A hybrid generative and transformer-based framework for anomaly detection in industrial sensor time-series for predictive maintenance

Nayab Asim, Irum Matloob, Zaid Khan, Rukaiya Rukaiya, Shoab Khan, Hessa Alfraihi · 2026

Read A hybrid generative and transformer-based framework for anomaly detection in industrial sensor time-series for predictive maintenance by Nayab Asim, Irum Matloob, Zaid Khan, Rukaiya Rukaiya, Shoab Khan, Hessa Alfraihi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13575108.txt.

Description

Nayab Asim and five colleagues build a hybrid system for spotting faults in factory sensor data in a 2026 paper. Predictive maintenance needs early warning of faults, yet fault examples are scarce and factory conditions keep changing. Their design mixes a generative model, transformer blocks that read time patterns, and classic statistical checks.

Tests ran on two public benchmark sets and on real current data from cranes with no labels. On the machine failure set the system reached 84.09 percent accuracy, and on the water treatment set it reached 92.25 percent accuracy. For the crane data, a weighted fusion of methods agreed on 14 high-confidence faults. The authors argue the combined design beats any single method alone.

Catalog record

CREATORNayab Asim, Irum Matloob, Zaid Khan, Rukaiya Rukaiya, Shoab Khan, Hessa Alfraihi
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE93 KB
LICENSEhttps://creativecommons.org/licenses/by/4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/article/MED/PMC13575108

Browse the full library · About the Tower of Babel