{"id":"arxiv/2602.00351","url":"https://arxiv.org/abs/2602.00351","tags":["math.OC","math.PR"],"score":0,"title":"Efficiency-Reward Trade-Off in Queues with Dynamic Arrivals","domain":"math.OC","source":"arxiv","authors":["Tianze Qu","Sushil Mahavir Varma"],"reviews":0,"subject":"math.OC","summary":"Motivated by applications in online marketplaces such as ride-hailing platforms and payment channel networks, we study a single server queue with service rate $μ$ and queue-length-dependent arrival rate $λ(q) \\in [0, λ_{\\max}]$. The system operator receives a reward $F(λ)$ by setting an arrival rate $λ$, where $F$ can represent objectives such as throughput, revenue, or social welfare. The goal is to characterize the Pareto frontier between long-run average reward and expected queue length ($\\mathbb{E}[q]$). We first establish a fluid upper bound on the reward achievable under any control…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-01-30","reactions":{}}