Machine-Learning Load Prediction for Message Sending Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing message sending systems struggle to effectively prevent failures due to insufficient time for administrators to address load issues, as current monitoring systems often fail to provide timely interventions.

Innovation Solution

A method and system utilizing a machine-learning model to predict future load of a message sending module, allowing for proactive control of message sending speed, schedule adjustments, and sending module speed adjustments based on predicted load, with optional administrator notifications.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a notification is sent to an administrator when an issue occurs in the message sending system, then the administrator is informed of the problem, but the administrator does not have sufficient time to resolve the issue before system failure occurs

Engineering Contradiction:
Improvesystem reliabilityVSAvoidresponse time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary action by predicting future load values before the system actually experiences overload. The load prediction module calculates anticipated load values at future time points and compares them against threshold values in advance, enabling the administrator to take corrective action before the system fails, thus resolving the contradiction between early warning and sufficient response time

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system applies preliminary anti-action by proactively notifying the administrator of predicted future load issues before they materialize. By comparing predicted load values with threshold values in advance and sending notifications beforehand, the system prevents the harmful effect of system failure while giving the administrator time to counteract the impending problem

Inventive Principle:
Principle #9Preliminary anti-action

2Productivity

If the message sending system operates without load prediction, then the system can send messages at current speed, but the system may experience failures due to unexpected load increases

Engineering Contradiction:
Improvemessage sending speedVSAvoidsystem reliability
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The load prediction module performs preliminary calculation of future load values based on current and historical data. By predicting load trends in advance and comparing them against threshold values, the system can maintain current message sending speed while proactively identifying potential overload conditions, thus preserving productivity while improving reliability

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by continuously monitoring current load values, predicting future load values, comparing predictions against thresholds, and sending notifications to the administrator. This closed-loop feedback mechanism enables the system to maintain high message sending speed while automatically detecting and warning about potential reliability issues before they cause failures

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4518287B1Method for controlling message sending and message service system using the same
Publication Date: 2025.09.17 SAMSUNG SDS CO LTD
  • EP4518287B1 patent drawingFigure 1
  • EP4518287B1 patent drawingFigure 2
  • EP4518287B1 patent drawingFigure 3

AI summary

There is provided a method for controlling message sending, performed by at least one computing device The method may comprise acquiring sending information of a target message; configuring input data of a machine-learning model based on the sending information, the machine-learning model being trained through a task of predicting a future load of a message sending module; predicting a load of the message sending module according to sending the target message from the input data through the machine-learning model; and controlling the message sending module based on the predicted load.