Manufacturing process monitoring system and manufacturing process monitoring method

• US 9,268,325 B2
• Filed: 08/11/2011
• Issued: 02/23/2016
• Est. Priority Date: 09/17/2010
• Status: Active Grant

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Ino, Tomomi
• Primary Examiner(s): Ali, Mohammad
• Assistant Examiner(s): Saavedra, Emilio J
• Application Number:
  US13/207,943
• Publication Number:
  US 20120072002A1
• Time in Patent Office: 1,657 Days
• Field of Search: 700/110, 700/51, 700/143
• US Class Current: 1/1
• CPC Class Codes: G05B 19/4183 characterised by data acquisition, e.g. workpiece identification
• CPC Class Codes: G05B 19/4184 characterised by fault tolerance, reliability of production system
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

According to one embodiment, a manufacturing process monitoring system for monitoring anomaly in a manufacturing process for products, the system includes an information storage section, a selection information section, a reference space formation section and a monitoring section. The information storage section is configured to store previously collected data. The selection information section is configured to create information used in classifying the data stored in the information storage section. The reference space formation section is configured to form a reference space based on data subjected to anomaly monitoring acquired in the manufacturing process and the data classified by the information from the data stored in the information storage section. The monitoring section is configured to monitor anomaly of the data subjected to anomaly monitoring based on the reference space.

Claims

1. A manufacturing process monitoring system for monitoring anomalies in a manufacturing process for products, the system comprising:

an information storage section configured to store data collected from plural lots of a manufactured product as previously collected data;

a selection information section configured to create a value of a weighting factor, wherein the value of the weighting factor pertains to a product yield for each lot of the manufactured product and establishes a priority associated with the weighting factors when classifying the previously collected data, wherein a higher yield is classified with a higher priority;

a detection apparatus configured to acquire data in the manufacturing process that is subjected to anomaly monitoring;

a reference space formation section configured to classify the previously collected data provided from the information storage section to produce classified data by using a product of a distance and the value of the weighting factor, the reference space formation section configured to form a reference space based on the classified data, the distance being between the data subjected to anomaly monitoring and each of the previously collected data provided from the information storage section; and

a monitoring section configured to monitor anomalies of the data subjected to anomaly monitoring based on the reference space.

2. The system according to claim 1, further comprising:

an inspection apparatus configured to inspect a product, the selection information section creating information on the product determined to be defective based on a result of the inspection, the reference space formation section excluding from the classified data, data of the previously collected data which corresponds to the information on the product determined to be defective, the reference space formation section forming the reference space based on the remaining classified data.

3. The system according to claim 1, wherein

the reference space formation section forms the reference space so that the reference space includes the classified data.

4. The system according to claim 1, wherein

the value of the weighting factor becomes smaller as the priority becomes higher, and

the reference space formation section uses a predetermined number of data from the classified data having the smallest products of the distance and the value of the weighting factor to form the reference space.

5. A manufacturing process monitoring method for monitoring anomalies in a manufacturing process for products, the method comprising:

acquiring data that is subjected to anomaly monitoring in a manufacturing process being executed;

acquiring data collected from plural lots of a manufactured product as previously collected data;

creating a value of a weighting factor, wherein the value of the weighting factor pertains to a product yield for each lot of the manufactured product and establishes a priority associated with the weighting factors when classifying the previously collected data, wherein a higher yield is classified with a higher priority;

classifying the previously collected data to produce classified data by using a product of a distance and the value of the weighting factor, and forming a reference space based on the classified data, the distance being between the data subjected to anomaly monitoring and each of the previously collected data;

and

monitoring anomalies of the data subjected to anomaly monitoring based on the reference space.

6. The method according to claim 5, further comprising:

displaying a result of a recognized anomaly and information on extracted data in combination on a display section.

7. The method according to claim 5, further comprising:

inspecting a product, creating information on the product determined to be defective based on a result of the inspection, excluding from the classified data, data of the previously collected data which corresponds to the information on the product determined to be defective, in the forming the reference space, the reference space formed based on the remaining classified data.

8. The method according to claim 5, wherein

in the forming the reference space, the reference space is formed so that the reference space includes the classified data.

9. The method according to claim 5, wherein

the value of the weighting factor becomes smaller as the priority becomes higher,

in the forming the reference space, a predetermined number of data from the classified data having the smallest products of the distance and the value of the weighting factor is used to form the reference space.