Grounding Time-Series Foundation Models in Digital Twin Topology for Predictive Maintenance

2026-09-30

AI Summary

This study addresses the performance limitations of existing time-series foundation models in digital twin-based predictive maintenance, where the absence of physical topology constraints hinders multivariate regression tasks. To overcome this, we propose a topology-aware fusion mechanism that explicitly embeds digital twin asset connectivity into the cross-attention computation of frozen-parameter foundation models, thereby guiding the precise fusion of multi-source sensor data. This work reveals the complementary effects among pretrained weights, task adaptation, and topological representations. Evaluations on the C-MAPSS benchmark demonstrate that incorporating topological constraints significantly outperforms unconstrained fusion approaches, enabling frozen foundation models to match or surpass state-of-the-art specialized models for remaining useful life prediction under complex operating conditions.

Abstract

Digital twins increasingly support downstream analytical tasks that depend on time-series data, motivating interest in time-series foundation models (TSFMs) as scalable backbones. However, TSFMs are primarily pretrained for temporal continuation and often underperform on unseen tasks such as regression, and systematic empirical comparisons against state-of-the-art dedicated models in digital twin contexts remain limited. This paper makes three contributions. First, we benchmark five well-known TSFMs with frozen backbones on remaining useful life (RUL) prediction using the C-MAPSS dataset, finding that multivariate architectures substantially outperform univariate ones, particularly under varying operating conditions. This raises a deeper question: when cross-channel dependencies can be modeled through pretrained weights, target-task adaptation, and digital twin-derived representations, how much does each contribute, and are they complementary? Second, we propose a topology-informed fusion approach in which topological constraints, derived from the asset structure the digital twin stores among its information models, explicitly shape cross-attention, so that fused representations respect the physical system's local connectivity rather than relying on unconstrained all-to-all interactions. Third, we conduct an ablation study across C-MAPSS subsets of varying operational complexity that isolates the three sources and their interactions. The sources prove complementary rather than redundant, and topology-constrained attention outperforms unconstrained fusion, though by a small margin, enabling a frozen TSFM informed by digital twin representations to remain competitive or in some cases exceed state-of-the-art performance on this regression task.

Problem

• Time-Series Foundation Models
• Digital Twin
• Predictive Maintenance
• Remaining Useful Life
• Topology-Informed Fusion

Innovation

• Time-Series Foundation Models
• Digital Twin Topology
• Topology-Informed Fusion
• Predictive Maintenance
• Remaining Useful Life