Bayesian Network-based Analysis of the Causes of Crowding in College Cafeterias

Abstract

In order to address the crowding problem in college cafeterias, this paper classifies and summarizes the factors that may cause the crowding in college cafeterias by reviewing a large amount of related literature, establishes a fault tree model of the crowding in college cafeterias on this basis, and transforms it into a Bayesian network to construct a Bayesian network model of the crowding in college cafeterias. Then, we collect the case data of students' opinions about the cafeteria crowding event in Dalian Maritime University by conducting a questionnaire survey and normalizing them, and use the two-way inference function of Bayesian network to calculate the influence degree of each influencing factor on the cafeteria crowding event in colleges and the importance degree of each influencing factor when the cafeteria crowding event occurs. Finally, we find that the short meal service time, few cafeteria stores, insufficient cafeteria staff, low level of informatization and serious seat occupation problem are the five key factors that are most likely to lead to the occurrence of cafeteria crowding events.