In manpower planning mathemathical models, based on estimations of transition probabilities between groups of employees, are used to predict future staff compositions. Although the quality of the model depends on the division of the staff in groups, this classification has been neglected in literature. The present paper investigates whether decision tree learning can be used as a classi- fication technique in manpower planning. The paper presents a method for dividing the population according to the available data in the HR database. The approach will improve predictions and validity of the model. Another advantage of our developed method is that it can be automated and implemented in software. Implementation of the method will make it possible for HR departments in a company to use the models in practice. The approach will be illustrated on a real life human resources database using statistical software such as R and WEKA
Original languageEnglish
Title of host publicationThe 16th Conference of the Applied Stochastic Models and Data Analysis International Society
Publication statusPublished - 2015
EventApplied Stochastic Models and Data Analysis - ASMDA2015 - University of Piraeus, Piraeus, Greece
Duration: 30 Jun 20154 Jul 2015


ConferenceApplied Stochastic Models and Data Analysis - ASMDA2015

ID: 5288677