Advances in Solutions on Public Transportation Problems with Artificial Intelligence
Main Article Content
Abstract
The public transportation service offered in medium and large cities has been under scrutiny in recent years due to declining user numbers. This trend is commonly attributed to internal factors such as poor management or suboptimal routes leading to extended waiting times, as well as external factors like competition from informal transportation services, which have gained popularity through mobility applications, among others. This article compiles a series of studies focused on the application of artificial intelligence to tackle various issues within public transportation services, published in recent years. The goal is to determine the most suitable technique among those implemented to address these challenges. Analyzing the research presented herein allowed for the identification of public transportation-related problems encountered by researchers and the techniques that yielded superior outcomes. It is concluded that techniques like Random Forest (RF) and Long Short-Term Memory (LSTM) neural networks are better suited for resolving problems requiring geographical coordinate data, latitude, longitude, and time series.