Abstract: Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: parameterizing with training data one or more model that expresses, for a set of historical activities, a known outcome as a function of a plurality of factors, the plurality of factors including one or more activity factor and a worker factor; solving the one or more model to learn by machine learning values of the worker random effect vector associated to the set of historical activities, wherein the values of the worker random effect vector provide an indication of learned worker skill level for respective workers identified by the worker identifiers; and providing, by the one or more processor, one or more output based on the solving.

Abstract: Methods, computer program products, and systems are presented. The method computer program products, and systems can include, for instance: parameterizing with training data one or more model that expresses, for a set of historical activities, a known outcome as a function of a plurality of factors, the plurality of factors including one or more activity factor and a worker factor; solving the one or more model to learn by machine learning values of the worker random effect vector associated to the set of historical activities, wherein the values of the worker random effect vector provide an indication of learned worker skill level for respective workers identified by the worker identifiers; and providing, by the one or more processor, one or more output based on the solving.