A large part of the population does not have access to Emergency Departments or, when they do, face a crowded environment, increasing wait time for the service without their risk situation being assessed. The Manchester Triage System was developed to identify the degree of priority of patients who come to the Emergency Department and to improve the quality of care in emergency services, redefining the flow of care by prioritizing patients who are in the most serious conditions. This work aims to make a comparison between six classifiers, based on the Manchester Triage System, with the data present during patient intake. The purpose is that the model can correctly classify their priority in emergency care. The experiments were conducted with a pediatric emergency database from hospitals in The Netherlands, Portugal and the United Kingdom. With the results obtained by the classifiers' performance, the best performing model was the Random Forest, with 78.20% for accuracy and 78.60% for F1-score. The expectation is that, by automating the classification process, health professionals will have a reliable tool to conduct risk assessment more accurately, having as a side-effect, less crowded Emergency Departments and reducing patient health deterioration due to misclassification and waiting time.