Abstract
Long waiting lines at polling stations have been a persistent issue on election days, often worsened by inadequate planning at consolidated polling locations. This paper reviews seven scientific methods from the literature that determine the minimum required voting resources at polling stations, analyzing their methodologies, key assumptions, practicality, and limitations. It also demonstrates that applying these methods could have prevented the excessive waiting times observed during the 2016 presidential primary election in Arizona. Queueing-formula-based methods are favored in data-limited scenarios due to their rigor, transparency, and reasonable approximation of real voting queues, while simulation-based methods can provide more precise queue performance estimates when sufficient data is available. Enhanced communication and training are needed to encourage election officials to adopt these methods.