JP6048082B2 - Extraction apparatus, extraction program, and extraction method - Google Patents

Description

  The present invention relates to an extraction device, an extraction program, and an extraction method, for example, an extraction device, an extraction program, and an extraction method for extracting a pass / fail judgment section.

  Work using robots such as industrial robots is widely performed. It is known that the quality of work using a robot is determined based on the feature amount associated with the quality of work (for example, Patent Document 1).

JP 2011-230245 A

  If it is determined that the work is good or bad, the work is interrupted or the work is defective. However, if an event that does not affect the downstream is determined to be defective, a delay due to interruption of work or a loss due to determining that a non-defective product is determined to be defective occurs.

  An object of the present extraction apparatus, extraction program, and extraction method is to appropriately perform work quality determination.

A step of performing a first work divided into a plurality of first sections in time series on an object, and determining whether the first work is good or bad in order to make the object on which the first work is performed good or bad For the work including the step of performing a second pass / fail determination and the step of performing a first pass / fail determination to determine pass / fail of the object that has performed the first work after the second pass / fail determination. a extractor for extracting a second interval for performing the second quality determination of the interval, when the first work to a plurality of first object was good in the first quality determination is performed sensor using a plurality of first data sequence is a data sequence outputted in time series, a calculation unit for calculating an extent rows of time series, among the plurality of first sections, a failure in the first quality determination the first film relative to the second object was Characterized by including the, an extraction unit second data sequence a section becomes the range Retsugai is extracted as the second section is a data sequence outputted in time series the sensor when the has been made An extraction device, an extraction program, and an extraction method are used.

  According to the extraction device, the extraction program, and the extraction method, it is possible to appropriately perform work quality determination.

FIG. 1A is a block diagram of a robot system in which the first embodiment is used, and FIG. 1B is a diagram illustrating an example of a robot in which the first embodiment is used. FIG. 2 is a block diagram of a computer that functions as an extraction apparatus according to the first embodiment. FIG. 3 is a flowchart of an operation for an object in which the first embodiment is used. FIG. 4 is a flowchart showing the processing of the computer. FIG. 5A to FIG. 5C are diagrams showing sensor outputs as time-series data strings. FIG. 6 is a flowchart of the extraction method according to the second embodiment. . FIG. 7A and FIG. 7B are diagrams showing sensor outputs as time-series data strings. FIG. 8 is a flowchart of the extraction method according to the third embodiment. FIG. 9A and FIG. 9B are diagrams showing sensor outputs as time-series data strings. FIG. 10 is a flowchart of the extraction method according to the fourth embodiment. Fig.11 (a) to FIG.11 (c) is a figure (the 1) which shows the content of work. FIG. 12A to FIG. 12C are diagrams (part 2) showing the contents of the work. FIG. 13 is a diagram showing the calculated range column. FIG. 14 is a diagram illustrating a second data string corresponding to the second object. FIG. 15 is a diagram illustrating a differential column of the average column. FIG. 16 is a diagram illustrating the first section and the second data string.

  Embodiments will be described below with reference to the drawings.

  FIG. 1A is a block diagram of a robot system in which the first embodiment is used, and FIG. 1B is a diagram illustrating an example of a robot in which the first embodiment is used. As shown in FIG. 1A, the robot system 100 includes a robot 22, a sensor 24, and a computer 10. The robot 22 is an industrial robot, for example. As shown in FIG. 1B, the robot 22 includes a stage 25 and a manipulator 26, for example. The manipulator 26 performs work using, for example, a tool 28. The tool 28 is tweezers, for example. The stage 25 supports the manipulator 26. The manipulator 26 is provided with a sensor 24. The sensor 24 is a strain sensor that detects strain of the manipulator 26, for example. As the sensor 24, for example, a 3-axis or 6-axis force sensor may be used. In this case, the sensor 24 detects the force vector and / or torque of the action point. The sensor 24 may be a load sensor, a pressure sensor, an acceleration sensor, or a microphone. Further, the sensor 24 may be a one-dimensional image sensor, or a two-dimensional image sensor such as a CCD (Charge Coupled Device) sensor or a CMOS (Complimentary Metal Oxide Semiconductor) sensor.

