JP2017126251A - Process evaluation method and device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide means capable of evaluating reliability of a product without use result data or number-of-test data, or with few use result data and number-of-test data.SOLUTION: When one product is manufactured by a plurality of mutually different processes, in a step S1, the plurality of processes are repetitively performed for manufacturing a plurality of products. In a step S2, on the basis of a result of the step S1, process data representing presence/absence of trouble in a process for each of number of execution times of a process in the respective processes. In a step S3, on the basis of the process data recorded in the step S2, trouble rate data of an object process subjected to evaluation, representing a trouble occurrence rate in each execution number section are calculated. In a step S4, on the basis of the trouble rate data calculated in the step S3, a value of a parameter m is calculated when a relation between the number t of execution times of the object process and a cumulative value H(t) of the trouble occurrence rate is represented by an approximate expression: H(t)=(t/η), or H(t)={(t-γ)/η}. The steps S1-S4 are repeated to repetitively calculate the value m.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a method and apparatus for evaluating the maturity stage of a product manufacturing process in order to evaluate the reliability of the product.

  In order to evaluate the reliability of a product, product use record data and product test result data are used. For example, in Patent Document 1 below, a population that is a set of devices shipped to a predetermined shipping destination is obtained, and within the population based on failure occurrence time information (usage record data) and other information of the devices. The failure record of the device that has occurred up to the time of reliability analysis is obtained, and the reliability analysis of the device is performed based on the population and the failure record.

JP 2005-327201 A

  However, it was not possible to evaluate the reliability of products that had little or no usage data or product test result data. For example, space products such as rocket engines and thrusters are required to have high reliability, but there are cases where there is not enough usage record data or test result data to confirm the reliability.

  Therefore, an object of the present invention is to provide a means that can evaluate the reliability of a product that has no or little use record data and test number data.

The inventor of the present application focused not on product use record data but on processes performed for manufacturing a product in order to evaluate the reliability of the product.
In other words, the maturity stage of the process is obtained by Weibull analysis (details will be described later), and the maturity stage of the process is in the stable period and the defect occurrence rate of the process at this time is low. We focused on the fact that it can be said to be highly reliable.

Based on this, the present invention has been made. That is, according to the present invention, a process evaluation method for evaluating a maturity stage of a process performed for manufacturing a product,
One product is manufactured by multiple different processes.
(A) In order to manufacture a plurality of products, the plurality of steps are repeated,
(B) Based on the result of (A) above, record process data indicating the presence or absence of a defect in the process for each process for each process execution count,
(C) Based on the process data recorded in (B) above, for the target process to be evaluated, obtain defect rate data indicating the failure occurrence rate in each execution frequency section,
(D) Based on the defect rate data obtained in (C) above, the relationship between the number of executions t of the target process and the cumulative value H (t) of the defect occurrence rate is expressed by an approximate expression H (t) = (t / η) obtains the value of the parameter m when represented by ^m or H (t) = {(t -γ) / η} m,
A process evaluation method is provided that repeats (A) to (D).

Further, according to the present invention, a process evaluation apparatus for evaluating a maturity stage of a process performed for manufacturing a product,
For a target process to be evaluated, a storage device that stores defect rate data indicating a defect occurrence rate in each execution frequency section;
Based on the failure occurrence data of the storage device, the relationship between the number of executions t of the target process and the failure occurrence rate cumulative value H (t) is approximated by H (t) = (t / η) ^m or H (t ) = {(T−γ) / η} An operation device that calculates a value of the parameter m when represented by ^m is provided.

According to the above-described present invention, in the process of repeatedly executing the process for manufacturing the product, the target process execution count t based on the fault rate data indicating the fault occurrence rate in each execution count section for the target process. parameter m in the case represented by the accumulated value of the defect occurrence rate H (t) approximation equation the relationship between H (t) = (t / η) m or H (t) = {(t -γ) / η} m Find the value of.
The obtained value of m indicates the maturity stage of the process. The value of m is repeatedly obtained. According to the Weibull analysis, when m is 1, it can be said that the maturation stage of the process has entered the stable period. If the defect occurrence rate takes a low value at this time, it can be evaluated that the product manufactured at this time has a particularly high reliability.
Therefore, it becomes possible to evaluate the reliability of products that have no or little use record data and test result data.

