A synergistic engine paradigm for real-time, context-aware decision-making: integrating declarative processes and event streams
Year: 2026
Type: article
Abstract: The abstraction gap between high-frequency IoT data and high-level business process logic creates a significant bottleneck for modern enterprises. Current architectures typically rely on separate middleware for event preprocessing, which introduces significant latency due to network hops and data serialization, and increases architectural complexity, creating multiple points of failure that hinder responsive operations. This paper introduces a synergistic engine paradigm that resolves this gap by leveraging a single complex event processing engine for both event abstraction and the direct execution of declarative MP-Declare models. Through a multi-level abstraction framework, process constraints are translated into executable queries, as demonstrated by a proof-of-concept. This unified approach provides a simplified architectural foundation for building highly responsive, event-driven applications that adapt intelligently to real-time conditions, as demonstrated by a proof-of-concept and a quantitative evaluation showing sub-millisecond latency at up to 10,000 events per second.
Source: Complex & Intelligent Systems
Authors: Leo Poss, Stefan Schönig
Institution: University of Regensburg
Language: English
Subfield: Management Information Systems
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $2,490
Funder: Universität Regensburg
OpenAlex ID: W7135220071