Ukg Inc.

Alexander L Krowitz

Brian R Muras

Keith A Briggs

David Henry Wayne

Martin Tapp

A computer-implemented method of training machine learning models for making simulated estimations is provided. The method includes collecting, from a database, a set of historical data, applying one or more transformations to the set of historical data to create a set of model features, and separating the set of model features into one or more pools, each pool comprising one or more model features of the set that are homogeneous with respect to a common value. the method further includes, for each pool, dynamically creating a training set that includes the one or more sets of model features of the pool and at least some of the historical data. The method further includes, for each training set, training a machine learning model on the training set.

Method of training machine learning models for making simulated estimations

An apparatus for extended horizon scheduling that includes a time-sequence interpretation circuit, a target interpretation circuit, a horizon target analysis circuit, and a promotive action provisioning circuit. The time-sequence interpretation circuit interprets time-sequence data, the target interpretation circuit interprets target data, and the time-sequence trend analysis circuit extracts a trend from the time-sequence data. The horizon target analysis circuit: determines whether the extracted trend furthers or impedes a target defined, in part, by the target data, and responsive to a determination that the extracted trend impedes the target, generates a promotive action command value structured to trigger an adjustment to the time-sequence data. The adjustment is structured to mitigate the extracted trend from impeding the target. The promotive action provisioning circuit transmits the promotive action command value.

Systems and methods for extended horizon time-sequences

Networks, methods, and apparatus to ensure schedule conformance are described. An example apparatus may include a schedule interpretation circuit to interpret schedule data, and a warden circuit to generate scores for the schedule data with respect to schedule properties. The warden circuit retrieves baseline values corresponding to schedule properties and determines whether the schedule data is out of alignment with the baseline value for any of the schedule properties. If so, a corrective action circuit generates a corrective action command value to trigger an adjustment to the schedule data to effect a change to a schedule property that is out of alignment. A corrective action provisioning circuit transmits the corrective action command value.

Growing community of inventors

Biddeford, ME, United States of America

Alexander L Krowitz