Mobile Spaghetti Analysis in Grocery Store with Smartphone Sensing

Spaghetti analysis is a survey conducted for the purpose of studying customer shopping behavior in a store. Each customer's behavior is observed visually by a hired researcher. The human cost of the survey is a significant challenge. In this study, we propose a framework for mobile spaghetti analysis that can be performed using sensor data obtained from customers' smartphones and a skeleton map, which is a map of the store interior. The performance of the proposed framework was evaluated in a rigorous laboratory environment and in an actual store to confirm its usefulness. In addition, a spaghetti analysis was conducted on more than 500 customers in an actual store using the proposed method, and from the results, it was possible to understand customer behavior patterns that were not known by the store staff without conducting a large-scale survey.