  FIG. 2 is a block diagram of a computer that functions as an extraction apparatus according to the first embodiment. The computer 10 executes an extraction method and an extraction program. The computer 10 includes a central processing unit (CPU) 11 that is a processor, a display device 12, an input device 13, and an output device 14. The computer 10 further includes a main storage device 15, a hard disk drive (HDD) 16, a storage medium drive 17, a communication interface 18, and an internal bus 19. The display device 12 includes a display panel such as a liquid crystal panel, for example, and displays processing results and the like. The input device 13 is, for example, a keyboard, a mouse, and a touch panel, and inputs processing data and the like. The output device 14 is a printer, for example, and outputs a processing result or the like. The main storage device 15 is a volatile memory such as a DRAM (Dynamic Random Access Memory), for example, and stores data being processed. The HDD 16 stores, for example, data during or after processing. The storage medium drive 17 is used when a program stored in the storage medium 21 is installed. Alternatively, the processed data is stored in the storage medium 21. The communication interface 18 is connected to other computers and transmits / receives data to / from other computers. The internal bus 19 connects each device in the computer 10. The computer 10 functions as a calculation unit, an extraction unit, a division unit, and a determination unit in cooperation with software.

  A portable storage medium can be used as the storage medium 21 readable by the computer 10 storing the program. As the portable storage medium, for example, a CD-ROM (Compact Disc Read Only Memory) disc, a DVD (Digital Video Disc) disc, a Blu-ray disc, or a USB (Universal Serial Bus) memory can be used. As the storage medium 21, a flash memory or an HDD may be used.

  FIG. 3 is a flowchart of an operation for an object in which the first embodiment is used. Referring to FIG. 3, the robot 22 performs a first operation on the object (step S10). There may be other work before step S10. Next, the determination device or the operator performs the second pass / fail determination of the first work in step S10 (step S12). If the determination is bad, the operation in step S10 is interrupted and the cause of the bad determination is investigated. Moreover, the product which performed the 1st operation | work is made into defect. If the determination is good in step S12, another second operation is performed on the object (step S14). The second work may be performed by the same robot. Another robot may do it. Alternatively, an operator may perform it. The second operation may not be performed. The determination device or the operator performs the first pass / fail determination of the object (step S16). For example, when the determination is bad, the object is determined as a defective product. If the determination is good, the object is determined to be good. The process from step S10 to S16 is performed on a plurality of objects.

  FIG. 4 is a flowchart showing processing of the computer 10. FIG. 5A to FIG. 5C are diagrams showing sensor outputs as time-series data strings. First, the robot 22 performs a first operation on the object based on the teaching data. The sensor 24 detects a detection amount related to the robot 22. For example, in the example of FIG. 1B, the robot 22 performs the first operation on the object by operating the manipulator 26 based on the teaching data. The sensor 24 detects distortion of the manipulator 26. As shown in FIGS. 5A to 5C, the period 52 during which the robot 22 performs work is divided into a plurality of first sections 50a to 50g.

  Referring to FIG. 4, the calculation unit of computer 10 obtains a plurality of first data strings (step S20). The first data sequence is a time-series data sequence output by the sensor 24 when the first operation in step S10 is performed on the first object determined to be good in the first pass / fail determination in step S16 of FIG. is there. Referring to FIG. 5A, one first data string 30 indicating sensor output with respect to time is shown. The calculation unit acquires the first data string 30 for a plurality of first objects. Referring to FIG. 4, next, a time series range string is calculated using a plurality of first data strings 30 (step S22). With reference to FIG. 5B, the calculation unit calculates the average column 32 and the range column 37. The average column 32 is an average of the plurality of first data columns 50. The range column 37 is a range between the upper limit column 34 positioned above the average column 32 and the lower limit column 36 positioned below. For example, the upper limit column 34 and the lower limit column 36 may be columns obtained by adding or subtracting a positive constant (for example, 3) × standard deviation from the average column 32 of the plurality of first data columns 30.

