Learning When to Automate: Queue Control in Human-AI Service Systems
Abstract
We study a human-AI service system in which tasks arrive sequentially and are processed through a two-stage architecture: an automated chatbot followed, when necessary, by a human agent. We consider T sequentially arriving tasks, each belonging to one of K heterogeneous types. For each task the decision maker chooses how many resources to allocate to the chatbot, whose type-dependent success probabilities are initially unknown. Tasks not resolved by the chatbot enter type-dependent human-service queues, where they are processed by a human agent with unknown service rates. This model captures a central tradeoff in hybrid service systems: relying more on automation reduces human congestion but increases chatbot costs, while insufficient automation may overload the human agent. We propose the UCB-DPP policy, which combines Upper Confidence Bounds with Drift-Plus-Penalty control to learn the unknown parameters of the system while making queue-aware decisions. We prove that UCB-DPP achieves regret 𝒪̃(K√T) and guarantees mean-rate stability of the human-service queues. Simulations on synthetic instances show that the proposed policy outperforms natural baselines.
Related
- Resilient Extremum Seeking Control for Cyber-Physical Systems Under Denial-of-Service Attacks
- Optimal Control of a Non-Stationary Inventory-Queue, with Applications to EV-Battery Swapping
- Capacity Control in a Single-Server Diagnostic--Treatment Queue with Heterogeneous Treatment Modes
- Robust Iterative Learning Control for Multi-Agent Systems with Random Impulsive Disturbances
- Learning and Control Beyond Linearity: Towards a Non-asymptotic Theory for Bilinear Systems
BibTeX
@misc{montanari2026learning,
title = {{Learning When to Automate: Queue Control in Human-AI Service Systems}},
author = {Giovanni Montanari and Marco Scarsini and Vianney Perchet},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2607.06017},
eprint = {2607.06017},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.06017},
}
RIS
TY - UNPB
TI - Learning When to Automate: Queue Control in Human-AI Service Systems
AU - Giovanni Montanari
AU - Marco Scarsini
AU - Vianney Perchet
PY - 2026
DA - 2026/10/21
PB - arXiv
DO - 10.48550/arxiv.2607.06017
UR - https://arxiv.org/abs/2607.06017
AB - We study a human-AI service system in which tasks arrive sequentially and are processed through a two-stage architecture: an automated chatbot followed, when necessary, by a human agent. We consider T sequentially arriving tasks, each belonging to one of K heterogeneous types. For each task the decision maker chooses how many resources to allocate to the chatbot, whose type-dependent success probabilities are initially unknown. Tasks not resolved by the chatbot enter type-dependent human-service queues, where they are processed by a human agent with unknown service rates. This model captures a central tradeoff in hybrid service systems: relying more on automation reduces human congestion but increases chatbot costs, while insufficient automation may overload the human agent. We propose the UCB-DPP policy, which combines Upper Confidence Bounds with Drift-Plus-Penalty control to learn the unknown parameters of the system while making queue-aware decisions. We prove that UCB-DPP achieves regret 𝒪̃(K√T) and guarantees mean-rate stability of the human-service queues. Simulations on synthetic instances show that the proposed policy outperforms natural baselines.
ER -