CMP Pilot Dispatch Scheduling to Reduce Head Idle Seasoning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The Chemical Mechanical Polishing (CMP) process faces inefficiencies due to labor-force dispatched Pilot approaches, leading to unnecessary machine allocation and increased production time, particularly exacerbated by Head Idle seasoning caused by discontinuous production and insufficient wafer load ports.

Innovation Solution

A CMP intelligent dispatch system utilizing a genetic algorithm to generate and optimize work schedules, automatically dispatching tasks to minimize idle time and optimize machine load, by generating initialization schedules, filtering based on adaptability, performing crossing and mutation operations, and selecting a target schedule that reduces Head idle seasoning and overall processing time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If labor-force dispatched Pilot approaches are used for machine scheduling, then ease of operation is improved, but productivity deteriorates due to unnecessary machine allocation and increased production time

Engineering Contradiction:
Improveease of operationVSAvoidproductivity
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent replaces manual labor-force dispatch with an automated computer-based dispatch management system that uses genetic algorithms to optimize machine scheduling. The system automatically generates initial schedules, performs filtering based on adaptability parameters, executes crossing and mutation operations, and produces optimized target schedules, eliminating the need for manual scheduling while significantly improving productivity through intelligent optimization

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes the scheduling approach by introducing adaptability parameters as filtering criteria and using genetic algorithm parameters (population size, mutation rate, crossover rate) to optimize machine allocation. By dynamically adjusting these parameters, the system achieves optimal scheduling that reduces production time while maintaining ease of operation through automation

Inventive Principle:
Principle #35Parameter changes

2Ease of operation

If inappropriate Pilot strategy is adopted, then ease of operation is maintained, but loss of time increases due to excessive production time consumption

Engineering Contradiction:
Improveease of operationVSAvoidloss of time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The dispatch management system incorporates feedback mechanisms by evaluating schedule adaptability parameters and using this information to guide the genetic algorithm's optimization process. The system continuously refines schedules based on performance feedback, ensuring optimal time utilization while maintaining operational simplicity through automated decision-making

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs preliminary actions by generating multiple initial schedules and pre-evaluating them against adaptability parameters before final optimization. This preliminary filtering and preparation work ensures that the subsequent genetic algorithm operations start from optimized candidates, reducing overall computation time and production time while maintaining ease of operation

Inventive Principle:
Principle #10Preliminary action

3Device complexity

If Head Idle seasoning occurs due to discontinuous production, then device complexity is reduced, but productivity deteriorates due to increased machine production time

Engineering Contradiction:
Improvedevice complexityVSAvoidproductivity
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent addresses Head Idle seasoning by optimizing the dispatch schedule to ensure continuous useful action of the CMP machine. The genetic algorithm generates schedules that maintain continuous wafer processing by properly coordinating load port operations and machine cycles, eliminating idle time caused by discontinuous production while keeping the device complexity manageable through software-based scheduling

Inventive Principle:
Principle #20Continuity of useful action

4Device complexity

If insufficient wafer load ports are allocated, then device complexity is reduced, but loss of time increases due to inability to continuously draw wafer pieces

Engineering Contradiction:
Improvedevice complexityVSAvoidloss of time
Core Design Contradiction:
Device complexityVSLoss of time

Solution Approach 1:

The system dynamically determines the optimal number of load ports to allocate based on production requirements and machine capabilities. The genetic algorithm adjusts load port allocation in the schedule to ensure continuous wafer processing, preventing idle time caused by insufficient load ports while avoiding unnecessary complexity by using software-based dynamic allocation rather than fixed hardware configurations

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS11397425B2CMP polishing system and associated pilot management system
Publication Date: 2022.07.26 UNITED SEMICONDUCTOR (XIAMEN) CO LTD
  • US11397425B2 patent drawing
  • US11397425B2 patent drawing
  • US11397425B2 patent drawing

AI summary

A dispatch management method for Pilot-run on a computer and applicable to chemical mechanical polishing machines includes: generating initialization work schedules; filtering the initialization work schedules according to respective adaptability parameters to generate intermediate work schedules; performing crossing operations on the intermediate work schedules to generate M sets of crossed work schedules; performing mutation calculations on contents of the intermediate work schedules and the M sets of crossed work schedules to generate mutated work schedules; performing optimization calculations on the intermediate work schedules, the crossed work schedules and the mutated work schedules to generate a target work schedule; and automatically performing dispatch on the CMP machines according to the target work schedule.