.png){kind=icon}
Return
A Situation-driven Synchronization Approach for Constructing Valid Process Digital Twins
.png){kind=icon}
PRE
AI
DOI：10.1109/jiot.2026.3743161.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Process digital twins (DTs) describe the execution of a process, such as a production line, by replicating the process in a digital or virtual environment. DT synchronization is a key enabling technology for maintaining correct state replication, which faces the challenge of synchronization validity. Due to the characteristics of DT synchronization, existing approaches have limitations in either accuracy or in consistency, possibly resulting in incorrect semantic interpretation of a process. To address the problem, this study proposes a situation-driven DT synchronization framework. It employs a situation graph model to enrich the semantics among operations with abstract situations. With the model, a situation inference method is then proposed, which combines both deterministic and probabilistic inference techniques to accurately infer inaccessible states with both field data and business process data. Next, real-time situations are correctly and consistently converted to simulation operations for process execution through situation reification. Both situation inference and situation reification routines are cyclically scheduled with a situation evolution scheme, which properly changes situation states to align changes in related situations. The evaluation results from theoretical analysis, experiments, and a case study demonstrate the effectiveness of the solution in improving DT synchronization validity, robustness to several adverse conditions, and usefulness in dynamic production scheduling, and its advantage in enhancing synchronization validity and industrial productivity.
Keywords:
Digital twin
data synchronization
situation graph
situation inference
dynamic scheduling
Journal
IF:
8.9
Papers:
1.4W
Citations:
7.8W
Organization
View more organizations
Cited Papers
No cited papers available