Restaurant Queuing Time Prediction Using Random Forest Regression

Unknown queuing time often brings negative consumer experience in restaurant service. Prediction of queuing time has been a critical task to benefit both consumers and restaurants. However, current research mainly focuses on the queuing theory, which requires collecting prior knowledge and survey. This paper proposes a queuing time prediction model based on Random Forest Regression with features selection method to ease the limitations of previous research. Random Forest Algorithm have achieved enormous success in various fields. Experiments on a real-life restaurant queuing database validate the practicality of the model. The proposed method outperforms the baseline model on the experimental data. We conclude the intrinsic information of the queuing model with the feature selection method, which will help improve other application models in queuing scenarios.