Adaptive control algorithm for improving AMHS push lot accuracy
AssigneeTaiwan Semiconductor Manufacturing Company, Ltd.
InventorsMing-Hsun Chung, Hung-I Chen, Wen-Cheng Chin
Life of this patent
Estimating the term…
Abstract
A self-adjusting prediction system that provides for the transmission and storage of push lots of work. It uses an adaptive control algorithm in it's methodology to improve Automated Material Handling Systems (AMHS) transmissions. The method of prediction is greatly enchanted to reduce overall cycle time, incorrect transmission of work lots, and idle manufacturing tools. It provides real-time updating that enables a complex manufacturing Fab to process work with optimum movement between tools.
Claims20
An AMHS lot transmission prediction method, comprising the steps of:
- a. providing a process that can predict the optimum transmission path to the next work position;
- b. providing an SQL algorithm that determines the key information records to retrieve for generation of optimized transmission and storage instructions to the AMHS;
- c. retrieving AMHS history records on a continuous basis so as to be able to quickly adapt the system according to recent past behavior;
- d. collecting time stamps, location, stage, tool capability, and next required tool capability of each lot processed in order to adapt the current Fab activity;
- e. running a stored programming procedure that analyzes current activity on the Fab and from that develops the lot distribution table of successful hit ratios for manufacturing transmissions.
A lot transmission prediction method, comprising the steps of:
- a. providing a flow control process that can predict the optimum transmission path for work lots to their next work position;
- b. providing a method that determines the key information records to retrieve for generation of optimized transmission and storage instructions for manufacturing;
- c. retrieving manufacturing history records on a continuous basis in order to quickly adapt the system according to recent past behavior;
- d. collecting the time stamps, location, stage, tool capability, and next required tool capability of each lot processed in order to adapt the current Fab activity;
- e. running a stored programming procedure that analyzes current activity on the Fab and from that develops the lot distribution table of successful hit ratios for manufacturing transmissions.
A prediction control system for controlling transmission of work lots in manufacturing comprising of:
- a. an AMHS type system that controls tools and transmission of work lots on the manufacturing floor;
- b. a data processing system that creates records containing the time stamps, location, stage, tool capability, and the next required tool capability of each lot processed;
- c. a set of analysis routines that continuously check past records to predict the optimum transmission path to the next tool so that processing can begin in a timely manner without a manual move by manufacturing personnel;
- d. a feedback procedure that notifies the system of successful and unsuccessful transmissions in real time so that subsequent transmissions have the advantage of this information in determining their optimum path.
A method of a software transmission prediction system that sorts a plurality of records that represent a plurality of activities monitored by a manufacturing control system so as to produce optimum transmission to the next capable tool by:
- a. storing a plurality of data records in a memory storage device with each record representing a plurality of entities from which a lot distribution table of successful hit ratios is built;
- b. identifying a set of data records containing the required information on which to build the lot distribution table of successful hit ratios, and
- c. providing the lot distribution table of successful hit ratios to the manufacturing control system so that the manufacturing control system can issue an optimized transmission command.
Dependencies are inferred from the claim text.