Artificial Neural Network And Their Application In The Prediction Of Absenteeism At Work

Research Article
Ricardo Pinto Ferreira., Andréa Martiniano., Domingos Napolitano., Edquel Bueno Prado Farias and Renato José Sassi
Artificial Neural Network; Prediction; Absenteeism; Rough Sets.

The high competitiveness in the market, professional development combined with the development of organizations and the pressure to reach increasingly audacious goals, create increasingly overburdened employees and end up acquiring some disturbance in the state of health related to the type of work activity, including depression considered the evil of the 21st century. Taking employees to absenteeism. Absenteeism is defined as absence to work as expected, represents for the company the loss of productivity and quality of work. The purpose of this paper was to apply an artificial neural networks to prediction of absenteeism at work. The database used in the experiment has 38 attributes and 2,243 records from documents that prove that they are absent from work and was collected from January 2008 to December 2016. The methodological synthesis of the paper consists of the modeling of an Artificial Neural Network (ANN), the 38 attributes were reduced to 17 attributes through the Rough Sets, these attributes were used in the experiments to prediction of absenteeism. ANN they are models consisting of simple processing units, called artificial neurons, these models are inspired by the structure of the brain and aim to simulate human behavior, such as learning, association, generalization and abstraction when submitted to training. The experiments with the ANN presented the expected results in prediction of absenteeism at work. Therefore, it is concluded that the ANN can be applied in the prediction of absenteeism at work and in other problems similar to that presented in this paper.