A survey of queueing systems with strategic timing of arrivals

TL;DR

This survey reviews research on queueing systems where customers strategically choose their arrival times, analyzing equilibrium behaviors, social optimality, and the impact of strategic decisions on congestion and costs.

Contribution

It provides a comprehensive overview of the main results and developments in the analysis of queueing systems with strategic timing of arrivals, including game-theoretic and social efficiency aspects.

Findings

Characterization of Nash equilibrium arrival strategies

Analysis of the price of anarchy in queueing systems

Insights into social optimality and congestion costs

Abstract

Consider a population of customers each of which needs to decide independently when to arrive to a facility that provides a service during a fixed period of time, say a day. This is a common scenario in many service systems such as a bank, lunch at a cafeteria, music concert, flight check-in and many others. High demand for service at a specific time leads to congestion that comes at a cost, e.g., for waiting, earliness or tardiness. Queueing Theory provides tools for the analysis of the waiting times and associated costs. If customers have the option of deciding when to join the queue, they will face a decision dilemma of when to arrive. The level of congestion one suffers from depends on others behavior and not only that of the individual under consideration. This fact leads customers to make strategic decisions regarding their time of arrival. In addition, multiple decision makers that affect each other's expected congestion, call for non-cooperative game theoretic analysis of this strategic interaction. This common daily scenario has prompted a research stream pioneered by the ?/M/1 model of Glazer and Hassin (GH1983) that first characterized an arrival process to a queue as a Nash equilibrium solution of a game. This survey provides an overview of the main results and developments in the literature on queueing systems with strategic timing of arrivals. Another issue is that of social optimality, namely the strategy profile used by customers that optimizes their aggregate utility. In particular, we review results concerning the price of anarchy (PoA), which is the ratio between the socially optimal and the equilibrium utilities.