Heterogeneity-Aware Personalized Federated Learning for Industrial Predictive Analytics

TL;DR

This paper introduces a personalized federated prognostic model tailored for industrial settings with heterogeneous degradation processes, enhancing failure prediction while preserving data privacy.

Contribution

It proposes a novel federated parameter estimation algorithm and a personalized modeling approach that accounts for client heterogeneity in industrial prognostics.

Findings

The model outperforms traditional federated models in simulations.

It effectively captures diverse degradation patterns across clients.

Validated on NASA turbofan engine data.

Abstract

Federated prognostics enable clients (e.g., companies, factories, and production lines) to collaboratively develop a failure time prediction model while keeping each client's data local and confidential. However, traditional federated models often assume homogeneity in the degradation processes across clients, an assumption that may not hold in many industrial settings. To overcome this, this paper proposes a personalized federated prognostic model designed to accommodate clients with heterogeneous degradation processes, allowing them to build tailored prognostic models. The prognostic model iteratively facilitates the underlying pairwise collaborations between clients with similar degradation patterns, which enhances the performance of personalized federated learning. To estimate parameters jointly using decentralized datasets, we develop a federated parameter estimation algorithm based on proximal gradient descent. The proposed approach addresses the limitations of existing federated prognostic models by simultaneously achieving model personalization, preserving data privacy, and providing comprehensive failure time distributions. The superiority of the proposed model is validated through extensive simulation studies and a case study using the turbofan engine degradation dataset from the NASA repository.