ML model detects elevator installation errors

ChangeBridge: Patent Grants - AI & Computing (G06N)

Published March 31st, 2026

Detected March 31st, 2026

Summary

The USPTO granted patent US12589974B2 to Wittur Holding GmbH covering a computer-implemented method for training a machine learning model to detect installation errors in elevators. The ML model combines Set Function and Fourier-Transform models with sensor data to identify elevator door and installation issues. The patent was filed on June 17, 2022, with 20 claims.

View original document View source feed page

What changed

Wittur Holding GmbH has been granted US Patent 12,589,974 covering a computer-implemented method for training a machine learning model to detect installation errors in elevators, specifically elevator doors. The method uses a hybrid ML architecture combining Set Function and Fourier-Transform models, processing physical parameter data from multiple sensors arranged on the elevator. The system extracts features from sensor datasets containing time series data through two separate input layers before feeding them to the respective models.

Patent holders and technology developers working with elevator systems or predictive maintenance should review the 20 granted claims to assess potential freedom-to-operate implications or licensing opportunities. There are no compliance deadlines or regulatory actions associated with this patent grant—it is an intellectual property notification. Companies developing similar ML-based elevator monitoring systems should consider the patent's scope when designing product architectures.

Source document (simplified)

← USPTO Patent Grants

Computer-implemented method for training a machine learning model to detect installation errors in an elevator, in particular an elevator door, a computer-implemented method for classifying installation errors and a system thereof

Grant US12589974B2 Kind: B2 Mar 31, 2026

Assignee

WITTUR HOLDING GmbH

Inventors

Martin Zellhofer, Giuseppe De Francesco

Abstract

A computer-implemented method for training a machine learning model to detect installation errors in an elevator, in particular an elevator door. The machine learning model being a combination of a Set Function model and a Fourier-Transform model. The method including arranging a plurality of sensors at the elevator and each sensor being configured to detect a physical parameter. The method also includes detecting values of the physical parameters by the sensors so as to obtain a dataset comprising at least one time series and obtaining a first input layer by extracting features from the dataset. The method also includes obtaining a second input layer by extracting features from the dataset, feeding the Set Function model with the first input layer, and feeding the Fourier-Transform model with the second input layer.