  Referring to FIG. 4, next, the extraction unit of the computer 10 acquires the second data string (step S24). The second data string is a time-series data string output by the sensor 24 when the first operation in step S10 is performed on the second object determined to be defective in the first pass / fail determination in step S16 of FIG. is there. Referring to FIG. 5C, a second data string 35 indicating the sensor output with respect to time is shown. Referring to FIG. 4, the extraction unit then extracts the first section where the second data string is out of the range string as the second section (step S <b> 26). The second section is a section for performing the second pass / fail determination in step S12 of FIG. Referring to FIG. 5C, as indicated by an arrow 54, the second data string 35 is out of the range string 37 in the first sections 50f and 50g among the first sections 50a to 50g. Therefore, the extraction unit sets the first sections 50f and 50g as the second sections 56a and 56b, respectively.

  Referring to FIG. 4, the determination unit of computer 10 determines the determination criteria in second sections 56a and 56b (step S28). The determination unit may use the range column 37 in the second sections 56a and 56b as a determination criterion for the second pass / fail determination. Also, other determination criteria may be determined. For example, the determination criterion may be determined based on the output of another sensor. Further, the quality may be determined by an inspection such as a visual inspection performed by an operator. Further, step S28 may be executed by a computer other than the computer 10. Further, the operator may execute step S28.

  According to the first embodiment, a time-series range column 37 is calculated using a plurality of first data columns 30 as in step S22 of FIG. As in step S26, among the plurality of first sections 50a to 50g, sections where the second data string 35 is outside the range string 37 are extracted as second sections 56a and 56b. Thereby, the 1st area which influences the 1st quality determination of step S16 among the period 52 of the 1st operation | work of step S10 of FIG. 3 can be extracted as a 2nd area. By performing the second quality determination in the extracted second section, the quality determination of the first work can be appropriately performed.

  In addition, the extraction unit extracts the earliest first section 50f outside the range column 37 as the second section 56a. As a result, the first pass / fail determination can be made at an early point in the work period 52.

  Further, as shown in FIG. 3, the first pass / fail determination (step S16) is performed after the second pass / fail determination (step S12) and the second work (step S14) different from the first work.

  FIG. 6 is a flowchart of the extraction method according to the second embodiment. FIG. 7A and FIG. 7B are diagrams showing sensor outputs as time-series data strings. With reference to FIG. 6, the range column 37 is calculated by the extraction method of the first embodiment, and the second section is extracted. The robot system 100 performs the first and second operations in FIG. 3 and the first and second pass / fail judgments on the third object (step S50).

  The extraction unit determines whether the first quality determination of the third object is bad and the second quality determination is good (step S52). If no, the process ends. In Yes, an extraction part acquires the 3rd data sequence which is the 1st data sequence of the 3rd subject (Step S54). Referring to FIG. 7A, the first section 50f is the second section 56a. The third data string 38 is in the range string 37 in the second section 56a. However, it is bad in the first pass / fail judgment.

  Referring to FIG. 6, the extraction unit newly extracts the second section based on the third data string 38 (step S56). With reference to FIG. 7B, the extraction unit extracts the first section 50d in which the third data string 38 is out of the range string 37 as the second section 56b. Then, the process ends.

  According to the second embodiment, as in step S52 of FIG. 6, when the work is performed on the third object, the second pass / fail judgment in the second section 56a is good and the first pass / fail judgment is bad. Determine if there is. In the case of Yes, as in step S <b> 56, the extraction unit selects a section in which the third data column 38 when working on the third object is out of the range column 37 among the plurality of first sections. Extracted as the second section 56b. Thereby, the 2nd area 56b can be added automatically.

