Statistical Significance Over Time Chart for Growing Process Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
It is challenging to assess how changes in business processes impact key metrics such as revenue, profit, or market share, as existing statistical tools often rely on the premise of a stable numerical goal, which is not applicable to all processes, particularly transactional processes like retail or service businesses that aim to grow continuously.
Innovation Solution
A graphical display is created using a statistical significance over time (SSOT) chart, which calculates and plots the 'area of common performance' for a control group, merging it with subject group data to visually highlight changes, using run charts or box plots, and performing hypothesis tests to verify statistical significance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional statistical tools are used to evaluate process changes, then measurement precision is improved, but adaptability deteriorates because these tools rely on the premise of stable numerical goals which do not apply to continuously growing transactional processes
Solution Approach 1:
The patent changes the fundamental parameters of statistical evaluation by replacing traditional hypothesis testing (which assumes stable means) with a growth-based framework. Instead of testing whether a mean differs from a target, the system evaluates whether observed growth trajectories significantly deviate from expected growth patterns, making statistical tools adaptable to continuously growing transactional processes while maintaining measurement precision
Solution Approach 2:
The patent introduces dynamic evaluation by comparing actual process growth trajectories against expected growth patterns over time. Rather than static hypothesis testing, the system uses time-series analysis to assess whether changes in process parameters produce statistically significant deviations from projected growth, enabling continuous monitoring and evaluation of process improvements in growing businesses
2Measurement precision
If hypothesis testing is performed to verify statistical significance, then measurement precision is improved, but loss of time increases due to the complexity of performing multiple tests over time periods
Solution Approach 1:
The patent merges multiple hypothesis tests into a unified growth trajectory analysis. Instead of performing separate statistical tests for each time period or metric, the system combines time-series data into integrated growth models that evaluate statistical significance across the entire evaluation period, reducing the number of tests required and minimizing time loss while maintaining measurement precision
3Measurement precision
If detailed statistical analysis is conducted to evaluate process changes, then measurement precision is improved, but device complexity increases due to the need for multiple statistical tools and methods
Solution Approach 1:
The patent creates a universal evaluation framework that can assess multiple process changes across different metrics and time periods using a single integrated growth-based statistical approach. This multi-functional system replaces the need for multiple specialized statistical tools, reducing device complexity while maintaining comprehensive evaluation accuracy through unified growth trajectory analysis
Data Source
AI summary
Method and apparatus for evaluating statistical significance over time are described. A graphical display helps to establish if an improvement produces effective results over a period of time relative to similar results from a control group. This graphical display can be achieved in part by calculating and plotting the control group's “area of common performance” representing confidence intervals around the mean of the metric being analyzed. Subject group data can be represented by either a run chart or a box plot, depending on whether the subject group data is discrete or continuous. In some embodiments a hypothesis test can be performed on the data to verify the representation.