Semiconductor Property Prediction System Using Supervised Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The manufacturing process of semiconductor elements is complex and time-consuming, requiring extensive labor and resources to optimize and examine causal relationships between processing steps and properties, making it difficult to predict and improve semiconductor element properties effectively.
Innovation Solution
A property prediction system using supervised learning with a learning data set to infer semiconductor element properties, incorporating a memory unit, input unit, processing unit, and arithmetic unit for data conversion and machine learning, enabling the prediction of properties such as atomic number, electron configuration, and reliability test results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If comprehensive examination of causal relationships between manufacturing process and semiconductor element properties is performed, then manufacturing optimization is achieved, but time consumption and labor requirements increase significantly
Solution Approach 1:
The patent applies preliminary action by performing comprehensive examinations of causal relationships between manufacturing processes and semiconductor element properties in advance, before actual mass production. This allows the accumulation of extensive data and knowledge about process-property relationships, which can then be reused for optimization without repeating the same time-consuming examinations repeatedly.
Solution Approach 2:
The patent creates a virtual model or digital twin of the manufacturing process and semiconductor element properties. By copying the complex physical examination process into a computational model, the system can simulate and analyze causal relationships without requiring repeated physical experiments, thus reducing time consumption while maintaining optimization quality.
2Manufacturing precision
If comprehensive examination of causal relationships between manufacturing process and semiconductor element properties is performed, then manufacturing optimization is achieved, but labor requirements increase significantly
Solution Approach 1:
The patent implements self-service by creating an automated system that performs the examination of causal relationships between manufacturing processes and semiconductor properties. The system uses computational algorithms and data analysis tools to automatically identify optimal manufacturing parameters without requiring extensive manual labor from engineers, thereby reducing labor requirements while achieving manufacturing optimization.
Solution Approach 2:
The patent replaces manual mechanical examination methods with computational and information processing systems. Instead of relying on human experts to manually analyze complex manufacturing data and identify causal relationships, the system uses automated algorithms, data mining, and computational modeling to perform these tasks, significantly reducing labor requirements.
3Loss of information
If human experts manually analyze huge amount of manufacturing data, then causal relationships can be identified, but the process becomes extremely difficult and labor-intensive
Solution Approach 1:
The patent introduces an intermediary computational system that acts as a bridge between raw manufacturing data and causal relationship identification. This intermediary system includes data processing modules, analysis algorithms, and visualization tools that transform huge amounts of complex manufacturing data into actionable insights about causal relationships, making the information accessible and usable without requiring manual analysis by human experts.
Data Source
AI summary
A property prediction system for a semiconductor element is provided. The property prediction system includes a memory unit, an input unit, a processing unit, and an arithmetic unit. The processing unit has a function of creating a learning data set from first data stored in the memory unit, a function of creating prediction data from second data supplied from the input unit, a function of converting qualitative data (a material name or a compositional formula) into quantitative data (the properties of an element and a composition), and a function of performing extraction or removal on the first data and the second data. The first data includes step lists of first to m-th semiconductor elements (m is an integer of 2 or more) and the properties of the first to m-th semiconductor elements. The second data includes a step list of an (m+1)-th semiconductor element. The arithmetic unit having a function of performing learning and inference of supervised learning performs learning on the basis of the learning data set and makes an inference of a semiconductor element from the prediction data.