  FIG. 8 is a flowchart of the extraction method according to the third embodiment. FIG. 9A and FIG. 9B are diagrams showing sensor outputs as time-series data strings. Referring to FIG. 8, the range column 37 is calculated by the extraction method of the first embodiment, and the second section is extracted. The robot system 100 performs the first and second operations in FIG. 3 and the first and second pass / fail judgments on the fourth object (step S60).

  The calculation unit determines whether the first quality determination of the fourth object is good and the second quality determination is good (step S62). If no, the process ends. In Yes, an extraction part acquires the 4th data sequence which is the 1st data sequence of the 4th subject (Step S64). With reference to FIG. 9A, the first section 50 f is the second section 56. As indicated by the arrow 58, the fourth data string 40 is out of the range string 37 in the first section 50d.

  Referring to FIG. 8, the calculation unit newly calculates range column 37 based on fourth data sequence 40 (step S66). With reference to FIG. 9B, the calculation unit calculates the range column 37 in the first section 50 d in which the fourth data sequence 40 deviates from the range column 37 as indicated by an arrow 42. For example, the calculation unit performs statistical processing again to calculate the range column 37. Alternatively, the range column 37 is calculated so that the fourth data column 40 does not fall outside the range column 37. Then, the process ends. Thereafter, when the extraction unit extracts the second section, the newly calculated range column 37 is used.

  According to the third embodiment, as in step S62 of FIG. 8, when the work is performed on the fourth object, the second pass / fail judgment in the second section is good and the first pass / fail judgment is good. Determine whether. In the case of Yes, as in step S66, the calculation unit calculates the range column 37 based on the fourth data column 40 in the plurality of first sections of the fourth object. Thereby, the range column 37 can be automatically recalculated.

  FIG. 10 is a flowchart of the extraction method according to the fourth embodiment. Referring to FIG. 10, the computer 10 acquires teaching data from the outside or the HDD 16 (step S30). The teaching data is data for operating the robot 22. The computer 10 operates the robot 22 in accordance with the teaching data for each object, and acquires the sensor output at this time as a data string (step S32). The computer 10 determines whether it is the last data string (step S34). For example, the number N of first data strings corresponding to first objects that are good in the first pass / fail judgment, the number M of second data strings corresponding to second objects that are bad in the first pass / fail judgment, Is determined to be Yes when each becomes more than a predetermined number. Further, it may be added on condition that the number N is larger than the number M. In No, it returns to step S32 and performs step S32 about the next target object.

  In step S34, in the case of Yes, the calculation unit acquires a plurality of first data strings 30 (step S20). The calculation unit calculates the range column 37 using the plurality of first data columns 30 (step S22). For example, as described in FIG. 5B, the calculation unit calculates the range row 37. The dividing unit of the computer 10 divides the period 52 into a plurality of first sections 50a to 50g based on the average string 32 obtained by averaging the plurality of first data strings 30 (step S36). For example, the dividing unit divides the period into the first interval in the interval where the differential sequence obtained by differentiating the average sequence 32 with respect to time is outside the threshold range and the peak time of the differential sequence. Details will be described later in the fifth embodiment.

  The extraction unit acquires one of the second data strings (step S24). The extraction unit extracts the second section (step S26). For example, as described in FIG. 5C, the extraction unit extracts the second section. The extraction unit determines whether it is the last second data string (step S40). If yes, end. In No, it returns to step S24 and performs step S24 to S26 about the next 2nd data sequence.

  By extracting the second section for each of the plurality of second data strings, it is possible to extract the second section corresponding to the plurality of failure modes of the second object that has failed in the first pass / fail determination. When the second section is extracted in step S26, only the earliest first section can be set as the second section. The extraction unit may further select a second section used for the second pass / fail determination from the second sections extracted from the plurality of second data strings.

