DOE Power Prediction Interface for Experiment Design

Solution Overview

Problem

Conventional Design of Experiments (DOE) processes lack the ability to predict the power of proposed experiments before they are run, leading to inefficiencies and increased costs due to insufficient experimental design, which can only be addressed post-experiment by adding trial runs, often resulting in time delays and loss of results.

Innovation Solution

A computer-implemented method and graphical user interface that allows users to input base number of trials, number of center point trials, and target effect size to generate a table or matrix displaying predicted power values for each combination of replicates count and effect size, enabling selection of experiment options with adequate power and cost considerations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional DOE processes are used without power prediction, then the experiment can be set up quickly with minimal planning, but the experiment lacks sufficient power to achieve target goals, requiring additional trial runs that increase time and cost

Engineering Contradiction:

Improveexperimental powerVSAvoidtime delay

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary power analysis and predictions before the experiment is executed. By calculating expected power values based on proposed design parameters (number of trials, factors, levels) before actual experimentation begins, the system enables experimenters to assess whether the design will achieve sufficient power to detect meaningful effects, thereby preventing wasted time on underpowered experiments

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system provides feedback on expected experimental power based on the proposed design parameters. This feedback mechanism allows experimenters to adjust their experimental design (e.g., increase number of trials, modify factor levels) before committing resources, creating a loop where power predictions inform design decisions that subsequently improve power adequacy

Inventive Principle:

Principle #23Feedback

2Reliability

If additional trial runs are added after experiment to address power deficiency, then the power adequacy may be improved, but the project incurs additional cost and time delay

Engineering Contradiction:

Improveexperimental powerVSAvoidadditional cost

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The system performs preliminary power analysis and predictions before the experiment is executed. By calculating expected power values based on proposed design parameters (number of trials, factors, levels) before actual experimentation begins, the system enables experimenters to assess whether the design will achieve sufficient power to detect meaningful effects, thereby preventing wasted time on underpowered experiments

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system provides feedback on expected experimental power based on the proposed design parameters. This feedback mechanism allows experimenters to adjust their experimental design (e.g., increase number of trials, modify factor levels) before committing resources, creating a loop where power predictions inform design decisions that subsequently improve power adequacy

Inventive Principle:

Principle #23Feedback

3Ease of operation

If experimenters rely on experience and insight to assess power adequacy, then the process remains simple and quick, but the assessment is subjective and may lead to insufficient power

Engineering Contradiction:

Improvesimplicity of processVSAvoidpower assessment accuracy

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The system performs self-service by automatically calculating power predictions based on input parameters without requiring the experimenter to manually perform complex statistical calculations. The system takes experimental design parameters as input and autonomously generates power analysis results, making precise power assessment accessible to users regardless of their statistical expertise

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system replaces the manual, experience-based mechanical assessment process with an automated computational system. Instead of relying on human expertise and subjective judgment, the system uses algorithms to calculate power predictions, substituting human cognitive processes with machine-based calculations that are more consistent and objectively measurable

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS9965721B2Computer-implemented system and method of performing a design of experiment, generating a product thereof, generating designed experiment options, and a graphical user interface

Publication Date: 2018.05.08 MORESTEAM DEV

AI summary

Disclosed is a computer-implemented method of generating a plurality of selectable design experiments in a design of experiments (DOE) process for analyzing at least one data set from a process to determine a relationship of a plurality of process factors of interest to a process output of interest. The method entails receiving as user input into a computing apparatus, a base number of trials and a number of center point trials. From the computing apparatus, a graphical user interface is generated on a display connected with the computing apparatus. This includes presenting a display containing a plurality of experiment each defined, at least partially, by a combination of number of replicates (replicates count) and effect size. The display further includes a predicted power value for each combination of replicates count and effect size.