It is a flowchart which shows the process evaluation method by embodiment of this invention. Process data is shown. (A) shows defect rate data, and (B) shows cumulative value data. It is a graph for showing the relationship between the execution frequency t of the target process and the cumulative value H (t) of the defect occurrence rate. It is a graph of the bathtub curve showing the relationship between the execution frequency t and a malfunction rate. It is a graph for showing the relationship between the frequency | count t of execution and the cumulative value H (t) by an Example. It is a block diagram of the process evaluation apparatus by embodiment of this invention.

  Embodiments of the present invention will be described with reference to the drawings. In addition, the same code | symbol is attached | subjected to the common part in each figure, and the overlapping description is abbreviate | omitted.

  In this embodiment, the maturity stage of the process is evaluated by Weibull analysis. Just as a product is a series of elements, a process can be considered as a series of elements such as procedures, workers, equipment (machines and jigs). Thus, the process can be evaluated in the same way as the product. A product failure occurs when a load greater than the strength of the product is applied. Looking at this phenomenon from the process side, not the product side, if the process procedure or equipment is set improperly, or if a force exceeding the actual capability (for example, rated value) is applied to the process equipment, the process will fail. It is thought to occur.

  Here, the defect means that the quality of the product is affected because the process is not properly executed, or the process is appropriately executed because the set process itself is inappropriate. Is also affecting the quality of the product. For example, in the former case, when a person drops a product part on the floor and the part is damaged during the execution of the process, when the person inputs the load condition of the environmental test of the product to the device, the condition exceeds the product ability If the product is erroneously input and the product is overloaded and damaged, the process may be a product performance test, and the measurement device used in the test may not operate. In the latter case, if the flow rate test is performed without setting a test method with sufficient accuracy for the allowable tolerance range of the flow rate test and only a non-reproducible result is obtained, the order of cutting is appropriate. In some cases, the residual distortion remains and is deformed without processing.

  The process evaluation method of this embodiment is based on the assumption that the reliability of the manufactured product is high when the maturity stage of the process is in the stable period and the defect occurrence rate of the process at this time is low. Assess the maturity stage of the process performed to produce the product.

  FIG. 1 is a flowchart of a process evaluation method according to an embodiment of the present invention.

  In step S1, a plurality of different processes are repeatedly performed in order to manufacture a plurality of identical (for example, the same model) products. Since one product is manufactured by a plurality of different processes, in step S1, a plurality of processes are performed (for example, once) each time one product is manufactured.

  In step S2, process data indicating whether or not a defect has occurred in the process is recorded for each process execution number for each process based on the result of step S1. This process data is, for example, shown in FIG. In FIG. 2, the presence / absence of occurrence of a defect is recorded for each process for each execution count. A cross indicates that a failure has occurred, and a blank indicates that no failure has occurred. The number of executions is a number indicating how many times the process has been performed since the flow of FIG. 1 was started.

  In the example of FIG. 2, ten steps are performed to manufacture one product. That is, the process of process numbers 1-10 is performed. Therefore, in this case, in step S1, ten different processes are repeated for the number of products to be manufactured. As these processes, there are assembly of each part constituting a product, cutting of product members and parts, welding of product constituent members, performance test of a finished product, and the like.

  In step S3, based on the process data recorded in step S2, the defect rate data indicating the defect occurrence rate in each of a plurality of continuous execution frequency sections included in the evaluation target range for the target process to be evaluated. Ask.

The evaluation target range is a range of the number of executions for evaluating the maturity stage. The evaluation target range may be freely specified. Step S3 and step S4 described later are performed on the evaluation target range. The evaluation target range may be a range of all execution times (in the example of FIG. 2, the execution number is a range of 1 to 10), or a specific range (that is, a part of the range of all execution times). It may be.
For example, in FIG. 2, it is assumed that the design change or systematic improvement regarding the target process is performed immediately before the number of executions 6 and the state of the target process becomes discontinuous. In this case, the specific range may be a range of 6 to 10 execution times.
Note that FIG. 3 to be described later is data in a case where the evaluation target range is the range of all execution times in FIG.

  In step S3, in the defect rate data, step S3 and step S4, which will be described later, may be performed with a process in which the occurrence frequency of the defect exceeds a set value as a target process. Here, the frequency of occurrence of defects may be, for example, the frequency over all execution times. Thereby, steps S3 and S4 can be omitted for a process with a low occurrence frequency of defects. That is, since the process with a low occurrence frequency of defects is highly reliable, steps S3 and S4 are not necessary.

