Workflow Flow Control Method for Dropout Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional workflow generation for measure planning often fails to consider the perspectives of both the implementing and receiving sides, leading to interventions that are not suitable for individuals, resulting in high dropout rates and discrepancies between planned and actual results.

Innovation Solution

A flow control method that extracts subjects who are likely to dropout from the current workflow, determines an insertion point for a new conditional branch, and inserts a new condition and node to guide these subjects to a more suitable intervention, enhancing the workflow's suitability for both sides.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If conventional workflow generation is used, then workflow creation is simple, but dropout rates increase and suitability for individual needs deteriorates

Engineering Contradiction:
Improveworkflow creation simplicityVSAvoiddropout rate
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The system performs preliminary analysis of subject characteristics and workflow paths before finalizing the workflow. It predicts dropout probabilities in advance and pre-determines appropriate conditional branches and nodes to insert, ensuring interventions are tailored to individual needs before execution begins

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system incorporates feedback loops that analyze actual subject responses and outcomes against predicted dropout patterns. This feedback is used to refine future workflow generation, adjusting conditional branches and node selections to reduce dropout rates while maintaining workflow creation simplicity

Inventive Principle:
Principle #23Feedback

2Device complexity

If conventional workflow generation is used, then workflow structure remains simple, but intervention suitability for individual perspectives deteriorates

Engineering Contradiction:
Improveworkflow structure complexityVSAvoidintervention suitability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The system segments the workflow into distinct conditional branches and nodes based on subject characteristics. Each segment can be independently analyzed and optimized for specific subject groups, enabling tailored interventions without overwhelming complexity in the overall workflow structure

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system applies local quality by customizing specific portions of the workflow (conditional branches and nodes) based on the characteristics of individual subjects or subject groups. This allows high adaptability in critical areas while maintaining simplicity in other workflow regions

Inventive Principle:
Principle #3Local quality

3Ease of operation

If subjects are not separated by conditions, then workflow is simple to execute, but intervention effectiveness deteriorates

Engineering Contradiction:
Improveworkflow execution simplicityVSAvoidintervention effectiveness
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The system introduces dynamic conditional branches that automatically adjust workflow paths based on real-time subject characteristics and predicted dropout probabilities. This dynamic adaptation improves intervention effectiveness while maintaining ease of execution through automated decision-making

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4390798A1Flow control method, flow control program, and information processing device
Publication Date: 2024.06.26 FUJITSU LTD
  • EP4390798A1 patent drawingFigure 1
  • EP4390798A1 patent drawingFigure 2
  • EP4390798A1 patent drawingFigure 3

AI summary

A flow control method executed by a computer, includes: extracting, from a plurality of subjects, one or more first subjects each satisfying a first condition, the one or more first subjects being extracted based on a result of assignment of the plurality of subjects to a node based on a workflow that includes the node connected to a conditional branch by a directed edge; determining an insertion point of a new conditional branch in the workflow, based on distribution of the one or more extracted first subjects in the workflow; inserting a second condition, as the new conditional branch, at the insertion point, the second condition separating, at a lower node that is lower than the determined insertion point, the one or more first subjects from a subject not satisfying the first condition among the plurality of subjects; and inserting a new node at a branch destination of the new conditional branch.