Substrate Scheduling System Using Precomputed Allocation Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional scheduling methods for substrate treating apparatuses require extensive time for schedule preparation, leading to reduced operating rates and inflexible scheduling, as they often necessitate moving back treating steps to minimize standby periods, which increases standby time and decreases the number of lots that can be scheduled.
Innovation Solution
A scheduling method that allocates treating steps based on a systematic chart, prioritizing blocks with early expected completion times and allowing for restarts at upstream branching points when waiting times exceed allowed times, to overlap operations and reduce trial-and-error allocations, thereby shortening scheduling time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the conventional scheduling method allocates treating steps by moving back subsequent steps to minimize standby periods, then the operating rate is improved, but the scheduling time is excessively extended
Solution Approach 1:
The patent pre-calculates and stores optimal allocation patterns for treating steps in a database before actual scheduling occurs. When scheduling is needed, the system retrieves pre-computed patterns based on current conditions rather than performing time-consuming real-time optimization, thus resolving the contradiction between achieving high operating rates and maintaining short scheduling times.
Solution Approach 2:
The patent divides the scheduling problem into segments by pre-calculating allocation patterns for different scenarios and storing them separately. The system then selects and combines appropriate pre-computed segments based on current treatment requirements, avoiding the need to recalculate everything from scratch and significantly reducing scheduling time while maintaining optimal operating rates.
2Loss of time
If the conventional scheduling method reduces the number of lots for scheduling to avoid extended scheduling time, then the scheduling time is reduced, but the operating rate decreases due to increased standby time
Solution Approach 1:
The system performs preliminary calculations and stores optimal allocation patterns for multiple lots in advance. This allows the scheduling system to handle a larger number of lots efficiently by retrieving pre-computed patterns rather than calculating allocations in real-time, thus maintaining high operating rates without extending scheduling time.
Solution Approach 2:
The patent creates and stores copy patterns of optimal allocations for different treatment scenarios in a database. During actual scheduling, the system copies and applies appropriate pre-computed patterns to current lots, enabling rapid scheduling of multiple lots without reperforming complex optimization calculations, thereby maintaining both short scheduling time and high operating rate.
3Productivity
If the conventional scheduling method attempts to improve operating rate by reducing standby periods through extensive schedule preparation, then the operating rate improves, but the scheduling becomes inflexible
Solution Approach 1:
The patent implements a dynamic scheduling system that retrieves pre-computed allocation patterns based on current treatment conditions and lot characteristics. This allows the system to maintain optimal operating rates through pre-calculated patterns while remaining flexible to adapt to different treatment scenarios by selecting appropriate patterns from the database, thus resolving the contradiction between improving operating rate and maintaining scheduling flexibility.
Data Source
AI summary
A scheduling method for a substrate treating apparatus having a plurality of treating units for treating substrates, wherein, based on a procedure including a plurality of treating steps, a controller determines an order of treating a plurality of lots successively in the treating units. The method includes a step of preparing a plurality of schedules through execution of a basic allocating method, based on a systematic chart representing combinations of the treating steps for the respective lots, for allocating a first treating step of one of the lots, and thereafter, while following the systematic chart from the treating step, for searching and allocating as a next treating step a treating step selected from among succeeding treating steps in the lots, whose preceding treating step has an earliest expected completion time of all preceding treating steps; and a first allocating method for allocating treating steps based on the basic allocating method, and when a waiting time occurring with one of the treating steps exceeds an allowed time for enabling a standby at one of the treating units, for avoiding allocation of the one of the treating steps and restarting a search at one branching point before a branching point upstream of the one of said treating steps. The method further includes a step of selecting one schedule having been completed within a specified time, from the plurality of schedules prepared.