Business Specification Generation via Multi-Model Association
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for generating business specifications are ineffective in extracting high-level business specifications from information systems, as they lack the ability to mechanistically assign meaning to programs, requiring semantic analysis by workers and failing to manage business processes as structured clusters.
Innovation Solution
A business specification generation assisting system is developed, utilizing a configuration with physical, business, and association model databases to store and manage relations between programs, physical data, and business functions, enabling the generation of business specifications by associating physical data with business functions and programs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If specification extraction processing is conducted using analysis of assets, then low level specification information close to computer system is extracted, but high level specification close to business cannot be extracted
Solution Approach 1:
The patent segments the specification extraction process into multiple hierarchical levels: low-level computer system specifications and high-level business specifications. By dividing the extraction target into distinct levels, the system can apply different analysis methods appropriate for each level, thereby achieving both precise low-level extraction and comprehensive high-level coverage.
Solution Approach 2:
The patent introduces an intermediary component that bridges computer system specifications and business specifications. This intermediary translates and connects the low-level technical specifications with high-level business requirements, enabling the system to extract both types of specifications effectively by serving as a mediator between different specification layers.
2Productivity
If mechanical analysis is conducted on programs, then analysis efficiency is improved, but semantic analysis capability deteriorates
Solution Approach 1:
The patent segments the analysis process into mechanical analysis for structural extraction and semantic analysis for meaning interpretation. By dividing the analysis into distinct phases with different methods, the system maintains high productivity through automated mechanical analysis while preserving semantic information through subsequent semantic processing.
Solution Approach 2:
The patent introduces an intermediary semantic analysis layer that processes the output of mechanical analysis. This intermediary component recovers and interprets semantic information that would otherwise be lost in purely mechanical analysis, thereby maintaining both efficiency and semantic fidelity.
3Adaptability or versatility
If business processes are managed as unstructured instructions in source programs, then implementation flexibility is maintained, but structure and comprehensibility deteriorate
Solution Approach 1:
The patent segments business processes from the monolithic source program structure into distinct, identifiable process clusters. By separating business logic into structured clusters while maintaining links to the original implementation, the system achieves both structure for comprehensibility and flexibility for implementation variations.
Solution Approach 2:
The patent creates a structured model copy of the business processes that mirrors the unstructured source program instructions. This copy provides a structured, comprehensible representation of business logic while the original unstructured instructions remain unchanged, preserving implementation flexibility while improving structure.
Data Source
AI summary
A business specification generation assisting system for generation of a business specification that implements a business function specified by a user, on the basis of association among the business functions, programs and physical data is provided. A physical model which is a graph having programs and input and output physical data as vertexes, a business model which is a graph having business functions and input and output logical data as vertexes, and an association model including an association table of business functions with program functions and an association table of logical data with physical data are stored. With respect to a business function specified by the user, an associated physical model is analyzed, and subgraphs associated with the business function are calculated and displayed. The business model and the association model are modified according to a modification order given by the user.