Method and system for managing semiconductor manufacturing equipment

• US 6,438,440 B1
• Filed: 03/26/1999
• Issued: 08/20/2002
• Est. Priority Date: 09/30/1998
• 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): PATEL, RAMESH B
• Application Number:
  US09/276,804
• Time in Patent Office: 1,243 Days
• Field of Search: 700/73-74, 700/96, 700/108, 700/78-80, 700/111, 700/117, 700/121, 700/21-22, 700/28, 700/32, 702/28-32, 702/82-85, 702/84, 702/184-185, 318/561, 318/560, 382/149, 382/155-159, 382/278, 382/190-197, 382/118, 382/309, 382/115, 382/227
• US Class Current: 700/121
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: H01J 37/32935 Monitoring and controlling tubes by information coming from the object and/or discharge

Abstract

A management method capable of making an accurate decision about a malfunction of the semiconductor manufacturing equipment includes sampling a plurality of data of at least one parameter under normal operating conditions of the semiconductor manufacturing equipment; generating a Mahalanobis space A from a group of sampled data; calculating a Mahalanobis distance D² from measured values of the parameter under ordinary operating conditions of the semiconductor manufacturing equipment; and deciding that a malfunction occurred in the semiconductor manufacturing equipment when the value of the Mahalanobis distance exceeds a predetermined value.

Claims

1. A method of managing operation of semiconductor manufacturing equipment comprising:

sampling a plurality of data of at least one parameter under a normal operating condition of said semiconductor manufacturing equipment;

generating a Mahalanobis space based on a group of sampled data;

calculating, based on said Mahalanobis space, a Mahalanobis distance from a group of measured values of said at least one parameter under an actual operating condition of said semiconductor manufacturing equipment; and

when a calculated Mahalanobis distance exceeds a predetermined value, making a decision that a malfunction occurred in said semiconductor manufacturing equipment.

2. A management method according to claim 1, further comprising:

sampling a plurality of data under abnormal conditions, each abnormal condition being generated by setting one parameter at an abnormal value and others of said at least one parameter at normal values;

generating Mahalanobis spaces under said abnormal conditions;

when, from a value of said Mahalanobis distance, a decision has been made that a malfunction occurred, calculating Mahalanobis distances corresponding to Mahalanobis spaces from said measured values, based on said Mahalanobis space under the abnormal condition; and

estimating that among said plurality of parameters, abnormality occurred in a parameter that gave a Mahalanobis space that has a Mahalanobis distance closest to 1.

3. A management method according to claim 1, wherein said sampling comprises sampling a plurality of mutually different parameters that include said at least one parameter, said management method further comprising:

generating Mahalanobis spaces at predetermined times from a data group of said plurality of mutually different parameters measured at predetermined times; and

calculating Mahalanobis distances at predetermined times from a group of measured values of said plurality of mutually different parameters, obtained from the actual operating condition of said semiconductor manufacturing equipment, by using said Mahalanobis spaces formed at predetermined times.

4. A management system of semiconductor manufacturing equipment comprising:

a memory unit that stores data on a Mahalanobis space obtained from a parameter showing a normal operating condition of semiconductor manufacturing equipment;

a detection mechanism that obtains data values of said parameter from said semiconductor manufacturing equipment during operation;

an arithmetic circuit that calculates a Mahalanobis distance from a data group of said parameter, obtained by said detection mechanism by using said Mahalanobis space stored in said memory unit; and

a circuit that decides whether or not a calculated value of said Mahalanobis distance by said arithmetic circuit exceeds a predetermined value, to determine occurrence of a malfunction of said semiconductor manufacturing equipment.

5. A management system according to claim 4, wherein said semiconductor manufacturing equipment is a plasma etching system utilizing plasma emission, and wherein said detection mechanism is a plasma emission intensity detector for measuring the intensity of a desired wavelength of the plasma emission of said etching system.

6. A management system according to claim 4, wherein said semiconductor manufacturing equipment is a plasma etching system equipped with a high frequency oscillator for plasma emission, and wherein said detection mechanism includes a current detector and a voltage detector for obtaining a current value, a voltage value, and a phase of high frequency output of said high frequency oscillator.

7. A management system according to claim 6, wherein said detection mechanism further includes a data converter for calculating current values, voltage values and phases of a fundamental wave and harmonics of said high frequency output, from a current value and a voltage value detected by said current detector and said voltage detector.

8. A management system according to claim 4, wherein said semiconductor manufacturing equipment is a plasma etching system utilizing plasma emission, and wherein said detection mechanism further includes an emission spectrometer for measuring the intensity of a plurality of desired wavelengths of plasma emission of said etching system.