Simulation-based estimation of cycle time using quantile regression

Production cycle time is an important performance measure in manufacturing systems, and thus it is of interest to characterize distributional properties, such as quantiles, for informative decision making. This article proposes a non-linear quantile regression model for the relationship between stationary cycle time quantiles and corresponding throughput rates of a manufacturing system. The statistical properties of the estimated cycle time quantiles are investigated and the impact of dependent data from simulation output on parameter estimations is analyzed. Extensive numerical studies are presented to demonstrate the effectiveness of the proposed methods.