JPH1054858A - Apparatus and method for inspection - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an apparatus and a method, for an inspection, which can self-evaluate a deviation in measured values and an irregularity in measurements by means of a statistical technique. SOLUTION: The characteristic of an object MS to be inspected is evaluated by a characteristic evaluation part 1, a mean value and an irregularity degree are computed on the basis of measured data or a result which is obtained by measuring a prescribed master sample repeatedly, the magnitude of a change amount with reference to an initial state is judged, the degradation of the accuracy of an inspection apparatus MQ itself is detected by a self-detection part 2, and a measured result or a self-detected result is displayed by a display means 3.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an inspection apparatus and an inspection method for inspecting characteristics of various electronic devices, and more particularly to an inspection apparatus and an inspection method capable of always performing a stable inspection by self-diagnosis.

[0002]

2. Description of the Related Art In general, a characteristic inspection apparatus for inspecting the characteristics of a mass-produced product performs a function measurement of each mass-produced product at the final stage of development of the device to know whether or not the characteristic inspection device is correctly inspecting. There is a need. In this case, conventionally, a single master sample is appropriately prepared from the object to be inspected, and the characteristic is inspected only once. Then, by comparing the actually measured value of the characteristic test of the master sample with the preset threshold value, it is determined whether or not the inspection result of the inspection device is out of order.

[0003]

The above-mentioned master sample is randomly extracted from a mass-produced product, and is itself a sample of a population having a certain variation. Therefore, there is a possibility that the characteristics of the master sample originally have a variation with respect to a preset threshold value. As a result, with the above-described method, it was not possible to accurately measure a change in the threshold value or variation of the characteristic inspection device.

However, in general, the accuracy of the characteristic inspection apparatus itself may change due to a change over time or the like. In this case, the threshold value fluctuates to some extent, and function measurement cannot be performed accurately. . As a result, there is a possibility that a product that once obtained a test result that is a non-defective product is later determined to be a defective product, or may be determined to be a defective product despite being a non-defective product.
Therefore, there is a need to more accurately detect the accuracy of such a characteristic inspection apparatus itself.

Japanese Patent Publication No. Hei 6-60918 discloses an example of a characteristic inspection apparatus that makes a self-determination. However, this apparatus calculates a failure rate according to a service period. Of failure data, and a huge amount of data is needed. In addition, it merely calculates the failure rate of the same type of characteristic inspection apparatus, and does not accurately determine whether or not each individual characteristic inspection apparatus has a specific failure. Therefore, it was not always possible to accurately detect the accuracy of the characteristic inspection device.

An object of the present invention is to provide an inspection apparatus and an inspection method which can self-evaluate a deviation of a measured value, a variation in measurement, and the like by a statistical method.

[0007]

Means for Solving the Problems In order to solve the above problems,
The present invention is an inspection apparatus for performing a characteristic inspection of an object to be inspected, a characteristic measuring unit for measuring characteristics of the object to be inspected, and determining whether the object to be inspected is a non-defective product based on the measured characteristics. Classification means for classifying the values, recording means for accumulating and recording the measured values in the characteristic measuring means, and a statistic for calculating at least one of the average value and the degree of dispersion of the measured values accumulated and recorded in the recording means The calculating means, the statistic calculated by the statistic calculating means, and a reference statistic obtained by previously measuring a plurality of the test objects or measuring the single test object a plurality of times. And a statistic error detecting means for detecting a difference between the statistic values and a statistic difference detected by the statistic error detecting means determines whether or not the statistic value exceeds a predetermined allowable range. Statistical tolerance judgment method When the statistic allowable range determining means determines that the statistic exceeds the predetermined allowable range, it indicates that the inspection apparatus itself is in a defective state and the statistic exceeds the predetermined allowable range. Display means for displaying the result of sorting by the sorting means when it is determined that the sorting has not been performed.

Preferably, the classification means has a function of stopping the classification operation when the statistic allowable range determination means determines that the statistic exceeds the predetermined allowable range.

In the above configuration, in the initial state, a plurality of the objects to be inspected are actually measured, and at least one of the average value and the degree of variation is calculated to obtain a reference statistic.

[0010] Thereafter, when inspecting an object to be inspected other than the object to be inspected, which is the object of the reference statistic, the actual measurement data measured by the characteristic measuring means is stored and recorded by the recording means. Calculate at least one of the average value and the degree of variation of the actually measured data. Then, a difference between the calculated statistic and the reference statistic is detected, and it is determined whether or not the statistic exceeds a predetermined allowable range based on the difference in the statistic.

