Controlling method for manufacturing process

• US 6,862,484 B2
• Filed: 04/19/2002
• Issued: 03/01/2005
• Est. Priority Date: 04/20/2001
• Status: Expired due to Term

• Current Assignee: LAPIS Semiconductor Co., Ltd. (ROHM Co., Ltd.)
• Original Assignee: OKI Electric Industry Company Limited
• Inventors: Hayashi, Shunji
• Primary Examiner(s): Knight, Anthony
• Assistant Examiner(s): BARNES-BULLOCK, CRYSTAL JOY
• Application Number:
  US10/125,403
• Publication Number:
  US 20020156549A1
• Time in Patent Office: 1,047 Days
• Field of Search: 700/73, 700/79, 700/110, 700/121, 702/183, 702/185, 702 81- 84
• US Class Current: 700/73
• CPC Class Codes: G05B 19/41875 characterised by quality surveillance of production
• CPC Class Codes: G05B 2219/31477 Display correlated data so as to represent the degree of correlation
• CPC Class Codes: G05B 2219/32221 Correlation between defect and measured parameters to find origin of defect
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

In a method of controlling a manufacturing process, a Mahalanobis space of plural manufacturing control parameters is generated on the basis of first sampled data. Then, a Mahalanobis distance from the Mahalanobis space and second sampled data is calculated. A manufacturing process is determined to be under a malfunction operating condition by comparing the Mahalanobis distance and a threshold value.

Claims

1. A method of controlling a manufacturing process, comprising:

providing first sampled data which indicates a sampled data group by sampling a plurality of manufacturing control parameters under a normal operating condition of the manufacturing process, which includes a plurality of process steps;

generating a Mahalanobis space of the plurality of the manufacturing control parameters, based first sampled data;

providing second sampled data which indicates a sampled data group by sampling the plurality of manufacturing control parameters at constant time intervals, during the manufacturing process;

calculating a Mahalanobis distance from the Mahalanobis space and the second sampled data;

making a decision that the manufacturing process is under a malfunction operating condition when the Mahalanobis distance is more than a threshold value;

calculating displacement quantities for each of the manufacturing control parameters from an average of the Mahalanobis space; and

determining a degree of incidence of the malfunction operating condition for each of the plurality of process steps, in accordance with the displacement quantities for each of the manufacturing control parameters.

2. The method according to claim 1, wherein the Mahalanobis space is generated in accordance with an inverse matrix of a correlation matrix of the first sampled data.

3. The method according to claim 1, further comprises:

storing information provided by a control method for the manufacturing process, into a host computer through a local area network, wherein the information is accessible from a plurality of servers through an intranet.

4. A method of controlling a manufacturing process, comprising:

setting ideal data of plural manufacturing control parameters of the manufacturing process which includes a plurality of process steps, and a permissible range of the ideal data;

generating random numbers within the permissible range of the ideal data;

generating Mahalanobis spaces for each of the plural manufacturing control parameters, based on the random numbers;

providing sampled data which indicates a sampled data group by sampling the plural manufacturing control parameters at constant time intervals, during the manufacturing process;

calculating Mahalanobis distances from the Mahalanobis spaces and the sampled data;

determining a degree of divergence from an ideal operating condition of the manufacturing process, by comparing the Mahalanobis distances and a threshold value;

calculating displacement quantities for each of the manufacturing control parameters from an average of the Mahalanobis spaces; and

determining a degree of incidence of a malfunction operating condition for each of the plurality of process steps, in accordance with the displacement quantities for each of the manufacturing control parameters.

5. A method of controlling a manufacturing process, comprising:

providing first sampled data which indicates a sampled data group by sampling a plurality of manufacturing control parameters under a normal operating condition of the manufacturing process, which includes a plurality of process steps;

generating a Mahalanobis space of the plurality of the manufacturing control parameters, on the basis of the first sampled data;

providing second sampled data which indicates a sampled data group by sampling the plurality of manufacturing control parameters during the manufacturing process, at constant time intervals;

calculating a first Mahalanobis distance from the Mahalanobis space and the second sampled data;

generating a selected group of combined parameters the plurality of the manufacturing control parameters;

calculating a second Mahalanobis distance from the Mahalanobis space and the selected group of combined parameters; and

determining a degree of incidence of a malfunction operating condition for each of the plurality of process steps, in accordance with a displacement quantity of the second Mahalanobis distance from the first Mahalanobis distance.

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

storing information provided by a control method for the manufacturing process, into a host computer through a local area network, wherein the information is accessible from a plurality of servers through an intranet.

7. A method of controlling a manufacturing process, comprising:

setting ideal data of plural manufacturing control parameters of the manufacturing process which includes a plurality of process steps, and a permissible range of the ideal data;

generating random numbers within the permissible range of the ideal data;

generating Mahalanobis spaces for each of the plural manufacturing control parameters, based on the random numbers;

providing sampled data which indicates a sampled data group by sampling the plural manufacturing control parameters at constant time intervals, during the manufacturing process;

calculating Mahalanobis distances from the Mahalanobis spaces and the sampled data;

determining a degree of divergence from an ideal operating condition of the manufacturing process, by comparing the Mahalanobis distances and a threshold value; and

storing information provided by a control method for the manufacturing process, into a host computer through a local area network, wherein the information is accessible from a plurality of servers through an intranet.

8. A method of controlling a manufacturing process, comprising:

setting ideal data of plural manufacturing control parameters of the manufacturing process which includes a plurality of process steps, and a permissible range of the ideal data;

generating random numbers within the permissible range of the ideal data;

generating Mahalanobis spaces for each of the plural manufacturing control parameters, based on the random numbers;

providing sampled data which indicates a sampled data group by sampling the plural manufacturing control parameters at constant time intervals, during the manufacturing process;

calculating Mahalanobis distances from the Mahalanobis spaces and the sampled data; and

determining a degree of divergence from an ideal operating condition of the manufacturing process, by comparing the Mahalanobis distances and a threshold value,

wherein the Mahalanobis spaces are generated in accordance with an inverse matrix of a correlation matrix of the sampled data.

9. A method of controlling a manufacturing process, comprising:

setting ideal data of plural manufacturing control parameters of the manufacturing process which includes a plurality of process steps, and a permissible range of the ideal data;

generating random numbers within the permissible range of the ideal data;

generating Mahalanobis spaces for each of the plural manufacturing control parameters, based on the random numbers;

providing sampled data which indicates a sampled data group by sampling the plural manufacturing control parameters at constant time intervals, during the manufacturing process;

calculating first Mahalanobis distances from the Mahalanobis spaces and the sampled data;

generating a selected group of combined parameters from the plural manufacturing control parameters;

calculating second Mahalanobis distances from the Mahalanobis spaces and the selected group of combined parameters; and

determining a degree of incidence of an ideal operating condition for each of the plurality of process steps, in accordance with a degree of divergence between the first and second Mahalanobis distances.

10. The method according to claim 9, further comprising:

storing information provided by a control method for the manufacturing process, into a host computer through a local area network, wherein the information is accessible from a plurality of servers through an intranet.