event tracking design

In manufacturing digital transformation, low-efficiency fusion of multi-source heterogeneous production event data and highly customized preprocessing hinder scalability and consistency. Method: This paper proposes a standardized reference model integrating the ISA-95 standard with an event knowledge graph, formally deriving reusable and automatable generic event data augmentation patterns. It pioneers the coupling of ISA-95’s hierarchical semantics with the structured representation of event knowledge graphs, employing empirically driven pattern engineering for modeling and validation. Contribution/Results: We establish an industrially applicable unified framework for event data storage and extraction, and deliver a validated suite of automation-ready data augmentation patterns. Evaluated across multiple real-world manufacturing scenarios, the approach significantly reduces preprocessing customization effort and deployment time, while enhancing scalability and analytical consistency of event data processing.

Traditional Business Process Management (BPM) struggles to unify discrete events with continuous sensor signals from Cyber-Physical Systems (CPS). Existing Signal Temporal Logic (STL)-based hybrid declarative approaches support only retrospective monitoring and lack real-time execution capabilities. To address this, we propose the first three-layer architecture enabling real-time execution of hybrid declarative processes, deeply integrating STL into a Complex Event Processing (CEP) engine. This integration supports joint temporal constraints over discrete events and real-valued signals, proactive activity triggering, and dynamic enforcement of process boundaries. Our approach achieves, for the first time in BPM, STL-driven online execution and closed-loop control—bridging the semantic and execution gap between declarative modeling and real-time physical-world operation. We validate its effectiveness and scalability in智能制造 and Industrial IoT scenarios.

Existing evaluation approaches for streaming process mining algorithms predominantly rely on static logs or synthetic event streams, which fail to capture the complexity of real-world event streams in IoT environments—such as out-of-order events, concurrency, incomplete cases, and concept drift. This work addresses this gap by introducing, for the first time, a feature framework from data stream research into streaming process mining. It proposes an intent-oriented event stream generation methodology, extends the conceptual model of event streams, and implements a prototype tool, Stream of Intent. This tool enables customizable configuration of key stream characteristics reflective of real-world scenarios, facilitating the generation of controlled, reproducible, and realistically complex event streams. Consequently, it significantly enhances the relevance and adaptability of algorithm evaluation and development in streaming process mining.

IoT sensors generate massive volumes of fine-grained streaming data, which are ill-suited for direct business-process-level analysis and mining. To address this, we propose Radiant—a domain-specific language (DSL) tailored for process activity recognition—along with a corresponding stream processing architecture. Radiant enables domain experts to declaratively specify abstraction rules mapping raw sensor data to high-level business events; these rules are executed in real time via complex event processing (CEP), yielding interpretable, low-latency mappings from perceptual data to process activities. Our key contributions are: (1) the first DSL dedicated to IoT-based process activity detection, balancing expressive power with usability; and (2) a feedback-driven quality assessment mechanism enabling continuous refinement of detection accuracy. Experiments in smart manufacturing and intelligent healthcare demonstrate significant improvements in recognition accuracy and latency, alongside strong configurability and maintainability.

In process intelligence, model abstraction (MA) and event abstraction (EA) are often misaligned, causing simplified process models to lose behavioral grounding in the original event log and undermining analytical reliability. Method: This paper introduces the first formal framework for synchronizing MA and EA. It employs behavior profiles to realize non-order-preserving model abstraction and derives a semantically equivalent event abstraction method therefrom. Contribution/Results: We formally prove that synchronized abstraction strictly preserves behavioral equivalence—overcoming the fundamental limitation of conventional abstraction approaches, which yield logs lacking support for the abstracted model. The proposed mechanism enables discovery of behaviorally equivalent process models, ensuring all process analyses remain grounded in actual event behavior. By guaranteeing semantic consistency between abstracted models and their underlying event data, our framework establishes a novel foundation for interpretable and verifiable process intelligence.

Modern OLTP systems often suffer from frequent schema changes, missing primary/foreign keys, and fragmented execution traces, rendering traditional approaches—reliant on fixed schemas and manual modeling—costly and error-prone. This work proposes a fully automated pipeline that operates without predefined schemas by identifying quasi-key and timestamp columns, discovering inter-table relationships through statistical signals, and assembling and ordering events accordingly. To capture long-range dependencies across system events, the method incorporates a Temporal Convolutional Network (TCN). By eliminating dependence on ER diagrams, domain-specific templates, and stable schemas, the approach enables generalizable and scalable reconstruction of execution traces in dynamic information systems. Experimental results on TPC-H/E, synthetic, and real-world industrial datasets demonstrate 85% accuracy in event prediction and recovery of approximately 82% of true predecessor relationships, yielding high-fidelity process traces.

This study addresses the challenge of cross-organizational business process modeling, where access to comprehensive event logs from multiple parties is often limited, while message interaction records are more readily available. To overcome this limitation, the authors propose a novel approach that transforms message logs into causally consistent event logs for the first time, enabling their integration with established process mining algorithms. By doing so, the method automatically synthesizes accurate Industry Process Nets that faithfully capture inter-organizational collaboration patterns. This advancement significantly broadens the applicability and data sources of process mining in cross-organizational settings, demonstrating that precise and efficient discovery of collaborative business processes can be achieved using only message exchange records.

This work addresses the challenge of effectively integrating Internet of Things (IoT) data with business process event logs, which stems from their heterogeneous origins and differing levels of abstraction. Direct incorporation of raw IoT data often leads to excessive log complexity, thereby hindering process analysis. To overcome this, the paper proposes a structured fusion approach within the Object-Centric Event Log (OCEL) framework, introducing a mapping and integration mechanism that seamlessly embeds process-relevant IoT data into standard OCEL logs without requiring domain-specific schemas. The resulting IoT-enhanced event logs, generated by an extensible tool, are fully compatible with mainstream process mining tools. Empirical validation in real-world scenarios demonstrates the method’s effectiveness in enabling downstream process analysis and visualization.

This work addresses the challenge of synchronously comparing event sequences with heterogeneous tabular attributes—such as numerical, categorical, and temporal data—in visual analytics. To this end, the authors propose EventColumn, a novel column type that unifies event sequences and diverse attribute types within a single coordinated view, enabling both instance-level and cohort-level joint analysis. Through interactive mechanisms including compact overviews, heatmap-based summaries, event alignment, and boxplots of historically similar items, the approach significantly enhances analytical efficiency. Implemented in Taggle and Microsoft Power BI, the method has been validated on real-world datasets from steel production logistics and public e-commerce domains, demonstrating its effectiveness in structured event data scenarios and substantially improving users’ ability to discern relationships between events and associated attributes.

The lack of systematic integration between Kafka design patterns and benchmarking methodologies hinders reproducible, evidence-based architectural decision-making for event-streaming systems. Method: This study systematically analyzes 42 academic and industrial publications (2015–2025) using bibliometric analysis and pattern induction, complemented by standardized (TPCx-Kafka, Yahoo Streaming Benchmark) and customized workload evaluations. Contribution/Results: We introduce the first unified Kafka architectural pattern taxonomy—comprising nine high-frequency patterns—and establish a pattern-to-benchmark mapping matrix. Additionally, we propose heuristic guidelines to support architecture-level decisions. Our work fills a critical gap in reproducible design methodology for event-streaming systems, significantly improving design quality and practical consistency across performance, fault tolerance, and cross-study comparability dimensions.