Wafer Testing Recipe Automation for AOI Consistency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Automatic Optical Inspection (AOI) systems face challenges in efficiently preparing and porting recipes for wafer testing, leading to increased preparation and review times, reduced machine utilization, and variability in inspection criteria across multiple sites.
Innovation Solution
A method and system for establishing a wafer testing recipe by acquiring die images, composing a reference image, defining zones of interest, determining detection and reporting policies, and integrating these components with pre-defined logical rules, allowing for manual, automatic, or combined definition techniques, and enabling fine-tuning and portability across inspection machines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual recipe preparation is performed for each wafer inspection, then inspection criteria can be customized, but recipe preparation time increases and machine utilization decreases
Solution Approach 1:
The system performs preliminary actions by automatically generating recipe components (detection policies, reporting policies, inspection policies) before actual wafer inspection begins. The reference image is created in advance, and zones of interest are pre-defined, allowing rapid recipe deployment without manual preparation during production
Solution Approach 2:
The system creates a reference image by copying and processing actual wafer images, then uses this reference to generate detection policies. This copying approach allows the system to learn inspection criteria from real data and replicate successful inspection patterns across multiple wafers, eliminating the need for manual recipe creation for each wafer
2Ease of manufacture
If multiple inspection sites use different manual recipes, then local optimization is possible, but consistency across sites deteriorates
Solution Approach 1:
The system creates a universal reference image that can serve multiple inspection sites simultaneously. The detection policies and reporting policies generated from this reference are designed to be site-agnostic, allowing the same recipe to be deployed across different locations while maintaining consistent inspection criteria, yet still adaptable to local requirements through parameter adjustment
3Measurement precision
If comprehensive inspection policies are defined manually, then inspection accuracy improves, but system complexity and setup time increase
Solution Approach 1:
The system performs self-service by automatically analyzing the reference image to identify zones of interest and generate appropriate detection policies without human intervention. The system autonomously determines which algorithms to apply, sets algorithm parameters, and creates reporting policies based on the reference data, eliminating the need for manual configuration of complex inspection parameters
Data Source
AI summary
A method and a system for establishing a wafer testing recipe are disclosed. The method may include acquiring images of a number of dice from a produced wafer; using at least part of the images to compose a reference-image; defining on the reference-image multiple “zones of interest; determining the Detection-Policy for each kind of zone of interest and determining the algorithm that will be used by each of the Detection-Policy; determining the parameters of each of the Detection-Policy's algorithms; determining the Reporting-policy; determining the Inspection-policy; and creating a “wafer testing recipe” by integrating of the testing reference of a typical die image, the zones of interest, the Detection-Policies, the parameters of the Detection-Policies' algorithms, the Reporting-Policies and the Inspection-Policies.