  Or step S3 and below-mentioned step S4 may be performed by making each process (all processes) for manufacturing a product into an object process. As a result, as described later, it can be evaluated whether or not the maturation stage has been stabilized for all the processes.

  FIG. 3A shows defect rate data obtained based on the process data of FIG. 2 with a process number of 8 as the target process.

For example, as shown in FIG. 3A, a section of the number of executions until a failure occurs may be set as one execution number section. That is, a process in which a problem has occurred may be the last execution number in one execution number section, and a process next to the process in which a problem has occurred may be a first execution number in the next execution number section. However, according to the present invention, each execution frequency section may be arbitrarily determined.
In FIG. 3A, an execution count section 1 (subscript 1 is the number of this section. The same applies hereinafter) is a section composed of one execution count (that is, execution count 1). Since a failure occurs at the number of executions of 1, the failure occurrence rate is 100%. The execution frequency section 2 is a section composed of two execution times (that is, the execution number 1 and the execution number 2). Since one defect occurs in this interval, the defect occurrence rate is 50%. The same applies to other execution frequency sections.

In step S4, based on the defect rate data obtained in step S3 by Weibull analysis, the relationship between the number of executions t of the target process and the cumulative value H (t) of the defect occurrence rate is expressed by an approximate expression H (t) = (t / eta) determining the values of the parameters m and eta when it represents by ^m. The obtained value of m is an index indicating the maturity stage of the target process.

For example, in step S4, for each of a plurality of execution counts t that are a part of all the execution counts t based on the defect rate data obtained in step S3, each execution count included in the range of the execution count t or less. Calculate the sum of the failure occurrence rates found in the section. This sum is taken as a cumulative value H (t) of the defect occurrence rate corresponding to the execution count t. The values of parameters m and η when the relationship between the plurality of cumulative values H (t) obtained in this way and the number of executions t corresponding thereto are expressed by the approximate expression H (t) = (t / η) ^m Ask for.

  FIG. 3B shows data of the cumulative value H (t) of the failure occurrence rate obtained based on the failure rate data of FIG. 3A (hereinafter referred to as cumulative value data). In the accumulated value data H (t) of FIG. 4, the accumulated value H (t) is obtained for each number of executions (t = 1, 3, 6, 7, 10) where a failure has occurred. That is, the cumulative value H (t) corresponding to the execution count t = 1 is 100% of the total failure occurrence rate 100% (see FIG. 3A) in the execution count section 1 up to the execution count 1. The cumulative value H (t) corresponding to the number of executions t = 3 is a total of 150% of the failure occurrence rates 100% and 50% (see FIG. 3A) in the execution number sections 1 and 2 up to the number of executions 3. is there. The same applies to the other execution times (t = 6, 7, 10).

When the cumulative value data in FIG. 3B is used, step S4 includes the number of executions of the target process (t = 1, 3, 6, 7, 10) and the cumulative value of the failure occurrence rate corresponding to each (H = 1). (T) = 100.0, 150.0, 183.3, 283.3, 316.7) When the relationship between the parameters m and η is expressed by the approximate expression H (t) = (t / η) ^m Find the value.

FIG. 4 shows a coordinate system in which the horizontal axis represents the number of executions t and the vertical axis represents H (t). In this coordinate system, the interval between the scales on the horizontal axis and the vertical axis is adjusted so that H (t) = (t / η) ^m is represented by a straight line. In FIG. 4, a plurality of black circles indicate a plurality of cumulative values H (t) in FIG. In the example of FIG. 4, parameters m and η for approximating the relationship between the horizontal axis coordinate and the vertical axis coordinate of each black circle by a straight line H (t) = (t / η) ^m are obtained. In FIG. 4, as a result of the human drawing the approximate straight line, η = 10.63 and m = 0.49. The parameters m and η can also be obtained so that the sum of the distances in the vertical axis direction between each black circle and the approximate expression graph is minimized.

  When step S4 is completed, the process returns to step S1. In this way, the value of m is repeatedly obtained by repeating steps S1 to S4.

(Operational effects according to the embodiment)
According to the above-described embodiment, in the process of repeatedly executing the process for manufacturing the product, the defect rate data indicating the defect occurrence rate in each execution frequency section is obtained for the target process, and the target is based on this data. The value of the parameter m in the case where the relationship between the process execution count t and the cumulative failure rate H (t) is expressed by the approximate expression H (t) = (t / η) ^m is obtained.

