Assisted Analytics Using Dependency Graphs for Variance Source Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems require substantial human expertise to analyze business performance data, leading to inefficient and non-transferable manual investigations for variance sources, posing a burden on organizations.

Innovation Solution

A computer-implemented measure factory system that collects, analyzes, and reports performance measures automatically, identifying outliers and their sources through assisted analytics using processors and peripherals, enabling rapid and comprehensive data processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual investigation is used to track down sources of variance, then expertise-based analysis can be performed, but the process is inefficient and creates ongoing organizational burden

Engineering Contradiction:

Improveanalysis accuracyVSAvoidanalysis efficiency

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces manual mechanical analysis processes with automated computer-based systems. The measure factory automatically collects data from multiple sources, calculates performance measures, identifies variances, and determines root causes without human intervention, thereby maintaining analysis accuracy while dramatically improving efficiency

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service analytics where the computer system automatically performs data collection, analysis, variance identification, and root cause determination. The measure factory autonomously processes performance data and generates insights without requiring ongoing manual investigation or expert analysis

Inventive Principle:

Principle #25Self-service

2Measurement precision

If manual analysis methods are used, then detailed investigation of variance sources is possible, but the work is rarely transferrable to other use requirements

Engineering Contradiction:

Improvevariance analysis depthVSAvoidanalysis transferability

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The measure factory is designed as a universal system that can analyze multiple types of performance measures across different business areas. The automated data collection and analysis engine can be applied to various data sources and performance metrics, making the analysis methodology transferable to different use requirements and business contexts

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Measurement precision

If comprehensive performance measures are analyzed, then deep identification of procedure sources is achieved, but the complexity of processing increases

Engineering Contradiction:

Improvesource identification accuracyVSAvoidprocessing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments the complex analysis process into distinct automated components: data collection from multiple sources, performance measure calculation, variance identification, and root cause determination. Each component handles a specific aspect of the analysis, reducing overall processing complexity while maintaining comprehensive source identification capability

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS12367445B2Assisted analytics

Publication Date: 2025.07.22 DIMENSIONAL INSIGHT INC

AI summary

Detecting a source of a change in a business performance measure includes a first measure outlier event detection step, an outlier event contributor determining step, an outlier event source data entry identification step based on an execution dependency graph, and a step of alerting a user of a business activity represented by data entries in the outlier event source data.