# Exclusive queueing model including the choice of service windows

## TL;DR

This paper develops a queueing model with multiple service windows incorporating agent choice behavior, using a logit algorithm, and analyzes how floor design and parameters affect transit times and block rates.

## Contribution

It introduces a novel queueing model that includes agent choice based on distance and number of agents, with simulation analysis for optimization.

## Key findings

- Optimal floor lengths minimize transit times.
- Transit times are mainly influenced by entrance block rates.
- Choice parameters significantly affect system performance.

## Abstract

In a queueing system involving multiple service windows, choice behavior is a significant concern. This paper incorporates the choice of service windows into a queueing model with a floor represented by discrete cells. We contrived a logit-based choice algorithm for agents considering the numbers of agents and the distances to all service windows. Simulations were conducted with various parameters of agent choice preference for these two elements and for different floor configurations, including the floor length and the number of service windows. We investigated the model from the viewpoint of transit times and entrance block rates. The influences of the parameters on these factors were surveyed in detail and we determined that there are optimum floor lengths that minimize the transit times. In addition, we observed that the transit times were determined almost entirely by the entrance block rates. The results of the presented model are relevant to understanding queueing systems including the choice of service windows and can be employed to optimize facility design and floor management.