Linking Automated Inspection Reliability to Digital Maintenance Optimization: Synergies Between Cutting-Edge Research and the AMADIT Subproject
The transition toward the digitalization of critical industrial infrastructures demands advanced tools that go beyond estimating raw technological performance to actively validating economic and operational viability in real-world environments. In this exact direction, a recent study developed by researchers from the Department of Industrial Management at the University of Seville alongside Patentes Talgo, published in Reliability Engineering and System Safety, introduces a pioneering framework to derive reliability requirements explicitly conditioned by fallback consequences.
This study directly aligns with the strategic objectives of the AMADIT subproject («Asset Management in the New Digital Twin Environment»), which is an integral part of the national DIGEST project. While the overarching DIGEST initiative promotes the development of comprehensive digital twins and optimization models for industrial asset management, the AMADIT subproject specifically focuses on providing these digital ecosystems with a robust analytical layer to support predictive maintenance and health-management decisions.
The Core of the Research: Quantifying the Cost of Unavailability
The research addresses a critical vulnerability in modern automated workshop environments: the operational impact of a system failure that forces maintenance routines to temporarily revert to traditional, manual procedures (workshop fallback). When automated inspection infrastructures experience downtime, manual workloads inside the workshop spike, asset availability decreases, and unexpected economic losses accumulate.
The proposed framework departs from traditional, heuristic reliability allocation methods. Instead, it establishes an economic admissibility constraint based on a clear break-even condition between the annualized net benefit of automation and the expected financial losses driven by fallback events. Utilizing an auditable back-propagation procedure (cast as constrained lifetime tuning), the model translates high-level operational targets into verifiable, element-level Mean Time Between Failures (MTBF) requirements for procurement and acceptance testing.
A Natural Fit for the DIGEST Architecture
This consequence-conditioned methodology strengthens the core concept of an industrial Digital Twin. A digital twin cannot simply assume that sensing infrastructures operate with perfect availability; it must factor in the reliability of the inspection assets that feed it data. If data acquisition loops are compromised by sensor failures, the optimized predictive maintenance pathways designed within the DIGEST project risk being bottlenecked by inaccurate or missing inputs.
By tying reliability engineering parameters directly to workshop capacity penalties and economic exposure, this methodology provides operators and technology buyers with a clear framework for contract negotiation, risk management, and the validation of digital maintenance investments.
.