ML and Combinatorial Optimization Framework for Dynamic Task Management

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Managing housekeeping services in medical care facilities, such as hospitals, is challenging due to the difficulty in estimating demand and turnaround times for tasks like bed cleaning, which affects patient flow and resource allocation, especially with limited resources and service level agreements to meet.

Innovation Solution

A system utilizing machine-learning and combinatorial optimization frameworks to forecast demand and turnaround times, prioritize tasks, and allocate resources in real-time, combining diagnostic analytics to optimize housekeeping operations and meet service level requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional manual methods are used to manage housekeeping tasks, then operational flexibility is maintained, but turnaround times are excessive and resource allocation is inefficient

Engineering Contradiction:

Improveturnaround timeVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent replaces manual mechanical management methods with an automated computer-based system that uses machine learning models and combinatorial optimization algorithms to forecast demand, predict turnaround times, and allocate resources dynamically, thereby reducing turnaround times while accepting increased system complexity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service through automated demand forecasting and resource allocation without continuous human intervention. The machine learning models continuously learn from historical data and automatically adjust predictions and allocations, allowing the system to manage itself while improving productivity

Inventive Principle:

Principle #25Self-service

2Reliability

If more resources are allocated to housekeeping tasks, then service level agreements are met, but resource imbalance and waste increase

Engineering Contradiction:

Improveservice level agreement complianceVSAvoidresource waste

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The patent implements dynamic resource allocation where the system continuously adjusts resource distribution based on real-time demand forecasts and predicted turnaround times. Resources are allocated dynamically to match actual needs rather than being statically distributed, ensuring service level agreement compliance while minimizing resource waste through optimized allocation

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes allocation parameters dynamically by adjusting the number and positioning of resources based on forecasted demand and predicted turnaround times. The combinatorial optimization algorithm evaluates multiple allocation scenarios and selects the optimal parameter configuration that meets service level requirements while minimizing resource waste

Inventive Principle:

Principle #35Parameter changes

3Reliability

If demand is overestimated to ensure service level compliance, then service level agreements are met, but resource allocation becomes inefficient

Engineering Contradiction:

Improveservice level agreement complianceVSAvoidresource allocation efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent implements feedback mechanisms where the system continuously monitors actual task completion times and compares them with predicted turnaround times. The machine learning models use this feedback to refine their predictions and adjust resource allocation dynamically, ensuring service level agreement compliance while improving resource allocation efficiency through continuous learning and adaptation

Inventive Principle:

Principle #23Feedback

4Productivity

If manual task prioritization is used, then operational simplicity is maintained, but turnaround times are excessive

Engineering Contradiction:

Improveturnaround timeVSAvoidoperational simplicity

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The patent replaces manual task prioritization with an automated system that uses combinatorial optimization algorithms to determine optimal task sequences. The system evaluates multiple prioritization scenarios and automatically selects the optimal sequence that minimizes turnaround times, accepting reduced operational simplicity in exchange for significantly improved productivity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11393577B2Machine-learning and combinatorial optimization framework for managing tasks of a dynamic system with limited resources

Publication Date: 2022.07.19 GE PRECISION HEALTHCARE LLC

AI summary

Techniques are described for managing tasks of a dynamic system with limited resources using a machine-learning and combinatorial optimization framework. In one embodiment, a computer-implemented method is provided that comprises employing, by a system operatively coupled to a processor, one or more first machine learning models to determine a total demand for tasks of a dynamic system within a defined time frame based on state information regarding a current state of the dynamic system, wherein the state information comprises task information regarding currently pending tasks of the tasks. The method further comprises, employing, by the system, one or more second machine learning models to determine turnaround times for completing the tasks based on the state information, and determining, by the system, a prioritization order for performing the currently pending tasks based on the total demand and the turnaround times.