Here, when it is determined that the statistic exceeds the predetermined allowable range, the display means indicates that the inspection apparatus itself is defective. On the other hand, when it is determined that the statistic does not exceed the predetermined allowable range, whether or not the inspection target is a non-defective product is classified by a classification unit, and the classification result by the classification unit is displayed. To display.

Alternatively, in the initial state, the single test object is actually measured a plurality of times, and at least one of the average value and the degree of variation is calculated to obtain a reference statistic.

[0013] Thereafter, data obtained by repeatedly measuring the same single test object identical to the test object set as the target of the reference statistic a predetermined number of times is accumulated and recorded by the recording means. After calculating at least one statistic of the average value and the degree of variation of the calculated data, a difference between the calculated statistic and the reference statistic is detected. Then, it is determined whether or not the statistic exceeds a predetermined allowable range based on the difference between the statistic, and when it is determined that the statistic exceeds the predetermined allowable range, Displays on the display means that the inspection apparatus itself is defective.

[0014]

BEST MODE FOR CARRYING OUT THE INVENTION

<< First Embodiment >><Configuration of Inspection Apparatus> FIG. 1 is a block diagram showing an inspection apparatus MQ according to a first embodiment of the present invention. This inspection device MQ
Is a characteristic evaluation unit 1 that measures and evaluates the characteristics of the inspection target MS, a self-detection unit 2 that detects deterioration of the inspection accuracy of the inspection device MQ itself based on the measurement result, Display means 3 for displaying a detection result.

The characteristic evaluation section 1 includes a characteristic measuring means 11 for measuring the characteristics of the inspection target MS, and determines whether the inspection target MS is a passing product or not based on the characteristic measurement result, and determines a non-defective product and a defective product. A classification unit 12 for performing classification and displaying the classification result on the display unit 3.

The self-detecting unit 2 is a recording unit 1 for recording the measured values of the characteristics measured by the characteristic measuring unit 11 of the characteristic evaluating unit 1.
4, average value calculating means 15 for calculating the average value of the measured values of the characteristics recorded in the recording means 14, and average value calculating means 15
Average error detecting means 16 for detecting the difference between the average calculated on the basis of the calculation result and the reference average recorded in advance.
The average error tolerance determining means 17 for determining whether or not the difference between the average values detected by the average value error detecting means 16 exceeds a preset first allowable range; Dispersion calculation means 1 for calculating dispersion of measured values of characteristics
8, a variance error detecting means 19 for detecting a difference between the variance calculated based on the calculation result of the variance calculating means 18 and a preset reference variance (maximum variance Smax), and a detection by the variance error detecting means 19 Based on the result, the variance allowance determining means 21 for determining whether or not the variance exceeds a second allowable range (maximum variance Smax) set by the variance allowance determination means 17 and the variance allowance determination means 21 When it is determined that at least one of the results exceeds the preset first or second allowable range, the output unit 2 that notifies the display unit 3 that the inspection apparatus MQ itself is in a defective state.
2 is provided.

Here, the classification means 12 of the characteristic evaluation section 1
When the average difference or variance determination unit 17 or the variance tolerance determination unit 21 of the self-detecting unit 2 determines that the difference or the variance of the average values is too large to be unacceptable, a function of stopping the measurement of the characteristics of the inspection target MS is provided. Have.

The information to be output from the output means 22 to the display means 3 when it is determined that the average error tolerance determination means 17 or the variance tolerance determination means 21 does not allow the error is, for example, "Error: device: Is too large and maintenance is required. ", And the output to the display means 3 is stopped when it is within the allowable range.

A general microcomputer chip having a CPU, a ROM, a RAM, and the like is used for the characteristic evaluation section 1 and the self-detection section 2, and all the arithmetic operations are executed by software stored in advance. Things.

The reference average value and reference variance used in the average value error detecting means 16 and the variance error detecting means 19 are as follows:
The measured values within a certain period of time at the beginning of mass production are recorded in the recording means 14 and the initial operation values calculated by the average value operation means 15 and the dispersion operation means 18 are used.

