An analytical assessment of demand effects in omni-channel assortment planning

The advent of omni-channel (OC) retailing makes assortments seamlessly available across channels and affects customer behavior. Whereas the demand effects within channels are well known, the effects across channels are less clear. A vast variety of assortment-related demand effects in bricks-and-mortar stores, webshops, and across channels has a potential impact on retailers' profitability. The multitude makes it costly to measure each effect empirically, which is further aggravated by interdependencies between products, channels and effects, and the resulting numerical complexity. Despite most retailers adopting OC assortments, the relevance and interplay of these demand effects is neither fully clear from an empir-ical nor an optimization perspective. It becomes necessary to better understand how assortment-related decisions impact customer behavior and optimal assortments.We approach this research gap by identifying and integrating demand effects in a novel model for OC assortment optimization. Our results show that in-channel effects matter more than cross-channel transitions. This also holds true when demand effects exceed the empirically measured values in a single -channel context. Generalizing, the impact depends on demand rates, channel package sizes, and channel size. These findings are relevant for the OR community, empirical researchers, and retailers and help streamlining further research in two ways. First, further advances of OC assortment models with cross -channel effects should be based on our findings. Second, given that the empirical tests for these effects are very voluminous and costly, our findings serve as "guardrails" to define the scope of such empirical investigations.(c) 2022 Elsevier Ltd. All rights reserved.