Machine-Learning Time-Shifted Action Scheduling for Resource Shortages

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing computing systems face inefficiencies and resource shortages due to the need to execute action requests immediately, which can lead to variability in workload and resource planning challenges, particularly in systems handling numerous requests daily.

Innovation Solution

A machine-learning-based system that utilizes historical data to train a model to predict a time-shifted target date for action execution, optimizing resource utilization by scheduling actions for execution at a future date when computing resources are available, thereby reducing the immediate demand on resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If action requests are executed immediately upon receipt, then the requesting party receives prompt service, but computing resources (memory, processor, user resources) are consumed at variable rates leading to resource shortages and inefficiencies

Engineering Contradiction:

ImproveExecution speed of action requestsVSAvoidResource utilization efficiency

Core Design Contradiction:

SpeedVSProductivity

Solution Approach 1:

The system performs preliminary actions by receiving and validating action requests immediately, then scheduling their execution for later times when resources are available. The request is processed in advance (validated, formatted, and queued) but the actual execution is deferred to an optimized time, allowing the system to prepare without immediately consuming full execution resources.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts the execution timing of action requests based on real-time and historical resource utilization patterns. By using machine learning models trained on historical data, the system flexibly schedules executions during periods of lower resource demand, making the execution timeline adaptive rather than fixed or immediate.

Inventive Principle:

Principle #15Dynamics

2Ease of operation

If action requests are executed immediately, then service responsiveness is high, but workload variability causes resource planning challenges

Engineering Contradiction:

ImproveService responsivenessVSAvoidResource planning complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The system introduces an intermediary scheduling layer between request reception and execution. This intermediary component (the scheduling system with ML models) buffers the direct connection between incoming requests and resource consumption, smoothing out workload variability and simplifying resource planning while maintaining service responsiveness through immediate request acknowledgment.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system uses feedback from historical execution data and resource utilization metrics to continuously optimize scheduling decisions. By training machine learning models on past performance data, the system learns from feedback loops to predict optimal execution times, reducing resource planning complexity through data-driven insights.

Inventive Principle:

Principle #23Feedback

3Adaptability or versatility

If actions are executed at variable rates based on incoming requests, then service flexibility is maintained, but resource consumption becomes unpredictable

Engineering Contradiction:

ImproveService flexibilityVSAvoidExecution timing optimization

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system implements periodic execution of action requests based on predicted resource availability patterns rather than immediate execution. By scheduling executions during identified optimal periods (when resource utilization is lower), the system maintains service flexibility while reducing overall resource consumption and improving execution timing efficiency.

Inventive Principle:

Principle #19Periodic action

Data Source

PatentUS20250272156A1Conserving computing resources by time shifting electronic action request execution operations

Publication Date: 2025.08.28 TRUIST BANK

AI summary

A system for processing a request to execute an action by a future date is disclosed. The system may access training data including historical action data, historical execution times, and historical utilization rate information for one or more execution objects, and can use the training data to train a machine-learning model. Upon receipt of a request to execute an action at a future date, action execution information provided in the request, the future date, and at least the current date, can be input to the trained machine-learning model, which may be configured to generate an output indicating a time-shifted target date on which the execution objects are available to execute the requested action that is no later than the future date. The system can then schedule the requested action for execution by the execution objects on the time-shifted target date.