<Operation> FIG. 2 shows a vehicle-mounted ECU 30 as an example of the inspection target MS of the inspection apparatus MQ.
The ECU 30 opens the door and sets the door switch 34 to O
When N is reached, the NPN transistor 35 is turned on by a command from the control unit 36 to turn on the interior light 31. Conversely, when the door switch 34 is turned off, that is, when the door is closed, the NPN transistor 35 is also turned on by a command from the control unit 36. The operation of the transistor 35 is controlled so that the lighting state of the interior light 31 is maintained for the time T and then turned off. The control of the time T at the time of turning off the light is performed by a timer means (timer) in the ECU 30.

It is assumed that the inspection device MQ is connected to a connection point 33 between the NPN transistor 35 and the vehicle interior light 31 with respect to the ECU 30.

The voltage level V applied to the connection point 33 of the ECU 30 is inspected as shown in FIG.

In general, products have variations in characteristics, but statistically the distribution of characteristics should be substantially uniform. Therefore, as long as the accuracy of the characteristic measurement does not change from the initial state, the frequency of occurrence of the actually measured data is shown in FIG.
It should approximate a given distribution, such as the curve in the middle. Conversely, when the accuracy of the characteristic measurement changes with respect to the initial state, the characteristic measurement result becomes the actual measurement data Br1, Br as shown in FIG.
2 change with respect to the average value (reference average value) VQ of the distribution Bp in the initial state, or as shown in FIG. 7, the variance Sv of the measured data Br1, Br2,.
.., Svn change with respect to the variance (reference variance) SvQ of the distribution Bp in the initial state. Taking advantage of this, in this embodiment, the self-evaluation of the inspection apparatus MQ itself is also performed at the same time by using the actual measurement data used for the actual characteristic evaluation as it is.

First, in step S1 in FIG.
The actual measurement data for each product of the CU 30 is collected by the characteristic measuring unit 11 of the characteristic evaluation unit 1 and the self-detection unit 2
Is recorded in the recording means 14. At this time, it is assumed that measured values Vi, Vii,..., Vn as shown in the following table are obtained.

[0026]

[Table 1]

The distribution of the measured values Vi, Vii,..., Vn in this case is as shown in FIG. 5, from which the average value Vav and the variance S of the voltage level can be obtained.

Now, assuming that actual measurement data has been accumulated for a predetermined number of lots, the average value calculating means 1 is used as shown in FIGS.
It is assumed that the average value Vav is obtained at 5 and the variance S is obtained at the variance calculation means 18 (step S2). The arithmetic expression at this time is as follows.

Vav = (Vi + Vii + Viii +... + Vn) / n (1) S = [{(Vi−Vav) ² +... + (Vn−Vav) ² } / (n−1)] ^1/2 (2) Now, as the average value Vav and the variance S, the average value V1 and the variance Sv1 are obtained at a certain opportunity (data Br1), and the average value V2 and the variance Sv2 are obtained at another opportunity (data Br2). It is assumed that an average value Vn (not shown) and a variance Svn have been obtained for (data Brn).

For each opportunity, in step S3, the differences Δv1, Δv2 (,..., V) between the calculated average values V1, V2 (,..., Vn) and the pre-recorded reference average value VQ.
Δvn) is obtained by the average value error detecting means 16 and the differences ΔSv1, ΔSv between the calculated variances Sv1, Sv2,..., Svn and the previously recorded maximum allowable variance Smax.
2,..., ΔSvn are obtained by the dispersion error detecting means 19. Then, in step S4, the average error tolerance determining means 17 determines whether or not the difference Δv1, Δv2 (,..., Δvn) between the average values exceeds a first allowable range Lv set in advance. Variance difference ΔSv1, ΔS
.., Svn are negative values, that is, the variances Sv1, Sv2,.
Whether or not x is exceeded is determined by the dispersion allowance determining means 21.

Average value differences Δv1, Δv2 (,..., Δvn)
Alternatively, if it is determined that the variances Sv1, Sv2,..., Svn are permissible, that is, smaller than the respective allowable ranges Lv, Smax, the fact is continuously transmitted to the classification unit 12 of the characteristic evaluation unit 1 and the classification unit 12 is notified. Is performed, the non-defective product and the non-defective product are separated for the inspection target, the result of the separation is displayed on the display unit 3, and the inspection is continued (Step S).
5). At this time, the output unit 22 does not operate, and the display unit 3
Does not display an error message. For example, in both cases of FIGS. 6 and 7, the average difference Δv1 and the variance Sv1 of Br1 are allowed because they are smaller than the allowable range Lv and the maximum variance Smax.

