Data processing: generic control systems or specific application – Specific application – apparatus or process – Product assembly or manufacturing
Reexamination Certificate
1998-09-04
2001-07-10
Grant, William (Department: 2121)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Product assembly or manufacturing
Reexamination Certificate
active
06259959
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention is directed to determining which tools or toolset have the greatest impact on manufacturing line performance and more particularly to a method for determining the performance components of a manufacturing line.
2. Description of the Related Art
Over the past two decades continuous flow manufacturing (CFM) has been the principle operational tool to help manage and improve the utilization of manufacturing assets. As the name connotes, the key focus of CFM is to measure and manage the throughput of tools/toolsets that comprise the manufacturing line. To this end, there have been a variety of systems proposed to help manage throughput with their attendant control methodologies.
Semiconductor manufacturing is faced with significant investment and cost challenges. The investment required to build and tool a semiconductor manufacturing line has steadily increased so that now in excess of $1 Billion is needed. In addition, the industry has been characterized by products (e.g., dynamic random access memories “DRAMS”) that have sustained long term price declines of some 27% per year over the past 20 years. A significant component of maintaining this price decline has been the ability of manufacturing/engineering to reduce costs by increasing productivity, not only in terms of good chips per wafer, which are driven by technology and yields, but by identifying and improving the utilization of the installed tool base.
One of the principle methods used to improve asset utilization in manufacturing has been continuous flow manufacturing (CFM) which includes a number of techniques that focus primarily on the flow of product through the line as the means to identify and fix problems in the line. One such example is the Theory of Constraints, which determines the toolset(s) which limit the throughput of the line and then provides a set of operational roles (Drum-Buffer-Rope) that seek to maximize the use of the constraining toolset(s). Another example, that was developed by the Japanese as part of their lean manufacturing, is the use of KANBANS, which are devices for limiting the movement of product between toolsets. The result is a limit on the amount of work in progress (WIP) in front of each toolset. Ideally, a KANBAN ensures that there is neither to much nor too little WIP; too much WIP results in line congestion while too little WIP allows a toolset to run out of work.
In the past few years there has been an increased awareness in semiconductor manufacturing that there is a relationship between the utilization of effective capacity and cycle time and the invention presents a new methodology to analyze this relationship and produce unexpected benefits when compared to conventional methods and structures.
SUMMARY OF THE INVENTION
It is, therefore, an object of the present invention to provide a structure and method for exploring how the X-factor (normalized cycle time), rather than throughput, is used as the prime production line control and production line parameter; hence, the invention is sometimes referred to herein as short cycle time manufacturing (SCM.).
More specifically, the invention includes a process, computer system and computer storage medium for optimizing a manufacturing line comprising, determining work center raw processing times of a plurality of work centers in the manufacturing line, summing the work center raw processing times to produce a manufacturing line raw processing time, determining work center cycle times of the work centers, dividing the work center cycle times by respective ones of the work center raw processing times to produce work center X-factors, weighting each of the work center X-factors by a percentage that a corresponding one of the work center raw processing times represents of the manufacturing line raw processing time to produce X-factor contributions for each of the work centers, and modifying at least one work center of the work centers having an X-factor contribution higher than others of the X-factor contributions to reduce the X-factor contribution of the at least one work center.
The step of determining work center raw processing times comprises measuring a time for a work center to perform work on a workpiece, and the step of determining work center cycle times comprises measuring a time from when the workpiece arrives at the work center until the workpiece leaves the work center. The step of modifying at least one work center further comprises determining whether the at least one work center is being utilized at approximately a maximum effective capacity. The step of modifying at least one work center comprises increasing effective capacity of the at least one work center. The step of modifying at least one work center comprises optimizing batch-size for at least one work center.
The work center comprises a group of tools and the determining work center raw processing times, summing the work center raw processing times, determining work center cycle times, dividing the work center cycle times by respective ones of the work center raw processing times, weighting each of the work center X-factors and modifying at least one work center are repeated for each tool within the work center to determine which tool of the tools has an X-factor contribution higher than others of X-factor contributions of the tools.
Because manufacturing lines have both throughput and X-factor commitments, the invention utilizes the fundamental relationships between throughput, capacity and X-factor. The X-factor is a much more sensitive indicator of capacity problems than throughput because X-factor increases rapidly as the throughput approaches the effective capacity, as shown in
FIG. 1
, discussed below. This sensitivity in X-factor is used by the invention as a powerful diagnostic tool to uncover unanticipated capacity issues.
Short cycle time manufacturing (SCM) allows each tool/toolset to be analyzed depending on its demonstrated X-factor and capacity versus a target to determine which tools/toolsets need improvement, since the overall X-factor of the line is just the weighted sums of the component toolset X-factors. In addition, the impact of mix and volume with a cycle time constrain the capacity of tools that are affected by batch or train size. Thus, SCM provides significant advantages over CFM in helping to manage and improve manufacturing asset utilization.
The present invention utilizes the X-factor as an important measurement and diagnostic parameter for semiconductor manufacturing lines and the X-factor forms the basis for short cycle-time manufacturing (SCM). The throughput constraint is not necessarily the performance constraint for the line. As a result, modifications to the production line that are driven primarily by continuous flow manufacturing can result in no improvement in line performance.
The invention relates the X-factor performance of each tool in the line to the overall line X-factor, which allows for the generation of customized performance targets per tool that still fulfill the overall line objective and prioritizes the tools so that the biggest contributors to line performance can be identified and improved.
In addition, the impact of the volume per recipe with an X-factor constraint is taken into consideration by the invention. This enables more accurate planning parameters and improved operational procedures. As a result, better decisions can be made with the invention concerning the application of scarce resources to increase capacity, whether by purchasing more equipment, applying more operators to run existing equipment, changing process characteristics, or by improving the lot selection capability.
REFERENCES:
patent: 5177688 (1993-01-01), Rentschler et al.
patent: 5229948 (1993-07-01), Wei et al.
patent: 5446671 (1995-08-01), Weaver et al.
patent: 5546329 (1996-08-01), Kurtzberg et al.
patent: 5748478 (1998-05-01), Pan et al.
patent: 5880960 (1999-03-01), Lin et al.
patent: 6104965 (2000-08-01), Lim et al.
Shea et al. “Development and Implementation the Ra
Gain Edward F.
Grant William
International Business Machines - Corporation
Kotulak, Esq. Richard M.
McGinn & Gibb PLLC
LandOfFree
Method for determining the performance components of a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for determining the performance components of a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for determining the performance components of a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2561915