Optimal staffing policy for queueing systems with cyclic demands

Although the profit loss from business resulting from queueing systems is quite difficult to estimate, this paper presents a creative and effective approach to formulating balking loss as waiting cost. The number of balking customers is estimated as the balking index θ multiplied by the expected queue length Lq and the mean arrival rate in that planning period λ. Estimating balking loss enables the decision maker to determine the optimal number of servers by minimising total cost, including service cost and balking loss. The practicality and the incurred benefits of this approach were analysed by using collected data from a fast food restaurant, using the SIPP M/M/S:(FCFS/∞/∞) queueing model. The results showed that this approach is an effective tool for aiding managers in formulating the optimal staffing policies of service systems.