On the other hand, the differences Δv1, Δv2 (,.
vn) or when the variances Sv1, Sv2,..., Svn are determined to be unacceptable, that is, larger than the respective allowable ranges Lv, Smax, the fact is transmitted to the classification means 12 of the characteristic evaluation unit 1 and the classification is performed. At step S6, a signal is transmitted to the display means 3 to display a predetermined error message (step S6). For example, in the case of FIG. 6, the average difference Δv2 of Br2 is larger than the allowable range Lv, so that it is not allowed. In the case of FIG. 7, the data Br1 and Br2 are allowed because they are smaller than the maximum variance Smax, but the data Brn are not allowed because they are larger than the maximum variance Smax. In this case, whether the worker performs appropriate maintenance or the inspection device M
Q itself will be exchanged.

The above is an example of the measurement relating to the voltage level V in FIG. 3. However, the difference in other transmission components such as the delay time T from the time when the door switch 34 is turned off to the time when the vehicle interior light 31 is turned off, etc. The evaluation of the functional characteristics
Characteristic measurement and self-detection of accuracy deterioration may be performed by exactly the same procedure.

As described above, the self-evaluation of the inspection apparatus MQ itself can also be easily performed simply by using the actual measurement data measured for evaluating the characteristics of the product.

Second Embodiment A second embodiment of the present invention uses an inspection apparatus MQ having the same configuration as that of the first embodiment. However, in the first embodiment, measurement is performed for characteristic evaluation. In contrast to the self-evaluation of the inspection apparatus MQ itself using the measured data directly, in this embodiment, a single master sample to be self-evaluated is determined, and the same master sample is continuously used. It is intended to detect whether there is no change in the measurement result by repeatedly performing the measurement.

The operation of this embodiment will be described with reference to the flowchart of FIG. First, in step T1, a predetermined master sample is prepared, and the characteristic measurement unit 11 of the characteristic evaluation unit 1 prepares the master sample.
A predetermined characteristic is measured repeatedly and continuously for a predetermined sufficient number n, and the measurement result is recorded in the recording means 14 of the self-detecting unit 2.

When the measured data is accumulated for a predetermined number of lots, the above-mentioned equations (1) and (2) (see FIGS. 6 and 7)
, The average value Vav and the variance S are calculated by the average value calculating means 15 and the variance calculating means 18 (step T).
2).

In step T3, the difference Δv between the calculated average value Vav and the previously recorded reference average value VQ is determined by the average value error detecting means 16, and the calculated variance S and the previously recorded maximum variance Smax Is obtained by the dispersion error detecting means 19. Then, in step T4, the average error allowance determining means 17 determines whether the average difference Δv exceeds a first allowable range Lv set in advance, and the variance difference ΔSv
Is a negative value, that is, whether the variance S exceeds a preset maximum variance Smax is determined by the dispersion allowance determination means 21.

At this time, if it is determined that the difference Δv of the average values or the variance S is allowable, that is, smaller than each allowable range Lv, Smax, the output means 22 does not operate, and
No error message is displayed on the display means 3.

On the other hand, if it is determined that the average difference Δv or the variance S is unacceptable, that is, larger than the allowable range Lv, Smax, a signal is transmitted to the display means 3 to display a predetermined error message. (Step T5).

As described above, by using a predetermined master sample, the self-evaluation of the inspection apparatus MQ itself can be easily performed.

In the above embodiments, the average value and the variance are measured as the actually measured data. However, only one of them may be measured, or the variation of the data may be replaced by the variance. Alternatively, another type of parameter such as a standard deviation may be calculated.

[0043]

According to the first aspect of the present invention, when an unacceptable change occurs in a statistic such as an average value or a degree of variation in the inspection, not only the evaluation of the product characteristics but also the inspection apparatus itself. Self-evaluation can also be performed automatically. Therefore, stable inspection can always be performed.

In particular, according to the second aspect of the present invention, the classification device is configured to stop the sorting operation when it is determined that the statistic exceeds the predetermined allowable range. Outputting the result can be prevented.

According to the third aspect of the present invention, the self-evaluation of the inspection apparatus itself is automatically performed by simply using the actual measurement data measured for the characteristic inspection of the inspection object. Since no special work is required for the self-evaluation of the inspection apparatus itself, labor can be reduced.

According to the fourth aspect of the present invention, when the self-evaluation of the inspection apparatus is performed, the same master sample is repeatedly and repeatedly measured a predetermined number of times, and the self-evaluation of the inspection apparatus is performed based on the measurement result. Is performed, the self-evaluation can be performed very accurately when any difference occurs in the inspection apparatus itself.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an inspection device of the present invention.

