Browse Prior Art Database

Exact Yield Prediction

IP.com Disclosure Number: IPCOM000084363D
Original Publication Date: 1975-Oct-01
Included in the Prior Art Database: 2005-Mar-02
Document File: 4 page(s) / 142K

Publishing Venue

IBM

Related People

DeVries, R: AUTHOR [+2]

Abstract

Data collection and analysis forms an integral part of any manufacturing process. In addition, a major consideration of a manufacturing process is an overall improvement of the quality and yield of a product, usually influenced by a set number of variables or parameters.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 55% of the total text.

Page 1 of 4

Exact Yield Prediction

Data collection and analysis forms an integral part of any manufacturing process. In addition, a major consideration of a manufacturing process is an overall improvement of the quality and yield of a product, usually influenced by a set number of variables or parameters.

The "YGAL" (Yield Gain and Loss) program is designed to identify the amount of loss or gain (in percent) contributed by an individual, or a combination of variables/parameters to the overall yield gain or loss of the product. YGAL is a terminal oriented conversational program that predicts yield when the imposed parameter limits are perturbed. YGAL employs that algorithm of extreme readings, thereby minimizing the amount of data collected and analyzed.

The program is described in a macro flow chart (Fig. 1) to represent the sequence of events that follow to obtain the desired results. The program assumes data to be resident online for analysis. It further assumes that the user has minimal knowledge of data processing and guides the user in a conversational interactive mode.

The principle of "extremes" is employed in a wide spectrum of applications. Two examples illustrate the use of YGAL. Example 1.

Company XYZ is in the business of producing memories.

Before the memories are shipped out to the customer, each individual memory is one hundred percent tested to guarantee its functionality.

The memory in this example is a 256 X 40 bits ROS (Read-Only Storage Memory).

The memory has: 8 binary inputs, 40 outputs, 2 gates, 1 Vcc and 1 V(EE).

The tests, to which each memory is subjected, are as follows:

(Image Omitted)

Assume that a lot consisting of 324 memories was tested according to the above test limits, and that 258 memories passed the test which translates to a
79.8% yield for this lot.

The question then becomes; why did the memories fail, and to what extent?

Once all the fails are pinpointed and categorized, by YGAL the yield losses can be minimized by either opening the limits by a minimum amount to gain a maximum yield increase and/or change the process to move the various distribu...