  Example 5 is an example in which the work of peeling the release paper of the double-sided tape is performed. Fig.11 (a) to FIG.12 (c) is a figure which shows the content of work. The robot 22 performs work using tweezers 76. The tweezers 76 are operated by the manipulator 26. The sensor 24 detects distortion of the manipulator 26. Referring to FIG. 11A, a double-sided tape 75 is affixed to the stage 70. The double-sided tape 75 has an adhesive portion 72 and a release paper 74. The bonding part 72 is bonded to the stage 70. The stage 70 and the double-sided tape 75 correspond to the object. With reference to FIG. 11B, the robot 22 places the tip of the tweezers 76 against the stage 70. With reference to FIG. 11C, the robot 22 inserts the tweezers 76 between the release paper 74 and the adhesive portion 72.

  Referring to FIG. 12A, the robot 22 picks the release paper 74 using the tweezers 76. Referring to FIG. 12B, the robot 22 peels the release paper 74 from the adhesive portion 72 using tweezers 76. Referring to FIG. 12C, the robot 22 moves the peeled release paper 74 to a designated position.

  In steps S30 to S34 in FIG. 10, it is determined whether or not the work is normally performed after the work in FIGS. 11 (a) to 12 (c). When the release paper 74 was peeled off, it was judged as good, and when the release paper 74 was not completely peeled off, it was judged as defective. The operation was continued until there were 10 good first data strings and 1 bad second data string.

  FIG. 13 is a diagram showing the calculated range column. Referring to FIG. 13, the voltage is an output voltage from the sensor 24, and one point corresponds to 20 milliseconds. The sections 60a to 60e correspond to the sections where the operations of FIGS. 11B to 12C are performed, respectively. In steps S20 and S22 of FIG. 10, the calculation unit calculates the average sequence 32 from the average of 10 first data sequences corresponding to the 10 first objects. The upper limit column 34 and the lower limit column 36 respectively correspond to columns obtained by lowering 3 × standard deviation of 10 first data columns from the average column 32. A range column 37 is between the upper limit column 34 and the lower limit column 36.

  FIG. 14 is a diagram illustrating a second data string corresponding to the second object. The second data string 35 is out of the range string 37 in the sections 60a to 60c.

  FIG. 15 is a diagram illustrating a differential column of the average column. Referring to FIG. 15, in step S <b> 36 of FIG. 10, the dividing unit calculates a differential sequence 44 obtained by differentiating the average sequence 32 with respect to time. The dividing unit extracts a section 47 where the differential sequence 44 is outside the range 46. The dividing unit sets the maximum or minimum time in the section 47 as the dividing point 48. When the interval between the plurality of sections 47 is within a predetermined period, the dividing unit sets the larger absolute value of the average column 32 as the dividing point 48. For example, sections 47a and 47b are close to each other in FIG. Therefore, the dividing unit sets a division point 48a that has a larger absolute value in the average column 32 among the peaks in the sections 47a and 47b. The dividing unit may set the higher absolute value of the differential sequence 44 as the dividing point 48a. The dividing unit sets the interval between the division points 48 as the first sections 50a to 50c. In the example of FIG. 15, the period 62 is divided into three first sections 50a to 50c.

  The dividing unit may divide the period 62 into first sections using a column obtained by differentiating the average column 32 a plurality of times. Furthermore, the period 62 may be divided into first intervals using the average column 32 itself. As described above, the dividing unit can divide the period 62 into the first sections based on the average column 32. Thereby, the period 62 can be automatically divided into the first sections. Further, the dividing unit may divide the period 62 into the first section based on the section where the second data string deviates from the range string 37 and the section where the second data string does not deviate.