FIG. 2 is a schematic circuit diagram showing an example of a device to be inspected.

FIG. 3 is a diagram showing the operation of an example of the apparatus as an object to be inspected.

FIG. 4 is a flowchart showing an operation of the inspection device according to the first embodiment of the present invention.

FIG. 5 is a diagram showing a distribution of measured data of the inspection device.

FIG. 6 is a diagram showing a state in which the average value of the actual measurement data of the inspection device has changed compared to the initial state.

FIG. 7 is a diagram illustrating a state in which the degree of variation in measured data of the inspection apparatus has changed compared to the initial state.

FIG. 8 is a flowchart showing an operation of the inspection device according to the second embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Characteristic evaluation part 2 Self-detection part 3 Display means 11 Characteristic measurement means 12 Classification means 14 Recording means 15 Average value calculation means 16 Average value error detection means 17 Average error tolerance judgment means 18 Dispersion calculation means 19 Dispersion error detection means 21 Dispersion tolerance Judging means 22 Output means 30 ECU 31 Interior light 32 Drive unit 33 Control unit 33a CPU 33b Connection point 34 Door switch 35 PNP transistor MQ Inspection device MS Inspection target

Claims (4)

[Claims] 1. An inspection apparatus for inspecting characteristics of an object to be inspected, comprising: a characteristic measuring means for measuring characteristics of the object to be inspected; and whether the object to be inspected is a non-defective product based on the measured characteristics. Classification means for classifying whether or not the measurement result is recorded; recording means for accumulating and recording the measured values by the characteristic measuring means; and calculating at least one of statistics of an average value and a degree of dispersion of the measured values accumulated and recorded in the recording means. A statistic calculation means, the statistic calculated by the statistic calculation means, and a reference statistic obtained by previously measuring a plurality of the test objects or measuring the single test object a plurality of times. A statistic error detecting means for detecting a difference between the statistic and the statistic, and determining whether or not the statistic exceeds a predetermined allowable range based on a difference in the statistic detected by the statistic error detecting means. Statistics to be judged Disconnecting means, when the statistic allowable range determining means determines that the statistic exceeds the predetermined allowable range, displays that the inspection apparatus itself is in a defective state, and the statistic is An inspection apparatus comprising: a display unit that displays a classification result by the classification unit when it is determined that the difference does not exceed the predetermined allowable range. 2. The inspection apparatus according to claim 1, wherein the classification unit includes a statistic allowable range determination unit.
An inspection apparatus having a function of stopping a sorting operation when it is determined that a statistic exceeds the predetermined allowable range. 3. An inspection method using the inspection apparatus according to claim 1 or 2, wherein a plurality of the objects to be inspected are measured in advance in an initial state, and at least one of an average value and a degree of variation is measured. Calculating a reference statistic, and then, when inspecting an object to be inspected other than the object to be inspected which is the object of the reference statistic, the actual measurement data measured by the characteristic measuring means A step of storing and recording by the recording unit; a step of calculating at least one statistic of the average value and the degree of variation of the actually measured data stored and recorded in the recording unit; and the calculated statistic and the reference statistic. Detecting a difference between the statistic and the statistic, and determining whether or not the statistic exceeds a predetermined allowable range based on the difference between the statistic; and Over When it is determined that the inspection device itself is in a defective state, the display means displays on the display means, and when it is determined that the statistic does not exceed the predetermined allowable range, the inspection target is A step of classifying whether or not the product is non-defective by a classification unit, and displaying a result of the classification by the classification unit on a display unit. 4. An inspection method using the inspection apparatus according to claim 1 or 2, wherein a single object to be inspected is actually measured a plurality of times in advance in an initial state, and the average value and the degree of variation are measured. A step of calculating at least one of the above to obtain a reference statistic, and thereafter, the same single test object as the test object which is the target of the reference statistic is repeatedly measured a predetermined number of times continuously Accumulating and recording data by the recording means; calculating at least one statistic of the average value and the degree of variation of the data accumulated and recorded in the recording means; and the calculated statistic and the reference statistic Detecting a difference between the statistic and the statistic, and determining whether or not the statistic exceeds a predetermined allowable range based on the difference between the statistic; and Over A step of displaying on the display means that the inspection apparatus itself is in a defective state when it is determined that the inspection apparatus is in a defective state.