Software Assurance Effort Estimation via Multidimensional Test Case Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing software estimation techniques for software assurance projects are unreliable due to their one-dimensional and person-dependent nature, failing to accurately categorize test cases and distribute effort across project phases, leading to improper effort estimation and inefficient resource allocation.
Innovation Solution
A multidimensional, person-independent system that classifies test cases based on two dimensions: Test Case Type (new, modified, existing) and Test Case Complexity (simple, average, complex), assigning weight ratios to estimate project size in Normalized Test Cases (NTC), and adjusts effort based on organizational baseline productivity and specific project factors for distribution across project phases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If one-dimensional categorization of test cases based on single parameters (number of steps, data, operations) is used, then estimation process is simple, but estimation accuracy is low
Solution Approach 1:
The patent transitions from one-dimensional test case categorization to multi-dimensional categorization by introducing multiple parameters (number of steps, data volume, operations performed, test case type) that form a dimensional matrix. This allows test cases to be classified across multiple axes simultaneously, significantly improving estimation accuracy while maintaining manageable process complexity through structured classification frameworks.
Solution Approach 2:
The patent segments the estimation process into distinct phases: test case classification based on multiple parameters, normalization of test cases into standard units, and separate estimation of different project phases (development, testing, maintenance). This segmentation allows each dimension to be handled systematically, improving overall estimation precision without overwhelming complexity.
2Ease of operation
If person-dependent estimation methods are used, then estimation can be easily performed, but reliability and consistency of estimates are poor
Solution Approach 1:
The patent transforms subjective person-dependent estimates into objective parameter-driven estimates by defining specific measurable parameters (number of steps, data volume, operations). Each test case is evaluated against these standardized parameters, and normalization factors convert them into consistent effort units. This removes personal bias while maintaining ease of operation through clear parameter definitions and calculation procedures.
Solution Approach 2:
The patent replaces the mechanical system of human judgment with a systematic calculation mechanism based on predefined parameters and normalization formulas. The estimation process becomes a deterministic calculation rather than a subjective assessment, improving reliability while keeping the process accessible through structured parameter collection and formula-based computation.
3Loss of time
If crude techniques like proportion of development effort are used, then effort estimation is quick, but effort distribution across project phases is improper
Solution Approach 1:
The patent divides the software assurance project into distinct phases (development, testing, maintenance) and estimates effort for each phase separately based on normalized test case counts and phase-specific factors. This segmentation enables accurate effort distribution across phases while maintaining quick estimation through standardized phase factor applications, replacing crude proportional methods with phase-aware systematic estimation.
Data Source
AI summary
A method and system is provided for estimating size and effort of software assurance project for distributing the estimated effort across the software assurance project phases. Particularly, the invention provides a method and system for estimating the software assurance project size based on the predefined weight ratios assigned to the test cases after classifying them into simple, medium and complex categories. Further, the invention provides a method and system for utilizing the estimated software assurance project size and organizational baseline productivity information for estimating the software assurance efforts. Further, the invention provides a method and system for distributing the estimated effort across the software assurance project phases.