Evaluating Staff Performance: A Markov Chain Approach

The purpose of this paper is to apply and validate an application of Markov chain models to measure the effects of different staffing levels on group performance whilst including the effects of absenteeism. Two models were formulated, one that models absenteeism in detail, and another that uses a simplified approach. Experiments and Monte Carlo simulations were conducted to confirm the validity of the models. Both Markov chain models provide results that fit within the standard error of the experiments. Limitations of this research are: (1) only one piece of work arrives at a time; (2) the arrival rate of work is a constant; (3) work needs to be executed on arrival, otherwise it leaves the system immediately and the group loses that piece of work; and (4) only one worker begins an absence at any time. This research adds to the literature on organisation management in two distinct ways. First, it shows that a Markov chain model is able to accurately include the effects of absenteeism on the relationship between staffing levels and performance of a group. Second, the Markov chain model can be simplified without loss of accuracy.