Multi-Layer Write-Time Verification of Data Provenance and Workflow Integrity for Blockchain-Based Robot Maintenance Records
Electronics · 12.11.2026 · 15 (22) · 5261
DOI: 10.3390/electronics15225261 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 12.11.2026. Verfasst von Jin Tian und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Blockchain-based robot maintenance records cannot be altered once written, but immutability does not prevent unjustified or fabricated records from being written. Contracts that enforce business processes on-chain guarantee that activities occur in the permitted order, yet accept whatever data participants submit. This paper proposes multi-layer write-time verification for robot maintenance records: before a record is stored, a source check verifies a device signature, a sequence number, and a raw-data commitment; a consistency check binds the confirmed diagnosis to the severity attested by an edge diagnostic model; a cross-contract prerequisite check confirms the preceding workflow stage; and lifecycle rules enforce unique orders, role re-checks, bounded rework, and dual sign-off. The mechanism was implemented as four Solidity contracts and evaluated on a local Ethereum network and on a four-node Hyperledger Besu network with the QBFT Byzantine fault-tolerant consensus protocol. In an ablation over 13 attack types (260 attempts), admitted invalid records fell from 260 with role-based access control alone to 0 with all layers. Invariant fuzzing found no violation of seven safety properties in the full design and counterexamples in every reduced variant. Full verification raised the gas per maintenance cycle from 594,644 to 736,810, 6.5% more than a single-contract design. On the four-node network, gas was unchanged, latency was set by the 2 s block period, and throughput grew linearly to 122 records per second with 256 concurrent uploads. In a case study on 885 Case Western Reserve University (CWRU) bearing windows recorded under an unseen load, the edge model graded 97.97% correctly and all 472 work orders reached verified acceptance. Write-time verification of both data provenance and workflow state is therefore a practical design pattern for multi-party maintenance records.
Zitieren
APA
Tian, J., Zhang, J., & Kim, S. (2026). Multi-Layer Write-Time Verification of Data Provenance and Workflow Integrity for Blockchain-Based Robot Maintenance Records. Electronics, 15(22), 5261. https://doi.org/10.3390/electronics15225261
Vancouver
Tian J, Zhang J, Kim S. Multi-Layer Write-Time Verification of Data Provenance and Workflow Integrity for Blockchain-Based Robot Maintenance Records. Electronics. 2026;15(22):5261. doi:10.3390/electronics15225261
BibTeX
@article{Tian2026,
author = {Tian, Jin and Zhang, Jing and Kim, Sanghyun},
title = {Multi-Layer Write-Time Verification of Data Provenance and Workflow Integrity for Blockchain-Based Robot Maintenance Records},
journal = {Electronics},
year = {2026},
volume = {15},
number = {22},
pages = {5261},
doi = {10.3390/electronics15225261}
}
EUVIMED