Data-driven decisions in employee compensation utilizing a neuro-fuzzy inference system

Abstract

Artificial intelligence assists organizations to carry out strategic management decisions especially in talent management. A firm’s overall compensation management is defined by its pay philosophy and process that has been a key component in employee engagement and satisfaction that also correlates with firm success. This neuro-fuzzy inference system was able to design an objective compensation algorithm that objectively identified relevant variables for qualified applicants in the hiring and selection stage that will be the baseline of an employee’s initial salary. The output is a salary grade matrix that allows adjustment discretion according to the standards of the HR department who may have preference to either one of the variables. This will now simultaneously function as an operational framework in the performance management stage for current employees and serve as a benchmark during annual salary reviews. An artificial neural network employed all parameters in the categorical traits in the performance evaluation of employees that targets errors that are not normally detected in the traditional review method that is subjected to preferential bias, favoritism or irregularities. The ANN structure output produced 5 numerical decisions to upgrade, maintain and downgrade the salary grade that will coincide with both organizational objectives and HR compensation policies.