  According to the Weibull analysis, the obtained value of m indicates the maturity stage of the process. That is, according to the Weibull analysis, the number of executions and the failure occurrence rate are represented by a bathtub curve shown in FIG. In FIG. 5A, when the value of m is smaller than 1, the execution count t is in the range A (initial failure occurrence period). In FIG. 5A, when the value of m is 1, the execution count t is in the range B (stable period). In FIG. 5A, when the value of m is greater than 1, the execution count t is in the range C (deterioration period).

  By repeatedly performing steps S1 to S4 as described above, the value of m is repeatedly obtained. As shown below, it is possible to evaluate the maturity stage of the process based on the value of m.

  When the obtained value of m is smaller than 1, the maturity stage of the process is at the time when the initial failure occurs (range A in FIG. 5A). Therefore, it is still required to correct the target process. That is, there is a need for a measure for eliminating a problem that tends to occur when the number of executions of the target process is small. In other words, the defect occurrence rate of the target process decreases as the procedure and equipment (machine or jig) for performing the target process are corrected, or a person with a higher skill level performs the target process.

  When the obtained value of m is 1, the maturation stage of the process is in the stable period (range B in FIG. 5A). If the defect occurrence rate takes a low value at this time, the product manufactured at this time can be evaluated as having high reliability. Since this evaluation is based on the maturity stage of the process, it is possible even for products that have little or no usage record data.

  In the stable period, even if the target process is corrected, the defect occurrence rate does not decrease. That is, in the stable period, troubles occur due to accidental factors. For example, when a person's skill is required to execute the target process, a human error that occurs regardless of the skill level or a malfunction due to an electrical factor occurs. In order to reduce the defect occurrence rate, improvement by design improvement and technological development is necessary.

  When the obtained value of m is larger than 1, the maturation stage of the process is in the deterioration period (range C in FIG. 5A). Therefore, at this time, the rate of occurrence of defects in the target process increases due to deterioration (for example, wear) of equipment used in the process.

(Example)
The process evaluation method described above was applied to the production of a two-component thruster. This thruster is a product manufactured by a plurality of processes. This thruster is mounted on a satellite launched from the ground by a rocket. In addition, this thruster mixes fuel and oxidant, and generates a propulsive force by a chemical reaction (combustion) of both. With this propulsive force, the satellite is put into orbit around the earth.

  The above-described process evaluation method was performed with a process having a relatively high defect occurrence rate as a target process among the manufacturing processes of the thrusters. Thereby, the result of FIG. 6 was obtained. FIG. 6 corresponds to FIG. 4, and a plurality of black circles in FIG. 6 are coordinates indicating a plurality of execution times t and a cumulative value H (t) corresponding thereto.

  The value of the parameter m obtained based on all the black circles in FIG. 6 was 1.49.

  On the other hand, in FIG. 6, a plurality of black circles in the vicinity of the black circle (hereinafter referred to as the latter half area) with the highest number of executions are located on substantially the same straight line. Therefore, it can be said that the maturity stage of the process is close to the stable period in the latter half. However, in the latter half area, the frequency of malfunctions was high, so the malfunction rate is stable and high. That is, it can be said that the latter half area is near the stable period in the solid bathtub curve shown in FIG. Note that the dashed bathtub curve shown in FIG. 5B indicates a case where the failure occurrence rate is low in the stable period.

  A high defect occurrence rate in the stable period as shown by the solid line in FIG. 5B indicates that the process content is inappropriate for the product. Therefore, in this case, it is required to improve the performance of equipment used in the process, review the standards in the process, or improve the design and technology related to the process.

(Process evaluation equipment)
Steps S2 to S4 of the process evaluation method described above may be performed by a person or may be performed using the process evaluation apparatus 10 described below.
FIG. 7 is a block diagram of the process evaluation apparatus 10 for performing the above-described process evaluation method.

  The input device 3 is for inputting the process data. The input device 3 includes an operation unit (such as a keyboard, a mouse, and a touch panel), and the above-described process data is input when a person operates the operation unit.

  The storage device 5 stores the process data input by the input device 3. That is, the above-described step S2 is performed using the input device 3, and the process data is recorded in the storage device 5.

  The arithmetic device 7 obtains the defect rate data for the target process based on the process data stored in the storage device 5. That is, the above-described step S3 is performed by the arithmetic unit 7. The defect rate data obtained by the arithmetic device 7 is stored in the storage device 5.

