Log in
Back to #federated learning
#federated learning

Toward a Federated Cross-Organizational Construction Safety Knowledge Recommendation: A Dual-Incentive Mechanism Integrating Worker Satisfaction

Shuai Zhang Botao Zhong Samad M. E. Sepasgozar Zhenzhao Xia Xueyang Zhong Qili Guan
Oct 2026 · Journal of construction engineering and management · 0 citations · 25 references

TL;DR

This study proposes a dual-incentive mechanism that integrates model training performance with worker satisfaction, encouraging providers to allocate more resources to safety knowledge recommendation and shows that the dual-incentive mechanism yields stronger incentive intensity, higher provider effort, and greater requester benefits versus single-model training incentives, while remaining robust across cost and risk variations.

Abstract

Despite the effectiveness of personalized safety knowledge enhancement training, existing research overlooks the challenges posed by fragmented worker data and strict privacy regulations. To overcome these challenges, this study proposes a novel FedCSKR framework aimed at achieving privacy-preserving cross-organizational construction safety knowledge recommendation (CSKR). Nonetheless, the effectiveness of CSKR depends on the decisions of both requesters and providers. We develop a game model to capture their interactions and show, through evolutionary analysis, that spontaneous collaboration is unsustainable, leading the system to a noncooperative equilibrium without external incentives. To promote cooperation, we propose a dual-incentive mechanism that integrates model training performance with worker satisfaction, encouraging providers to allocate more resources to safety knowledge recommendation. Modeling and simulations show that the dual-incentive mechanism yields stronger incentive intensity, higher provider effort, and greater requester benefits versus single-model training incentives, while remaining robust across cost and risk variations. Specifically, incentive intensity can be increased by over 98.3% and provider effort levels by more than 197.4%, while simultaneously helping to mitigate the requester’s exposure to uncertainty arising from model performance variability. Theoretical contributions of this study include the integration of federated learning with incentive-driven contract design in CSKR, representing the first effort, to our knowledge, that incorporates worker satisfaction into the incentive mechanism of the FedCSKR. Practically, this study offers a structured decision-making framework to assist construction managers in implementing privacy-compliant, effective, and collaborative CSKR. Overall, this study facilitates cross-organizational collaboration in CSKR and contributes to advancing personalized safety training.

DOI Semantic Scholar
View source

Similar papers

#federated learning Open access Oct 2026

SoK: Verifiable Integrity Claims for Privacy-Preserving Federated Learning

This SoK model federated learning as an append-only transcript of submissions, admissions, aggregation, and finalization events, and formalize verifiability as a collection of integrity claims issued by clients and the aggregator, and checked by different verifier classes.

Andrea Rizzini, Marco Esposito, Tommaso Gagliardoni et al. · 0 citations
#federated learning Sep 2026

A Decentralized Learning Architecture for Medical Prescription Anomaly Detection via Hybrid Federated-Swarm Learning

The increasing use of electronic medical records (EMRs) has improved efficiency, accuracy, and accessibility of patient data. However, conventional centralized architectures suffer from single points of failure and data privacy issues. To address these challenges, this study proposes a decentralized machine learning architecture that combines concepts from Federated Learning (FL) and Swarm Learning (SL) for anomaly detection in medical prescriptions. The proposed architecture leverages blockchain and the InterPlanetary File System (IPFS) to enable secure model sharing and decentralized storage, thereby reducing communication complexity and establishing a transparent, decentralized parameter repository. Experimental evaluations were conducted using logistic regression (LR), a multi-layer perceptron (MLP), and a decision tree (DT) model. Compared with the FL baseline, the proposed system achieved superior efficiency, lower resource consumption, and improved latency, along with smaller block sizes. It, however, exhibited slightly lower transaction throughput and longer training rounds, reflecting the added complexity of decentralization. In predictive performance on the anomaly classification task, DT achieved the highest precision and recall under the evaluated dataset (F1-score=0.9912), followed by MLP (0.5504) and LR (0.2442). The decentralized training approach led to negligible performance loss relative to centralized models, less than 4% for LR and below 1% for both MLP and DT. Overall, the proposed system demonstrates a robust and efficient alternative for decentralized learning in healthcare applications, maintaining strong predictive performance while enhancing architectural transparency.

Ravelly Zanatta, Vinícius Vanzin, S. Matos et al. · 0 citations

Related blog posts