies a cause candidate that is a main factor element characteristic from a configuration record with random element traceability, automatically answers it, further verifies and confirms it, and returns to cause countermeasure In order to make an evaluation, the following six functions are configured into a program: a database function unit, a secondary database function unit, a statistical comparison table function unit, a histogram function unit, an X-R control chart function unit, and a scatter diagram function unit.

On the X-R control chart, not only the time axis but also the traceability that you want to check can be monitored around the traceability you want to check, and the fluctuations in the product performance characteristics and operation output characteristics in question can be monitored. To be able to identify
Not only the transition of one characteristic average such as product performance characteristics and operation output characteristics, but also the average transition of element characteristics such as part dimensions, materials and fuel component ratios, process and equipment operation setting condition parameters on the same graph, Along with the conventional one-characteristic monitoring function, it is possible to monitor the synergistic relationship with the transition of element characteristics, correlating with the main factor element characteristic name of the cause candidate that shows the most correlation between the averages of element characteristics and problem characteristics X-R control chart function unit that makes the index respond with a monitor display, a message diagram, and voice,
Regression evaluation between stratified averages of analysis target characteristics and element characteristics confirmed by the X-R control chart function unit to review and correct the tolerance standard and offset amount of the factor element characteristics from the regression line formula and the scatter chart including the regression line A scatter diagram function unit is provided for performing monitor display.

The traceability mixed production in the flow diagram 1 is an example in the table of FIG. 3 in which data is collected in order to investigate the mold effects of the traceability differences in the components, materials, and production processes that are components of the product (2) Parts A production record that is randomly combined, such as CV (production use model number) for each product, and a product assembly instruction table that randomly incorporates element traceability in the production return history is created in this program, and the element traceability from the accumulated data A record of products with the same combination is collected, and when there is a shortage , additional production is performed in accordance with the product assembly instruction table created in the same way by using the existing elements.

Claims (3)

  Collecting production records and operation records with random (even) traceability differences (evaluation environment, device parts, fuel, etc.) that are fluctuation factors of product performance and device operation output analysis target characteristics The traceability identification of one element is sorted into eyes (keywords) and divided into groups. If the analysis target characteristic and its one element part, the average of each element characteristic such as dimensions is calculated, and the group element Traceability records of the target product or operation, and the elements that comprise them, in order to perform cause analysis according to the average control analysis method by random traceability stratification that distinguishes the influence contrast of element characteristics, and performs correlation and regression evaluation between the averages Traceability records, actual measurement records of characteristics to be analyzed, and element characteristics linked to element traceability Database function section for storing four types of records, secondary database function section for temporarily storing database records by element traceability identification, and statistic values for all characteristics of both database and secondary database In order to verify the causes and factors together with the statistical comparison table function unit for statistical comparison, and the statistical comparison table function unit, the analysis target characteristics of both the database and the secondary database sorted by the traceability of the specified cause candidate elements Histogram function section that compares and monitors the frequency distribution of problem characteristics, and the secondary database function section makes it possible to use group traceability by element traceability to record group average transitions for both problem characteristics and element characteristics including causes and factors Unlike the conventional case, the average transition from one characteristic to two characteristics is displayed on the same graph. The X-R control chart function unit with the function to automatically search for the cause candidates by checking the correlation between the averages of the two characteristics and reading the permissible confirmation and corrective measure values of the element characteristics of the cause factors The scatter diagram function section that performs regression evaluation for the purpose is linked and driven on a computer such as a personal computer while monitoring the processing results of each function section, to maximize the influence of variations in analysis target characteristics and fluctuation problems. , Random correlation traceable average control cause analysis program that automatically identifies and answers to cause candidates that are element characteristics of maximum correlation and main factors   The trace axis of the X-R control chart for managing product and event evaluations and output characteristic transitions can be replaced with traceability items that you want to check, not limited to time-series year / month / date / time. It is possible to monitor and check changes in management characteristics such as product characteristics depending on the same graph, not only with the one-character average transition of conventional product performance characteristics, but also with the addition of one axis to the same graph with the average transition of element characteristics It is possible to monitor the synergistic relationship between the two, and add a function to confirm the correlation between the averages of the two characteristics, and display the correlation index at the bottom of the graph, sorting by the traceability of the element where the problem characteristics such as products vary most The element characteristic item of the same element in the stratified secondary database is automatically circulated within the specified range, and the correlation between the problem characteristic of the product and each element characteristic is revealed. And a function for reporting the maximum influence on variations and fluctuations of problem characteristics of products and the like, the cause candidate which is the maximum correlation element characteristic, and its correlation index by a monitor display, a message diagram and voice. X-R control chart function part of description   Product characteristics and device operation output analysis target characteristics are biased in element characteristics such as evaluation environment such as outside air temperature, equipment operating conditions such as fuel supply, equipment part shape dimensions, device settings within equipment, and fuel component ratio Evaluation date / time, equipment number, equipment production date / time, equipment production process number, parts production plant, so as not to interfere with the statistical impact assessment of the single element characteristics on the characteristics to be analyzed due to the effects of variation and interaction Record history traceability and element characteristics of all quality processes, such as name, parts production date and time, parts production machine and model number, and create production and operation records in which the possible traceability differences of all quality processes are mixed evenly and randomly The records are categorized into stratified groups with the identification of traceability of one element as an eye (keyword), and the average for each group of characteristics to be analyzed is calculated. Since element traceability within a group that is not equivalent to a group is randomly equal to each other, the average of the element characteristics of each element within each group is always the same in the middle of the range of possible fluctuations. The average effect of the element characteristics is also the same, and the difference in the influence of each group average of the analysis target characteristics is only the difference of the group average of the element characteristics of the element in the group. Since it was possible to clarify the influence contrast (contrast) on the analysis target characteristics, it was not possible to simply use the group average of the analysis target characteristics and element characteristics, and simply perform the stratified statistical analysis of random information. 2. Random traceability stratified average according to claim 1, wherein the correlation between the analysis target characteristic and the element characteristic within a certain accuracy range and the statistical evaluation of regression are performed. Irradiation analytical methods

Images