  FIG. 16 is a diagram illustrating the first section and the second data string. Referring to FIG. 16, in steps S <b> 24 and S <b> 26 of FIG. 10, the extraction unit extracts the first section where the second data string 35 deviates from the range string 37 as the second section. In the first sections 50a and 50b, the second data string 35 deviates from the range string 37. Therefore, the extraction unit extracts the first first section 50 among the first sections 50 a and 50 b as the second section 56.

  As in the fifth embodiment, the second section 56 can be extracted from the average column and the range column. Thereby, the quality of work can be determined appropriately.

  Although the embodiments of the present invention have been described in detail above, the present invention is not limited to such specific embodiments, and various modifications and changes can be made within the scope of the gist of the present invention described in the claims. It can be changed.

In addition, the following additional notes are disclosed regarding the above description.
(Supplementary note 1) Using a plurality of time-series first data strings output by the sensor when a first operation is performed on a plurality of first objects that are good in the first pass / fail judgment, a time-series range string is obtained. Of the plurality of first sections obtained by dividing the period of the plurality of first data strings in time series, the first operation is performed on the second object that is defective in the first pass / fail determination. An extraction unit that extracts a section in which the time-series second data string output by the sensor is out of the range string as a second section for performing a second pass / fail determination in the first work, An extraction apparatus comprising:
(Additional remark 2) The said extraction part extracts the earliest 1st area which became out of the said range row | line | column as said 2nd area, The extraction apparatus of Additional remark 1 characterized by the above-mentioned.
(Supplementary note 3) The section extracting device according to supplementary note 1 or 2, wherein the first pass / fail determination is performed after a second work different from the first work is performed after the second pass / fail determination.
(Additional remark 4) Based on the average row | line | column which averaged the said some 1st data row | line | column, it has a division part which divides | segments the said period into the said several 1st area, It is any one of Additional remark 1-3 characterized by the above-mentioned. The extraction device described.
(Additional remark 5) When the said 2nd quality determination in the said 2nd section is good when the said 1st operation | work is performed with respect to the 3rd target object, and the said 1st quality determination is unsatisfactory, the said extraction part is The time series third data string output by the sensor when the first work is performed on the third object among the plurality of first sections is a section where the third data string output from the sensor is outside the range column. The extraction device according to any one of appendices 1 to 4, wherein the extraction is performed as a second section.
(Additional remark 6) When the said 2nd quality determination in the said 2nd area is favorable when the said 1st operation | work is performed with respect to the 4th target object, and the said 1st quality determination is favorable, the said calculation part is The range column is calculated based on a fourth data sequence of time sequences output by the sensors in the plurality of first sections of the fourth object. 6. The supplementary note 1, wherein the range sequence is calculated. Extraction device.
(Additional remark 7) From the additional remark 1 characterized by comprising the determination part which determines the range for performing the said 2nd quality determination in the said 1st operation | work based on the said some 1st data sequence of the said 2nd area. The extraction device according to claim 6.
(Supplementary note 8) The extraction device according to any one of Supplementary notes 1 to 7, wherein the first work is a work using a manipulator, and the sensor is a sensor that detects distortion applied to the manipulator. .
(Supplementary note 9) Using a plurality of time-series first data strings output by a sensor when a first operation is performed on a plurality of first objects that are good in the first quality determination, The first operation is performed on the second object that is defective in the first pass / fail judgment among the plurality of first sections obtained by calculating the range row and dividing the periods of the plurality of first data rows in time series. A time series second data string output by the sensor at a time is outside the range string is extracted as a second section for performing a second pass / fail judgment in the first work. Extraction program (Appendix 10) An extraction method executed by a computer, wherein a plurality of time-series outputs output by a sensor when a first operation is performed on a plurality of first objects that are good in the first pass / fail determination Using one data string A step of calculating a time-series range sequence, and a plurality of first sections obtained by dividing a period of the plurality of first data sequences into a time-series, the second object being defective in the first pass / fail determination A section in which the time-series second data string output by the sensor when the first work is performed is out of the range string is extracted as a second section for performing the second pass / fail judgment in the first work. And an extraction method comprising: steps.

