中航大供应链研究团队在IEEE Transactions on Engineering Management发表重要研究成果

1. Qingguo Bai

School of Management, Qufu Normal University, Rizhao 276826, PR China

2．Lei Xu

Economics and Management College, Civil Aviation University of China, Tianjin 300300, PR China.

随着线上下单门店提货（BOPS）服务的推出，当制造商的运营过程中存在过度自信行为、碳减排要求以及不完全需求信息时，实体店是否应拥有自营在线渠道尚不明确。本文的研究对象为一个受限额与抵消政策约束的过度自信制造商，其通过全渠道零售模式销售产品。在该模式中，顾客被分为纯线上顾客、纯门店顾客和BOPS顾客，三类顾客仅能获取随机需求的期望与方差信息。与生产、库存和运输相关的运营环节均会产生碳排放。考虑自营在线渠道和独立渠道两种模式，并分别建立相应的分布鲁棒报童模型，求解出每种模型的最优鲁棒生产量的解析表达式。本文比较了过度自信制造商在自营在线系统和独立系统中的运营效能。结果表明，向线上顾客销售每单位产品的净利润是比较两个全渠道零售系统中过度自信制造商利润与碳排放的关键因素。本文还进行了数值实验以补充理论模型，并考察了全渠道零售模型中若干主要参数对两个系统的影响。本文为制造商在全渠道零售模式下制定鲁棒的生产策略以及实现线上与线下渠道的整合提供了有价值的见解。

Omnichannel retailing, operations management, overconfidence, carbon abatement, robust analysis.

The proliferation of omnichannel retailing, particularly the adoption of Buy-Online-and-Pick-up-in-Store (BOPS) services, presents manufacturers with complex operational decisions. This complexity is compounded by the prevalent behavioral bias of overconfidence, stringent environmental regulations like the cap-and-offset policy, and the inherent demand uncertainty associated with short life-cycle products. Critically, for such products, manufacturers often possess only partial demand information, characterized solely by its mean and variance, rather than the full distribution. Existing literature on overconfident newsvendors predominantly assumes complete demand information, creating a significant research gap. Therefore, a pressing need exists to investigate robust production strategies for an overconfident manufacturer operating under a cap-and-offset policy within an omnichannel framework, where strategic channel integration decisions must be made amidst severe demand information incompleteness.

A distributionally robust newsvendor model is developed to capture the decision-making process of an overconfident manufacturer under a cap-and-offset policy. The model considers two omnichannel systems: a self-operated online channel and an independent channel system. In both systems, the manufacturer faces stochastic demand with known mean and variance but unknown full distribution. Carbon emissions occur during production, inventory holding, and online transportation. The objective is to determine the optimal manufacturing quantity that maximizes profit while complying with carbon emission limits. The analysis derives closed-form solutions for robust production quantities and examines the impact of overconfidence, demand variability, carbon tax, and quota on operational decisions.

The findings reveal that overconfidence can lead to more conservative production decisions, which in turn reduce both excess inventory and carbon emissions. Contrary to traditional belief, overconfident manufacturers may achieve higher profits and lower emissions than unbiased ones when only partial demand information is available. The self-operated online channel system tends to yield higher profits but also higher emissions, whereas the independent system is more carbon-efficient. The cap-and-offset policy is effective only when the carbon quota is binding; otherwise, it has limited impact on production decisions. Managers should adjust production quantities based on their confidence level and the stringency of emission regulations.

This study provides a robust optimization framework for overconfident manufacturers operating under omnichannel retail environments with carbon emission constraints. It highlights the interplay between behavioral biases, limited demand information, and environmental policies. The results offer practical guidance for manufacturers to balance economic and environmental objectives through appropriate channel selection and production planning. Future research could extend the model to include supply chain coordination, consumer environmental awareness, and other omnichannel strategies such as buy-online-return-to-store (BORS).