System Design Learning Models for Functional and Performance Requirements
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ICT system design methods struggle with efficiently deriving a concrete system configuration due to the application of multiple concretization rules, leading to numerous configurations that may not meet requirements, and lack comprehensive learning of both functional and nonfunctional requirements.
Innovation Solution
A system design learning apparatus and method that comprehensively learns functional and nonfunctional requirements through a functional requirement learning unit, performance measurement unit, and nonfunctional requirement learning unit, using learning models to evaluate and select optimal configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple concretization rules are applied to convert abstract portions into concrete portions, then the system can generate diverse system configurations, but the number of generated configurations becomes huge and many do not meet requirements
Solution Approach 1:
The patent applies preliminary action by using a learning model to evaluate and score concrete system configurations before final selection. The learning model is trained in advance on requirements and configurations, enabling rapid assessment of which configurations meet requirements without exhaustively checking all possible configurations generated by multiple concretization rules.
Solution Approach 2:
The patent implements feedback through the learning model that continuously refines its evaluation based on the relationship between system configurations and requirements. The learning model receives feedback about which configurations satisfy requirements and uses this to improve future evaluations, enabling efficient filtering of valid configurations from the large set generated by multiple concretization rules.
2Adaptability or versatility
If multiple concretization rules are applied in different orders, then more system configurations can be generated, but the derivation process becomes time-consuming and inefficient
Solution Approach 1:
The learning model performs preliminary evaluation of configurations before they are fully generated or selected. By pre-training the model on the relationship between requirements and valid configurations, the system can rapidly assess the potential validity of configurations without time-consuming exhaustive generation and checking of all possible configurations.
Solution Approach 2:
The patent replaces the mechanical process of exhaustively generating and checking all possible configurations through a learning-based system. Instead of mechanically applying concretization rules in all possible orders and verifying each configuration, the learning model substitutes this process with intelligent prediction and evaluation, significantly reducing the time required to derive valid concrete system configurations.
3Reliability
If a large number of concretization rules are applied, then comprehensive coverage of system requirements is achieved, but the complexity of the derivation process increases
Solution Approach 1:
The learning model serves as an intermediary between the concretization rules and the final configuration selection. Instead of directly managing the complexity of multiple concretization rules and their interactions, the learning model mediates by evaluating the outcomes of applying these rules and filtering configurations based on learned patterns of requirement satisfaction, thereby reducing the operational complexity while maintaining comprehensive coverage.
Solution Approach 2:
The learning model provides feedback about which configurations satisfy requirements, enabling the system to focus computational resources on promising configurations rather than exhaustively processing all possible configurations. This feedback mechanism maintains reliability by ensuring requirement compliance while reducing the effective complexity of the derivation process through intelligent filtering.
Data Source
AI summary
A system design learning apparatus includes: a functional requirement learning unit that, by executing a concretization processing, learns with regard to a first learning model that outputs an evaluation value representing a potentiality of the system requirement or proposed system configuration meeting a functional requirement of the system and, furthermore, being converted into a concrete system configuration, which is information representing the configuration of the system not including the abstract portions; a performance measurement unit that, executes a performance measurement processing on the performance of the concrete system configuration to measure performance data; and a non-functional requirement learning unit that, learns with regard to a second learning model that outputs an evaluation value representing a potentiality of the concrete system configuration being able to exhibit performance meeting a nonfunctional requirement, by using the performance data.