In addition, the arithmetic unit 7 calculates the relationship between the number of executions t of the target process and the cumulative value H (t) of the defect occurrence rate based on the defect rate data in the storage device 5 as an approximate expression H (t) = (t / η ) Determine the values of the parameters m and η when represented by ^m . That is, the above-described step S4 is performed by the arithmetic unit 7. The arithmetic unit 7 outputs the obtained value of m. The output value of m may be displayed on a display or stored in an appropriate storage device.

For example, the arithmetic unit 7 obtains the above cumulative value data based on the defect rate data of the storage device 5, and based on this cumulative value data, the relation between t and H (t) is expressed by the approximate expression H ( ^t) = (t / η) may determine the value of the parameter m when represented by ^m.

  Preferably, the process evaluation device 10 includes a range specifying device 9 that specifies the above-described evaluation target range. The range specifying device 9 has an operation unit (for example, a keyboard, a mouse, a touch panel, etc.) operated by a person. A person can specify the evaluation target range by operating the operation unit. The designated evaluation target range is input to the arithmetic device 7. As described above, the arithmetic unit 7 obtains the defect rate data for the target process and obtains the value of the parameter m for the evaluation target range.

In addition, the arithmetic device 7 identifies a process whose defect occurrence frequency exceeds a set value as a target process based on the process data in the storage device 5, and obtains the defect rate data for the target process as described above. Then, the value of m may be obtained and output.
Alternatively, the arithmetic unit 7 specifies all processes as target processes based on the process data in the storage device 5, obtains the defect rate data as described above for each target process, and then calculates the value of m. You may ask and output.

  The present invention is not limited to the above-described embodiment, and various changes can be made without departing from the scope of the present invention. For example, any of the following modified examples 1 to 3 may be adopted, or two or more of modified examples 1 to 3 may be arbitrarily combined and employed. In this case, the other points may be the same as described above.

(Modification 1)
In the process evaluation device 10, the arithmetic unit 7 does not have to perform processing for obtaining defect rate data based on the process data. In this case, the defect rate data may be calculated by a person and stored in the storage device 5 using an appropriate input device.

(Modification 2)
There are two types of process defects, one related to the person performing the process, one related to equipment used in the process, and one related to the procedure of the process. For each of these types, process data may be recorded in step S2. That is, steps S2 to S4 may be performed for each type of defect.

(Modification 3)
H (t) = {(t−γ) / η} ^m may be used instead of the approximate expression H (t) = (t / η) ^m described above. In this equation, the symbol γ is a parameter, and the other symbols are the same as described above.

3 input device, 5 storage device, 7 arithmetic device, 9 range specification device, 10 process evaluation device

Claims (5)

In order to make this evaluation, according to the present invention, a process evaluation method for evaluating the maturity stage of a process performed for manufacturing a product,
One product is manufactured by multiple different processes.
(A) In order to manufacture a plurality of products, the plurality of steps are repeated,
(B) Based on the result of (A) above, record process data indicating the presence or absence of a defect in the process for each process for each process execution count,
(C) Based on the process data recorded in (B) above, for the target process to be evaluated, obtain defect rate data indicating the failure occurrence rate in each execution frequency section,
(D) Based on the defect rate data obtained in (C) above, the relationship between the number of executions t of the target process and the cumulative value H (t) of the defect occurrence rate is expressed by an approximate expression H (t) = (t / η) obtains the value of the parameter m when represented by ^m or H (t) = {(t -γ) / η} m,
The process evaluation method which repeats said (A)-(D).   The process evaluation method according to claim 1, wherein (C) and (D) are performed on a target process with respect to a specific range of the number of executions.   The process evaluation method according to claim 1, wherein the steps (C) and (D) are performed with a process in which the occurrence frequency of a defect exceeds a set value as a target process.   The process evaluation method according to claim 1 or 2, wherein the processes (C) and (D) are performed with each process for manufacturing a product as a target process. A process evaluation apparatus for evaluating a maturity stage of a process performed for manufacturing a product,
For a target process to be evaluated, a storage device that stores defect rate data indicating a defect occurrence rate in each execution frequency section;
Based on the failure occurrence data of the storage device, the relationship between the number of executions t of the target process and the failure occurrence rate cumulative value H (t) is approximated by H (t) = (t / η) ^m or H (t ) = {(T−γ) / η} An arithmetic device that obtains the value of the parameter m when represented by ^m .

Images