Forecasting Operator for Event Stream Prediction in CEP

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current Complex Event Processing (CEP) systems lack effective methods for forecasting future event streams in real-time, particularly in high-volume environments, leading to inadequate proactive system management and decision-making due to limitations in existing database approaches and lack of integrated forecasting functionality in Event Query Languages.

Innovation Solution

An online forecasting operator is introduced that computes forecasts within a continuously moving time window, allowing for flexible adaptation and integration into the Event Query Language, enabling the estimation of future event values and temporal information, and supporting different event stream representations and forecasting strategies.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If traditional CEP systems process high-volume event streams in real-time, then system responsiveness and detection speed are improved, but the ability to forecast future behavior and enable proactive management deteriorates

Engineering Contradiction:
Improveevent processing speedVSAvoidfuture behavior prediction capability
Core Design Contradiction:
SpeedVSLoss of information

Solution Approach 1:

The patent applies preliminary action by computing forecasts of future event stream behavior in advance, before the actual events occur. The forecasting operator analyzes historical patterns within a reference window and generates predictions about future event characteristics, enabling the CEP system to proactively prepare for upcoming load variations and optimize resource allocation beforehand rather than reacting after events occur.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If a CEP system integrates forecasting functionality to estimate future stream behavior, then proactive system management and resource allocation are improved, but system complexity increases

Engineering Contradiction:
Improveproactive system management capabilityVSAvoidsystem architecture complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The forecasting operator is designed as a universal component that can be integrated into the existing CEP system architecture without requiring separate dedicated forecasting systems. It multi-functionally serves both traditional event processing roles and forecasting roles, working with various event stream representations and supporting different forecasting strategies through configurable parameters, thereby reducing overall system complexity while enabling proactive management.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The forecasting operator acts as an intermediary component between historical event data and future behavior predictions. It mediates by taking events from the input stream, applying forecasting algorithms within a reference window, and producing forecasted events that represent future stream behavior. This intermediary layer integrates forecasting functionality into the existing CEP pipeline without fundamentally restructuring the entire system architecture.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If forecasts are computed in an online manner with continuously moving time windows, then forecast accuracy and timeliness are improved, but computational overhead increases

Engineering Contradiction:
Improveforecast accuracyVSAvoidcomputational resource consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent applies dynamics by implementing a continuously moving reference window that adapts to changing event stream characteristics. The reference window dynamically adjusts its position and content as new events arrive, allowing the forecasting operator to capture temporal patterns and trends in real-time. This dynamic approach improves forecast accuracy compared to static windows while managing computational overhead through efficient window update mechanisms.

Inventive Principle:
Principle #15Dynamics

4Adaptability or versatility

If the forecasting framework is flexibly defined and parameterized to allow tailored adaptation, then adaptability to different event representations is improved, but device complexity increases

Engineering Contradiction:
Improveevent stream representation flexibilityVSAvoidforecasting framework complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The forecasting framework employs parameter changes to achieve flexibility, allowing users to configure various parameters such as reference window size, forecasting horizon, and algorithmic parameters based on specific event stream characteristics and requirements. By adjusting these parameters rather than redesigning the entire framework, the system adapts to different event representations and forecasting needs while maintaining a consistent underlying architecture, thereby limiting the increase in device complexity.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2560106B1Systems and/or methods for forecasting future behavior of event streams in complex event processing (CEP) environments
Publication Date: 2019.11.13 SOFTWARE AG
  • EP2560106B1 patent drawingFigure 1
  • EP2560106B1 patent drawingFigure 2
  • EP2560106B1 patent drawingFigure 3

AI summary

Certain example embodiments described herein relate to forecasting the future behavior of event streams in Complex Event Processing (CEP) environments. For each received event in an event stream, a reference window indicative of a predefined temporal range during which the forecast is to be computed is updated so that the reference window ends with the received event, with the reference window moving with the event stream. Within this processing loop, when a forecasting update policy indicates that the forecast is to be updated based on the received event: a forecasting window indicative of a temporal range in which events are to be forecasted is updated; and while the time period of the forecasting window is not exceeded, (a) a next forecasted event is generated via at least one processor and (b) the next forecasted event is inserted into the forecast window; and the forecast window is published.