Towards Optimal Inventory Control under Censored Demand: A Biased Sample-Average Approximation Approach
Abstract
We study data-driven multi-period lost-sales inventory control under censored demand, where a stockout reveals only that demand exceeded the stocking level. We develop a unified, model-based framework for policy learning from censored data, built on a new cost decomposition for base-stock policies and a biased sample-average approximation (SAA) approach. The cost decomposition allows us to propose a new coverage condition under which censored observations are informative enough for sample-efficient policy learning. Guided by this coverage condition, we design two biased SAA algorithms: an upper-biased one that achieves near-optimal sample complexity under the offline coverage condition, and a lower-biased one that actively generates the required coverage and achieves near-optimal regret online. More broadly, this biased SAA approach provides a general principle for implementing pessimism and optimism under censored feedback, which may be of independent interest.
Related
- Improved Approximation Guarantees for Periodic-Review Stochastic Inventory Control
- Optimal Control of a Non-Stationary Inventory-Queue, with Applications to EV-Battery Swapping
- The BAR-SOT Method: Long-term Average Cost Control as Stochastic Optimal Self-Transport
- Numerical approximation of distributed optimal control problems governed by linear thermo-poroelasticity
- Optimal subsampling approach for Cox models under large-scale length-biased right-censored data
BibTeX
@misc{han2026towards,
title = {{Towards Optimal Inventory Control under Censored Demand: A Biased Sample-Average Approximation Approach}},
author = {Yuxuan Han and Xiaoyu Fan and Jiawei Zhang and Zhengyuan Zhou},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.39397},
eprint = {2609.39397},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.39397},
}
RIS
TY - UNPB
TI - Towards Optimal Inventory Control under Censored Demand: A Biased Sample-Average Approximation Approach
AU - Yuxuan Han
AU - Xiaoyu Fan
AU - Jiawei Zhang
AU - Zhengyuan Zhou
PY - 2026
DA - 2026/09/30
PB - arXiv
DO - 10.48550/arxiv.2609.39397
UR - https://arxiv.org/abs/2609.39397
AB - We study data-driven multi-period lost-sales inventory control under censored demand, where a stockout reveals only that demand exceeded the stocking level. We develop a unified, model-based framework for policy learning from censored data, built on a new cost decomposition for base-stock policies and a biased sample-average approximation (SAA) approach. The cost decomposition allows us to propose a new coverage condition under which censored observations are informative enough for sample-efficient policy learning. Guided by this coverage condition, we design two biased SAA algorithms: an upper-biased one that achieves near-optimal sample complexity under the offline coverage condition, and a lower-biased one that actively generates the required coverage and achieves near-optimal regret online. More broadly, this biased SAA approach provides a general principle for implementing pessimism and optimism under censored feedback, which may be of independent interest.
ER -