A mean–variance model to optimize the fixed versus open appointment percentages in open access scheduling systems

Although healthcare quality may improve with short-notice scheduling and subsequently higher patient show-up rates, the variability in patient flow may negatively impact the service design. This study demonstrates how to select the percentage for short-notice or open appointments in an open access scheduling system subject to two quality performance metrics. Specifically, we develop a mean–variance model and an efficient solution procedure to help clinic administrators determine the open appointment percentage subject to increasing the average number of patients seen while also reducing the variability. Our numerical results indicate that for cases with high patient demand and high patient no-show rates for fixed appointments, one or more Pareto optimal percentages of open appointments significantly decrease the variability in the number of patients seen with only a negligible decrease in the expected number of patients seen. While our method provides a useful tool for clinic administrators, it also presents a modeling foundation for open access scheduling with quality management objectives to smooth patient flow and improve capacity utilization.

► We develop a decision tool for the scheduling systems of open access clinics. ► It includes a mean–variance model and the procedure to obtain Pareto solutions. ► The model optimizes both the expected number of patients seen and its variance. ► Pareto solutions provide better options than solutions only maximizing the mean.