Do restaurant customers and owners differ in their preferences for serving robots? A choice experiment and latent class approach

Serving robots are increasingly being adopted by restaurants to address labor shortages and improve operational efficiency. These robots offer a range of benefits to restaurant owners, including increased productivity and a novel customer experience. However, there are barriers to adopting these technologies, including the digital literacy gap, emotional resistance, and social concerns regarding the impact of automation. Nevertheless, most existing research has focused on customer acceptance of serving robots, overlooking the perspectives of restaurant owners, who are the actual decision makers driving their adoption. This study addresses this gap by analyzing the heterogeneous preferences of customers and restaurant owners regarding serving robot types. Using a discrete choice experiment (DCE) and latent class analysis (LCA), this study identified two segments in each group. Customers were technology-progressive (43.6 %) or technology-conservative (56.4 %); owners were interaction-oriented (60.3 %) or appearance-oriented (39.7 %). The results showed that customers generally prefer non-humanoid robots, whereas owners favor fixed-type or humanoid robots. Notably, customers were the least likely to accept humanoid robots, indicating a potential “uncanny valley” effect. These divergent preferences suggest that successfully introducing serving robots to restaurants requires a strategic alignment between customer experience and owner priorities. By comparing these two stakeholder groups, this study provides practical insights for restaurant owners and robot manufacturers seeking to facilitate the effective introduction of serving robots in customer-facing environments.