DESCRIPTION OF SYMBOLS 10 Computer 22 Robot 24 Sensor 30 1st data sequence 32 Average sequence 35 2nd data sequence 37 Range sequence 38 3rd data sequence 40 4th data sequence 50a-50g 1st area 56a-56b 2nd area

Claims (8)

A step of performing a first work divided into a plurality of first sections in time series on an object, and determining whether the first work is good or bad in order to make the object on which the first work is performed good or bad For the work including the step of performing a second pass / fail determination and the step of performing a first pass / fail determination to determine pass / fail of the object that has performed the first work after the second pass / fail determination. An extraction device for extracting a second section for performing the second pass / fail judgment among the sections,
A plurality of first data string sensor is data string outputted in time series when the first work to a plurality of first object was good in the first quality determination is performed, the time series A calculation unit for calculating a range column of
Among the plurality of first sections, the second data is a data string the sensor has output in time series when the first work to the second object was not good in the first quality determination is performed a section column becomes the range Retsugai, an extraction unit that extracts as said second section,
An extraction apparatus comprising:   The extraction device according to claim 1, wherein the extraction unit extracts the earliest first section outside the range column as the second section. The operation is the first quality determination and before after the second quality determination, according to claim 1 or 2 wherein the first work on the object is characterized in that it rows of the second work with another Extraction device. 2. The apparatus according to claim 1, further comprising a dividing unit that divides a period corresponding to the plurality of first data strings into the plurality of first sections based on an average string obtained by averaging the plurality of first data strings. 4. The extraction device according to any one of 3. The second pass / fail judgment in the second section is good when the first work is performed on a third object that has performed the work after the extraction unit has extracted the second section, When the pass / fail judgment is poor, the extraction unit outputs data output by the sensor in time series when the first work is performed on the third object in the plurality of first sections. a section third data column is a column becomes the range Retsugai extraction apparatus of any one of claims 1, wherein the extracting the second section to add 4. The second pass / fail judgment in the second section is good when the first work is performed on a fourth object that has performed the work after the extraction unit has extracted the second section, When the 1 pass / fail judgment is good, the calculation unit includes a fourth data string that is a data string output in time series by the sensor when the first work is performed on the fourth object. 6. The extraction device according to claim 1 , wherein a new range column is calculated so as not to be out of the range column. A step of performing a first work divided into a plurality of first sections in time series on an object, and determining whether the first work is good or bad in order to make the object on which the first work is performed good or bad For the work including the step of performing a second pass / fail determination and the step of performing a first pass / fail determination to determine pass / fail of the object that has performed the first work after the second pass / fail determination. An extraction program for extracting a second section for performing the second pass / fail judgment among the sections,
On the computer,
A plurality of first data string sensor is data string outputted in time series when the first work to a plurality of first object was good in the first quality determination is performed, the time series Let's calculate the range column of
Among the plurality of first sections, the second data string is a data string in which the sensor has output in time series when the first work to the second object that was bad in the first quality determination is performed There extraction program, characterized in that to extract the segment became the range Retsugai, as the second section. A step of performing a first work divided into a plurality of first sections in time series on an object, and determining whether the first work is good or bad in order to make the object on which the first work is performed good or bad For the work including the step of performing a second pass / fail determination and the step of performing a first pass / fail determination to determine pass / fail of the object that has performed the first work after the second pass / fail determination. An extraction method executed by a computer that extracts a second section for performing the second pass / fail judgment among sections ,
A plurality of first data string sensor is data string outputted in time series when the first work was to a plurality of first object was good in the first quality determination is performed, when Calculating a range column of the series;
Among the plurality of first sections, the second data string is a data string in which the sensor has output in time series when the first work to the second object that was bad in the first quality determination is performed a step but of extracting the section became the range Retsugai, as the second section,
The extraction method characterized by including.

Images