Real-Time Bus Departure Prediction Using Neural Networks for Smart IoT Public Bus Transit

TL;DR

A neural network-based approach for real-time bus departure time prediction tailored for smart IoT public transit applications that improves the reliability of bus transit schedules and can facilitate the integration of IoT devices, such as smart bus stops and passenger information systems, that rely on precise data for optimal performance.

Abstract

Bus transit plays a vital role in urban public transportation but often struggles to provide accurate and reliable departure times. This leads to delays, passenger dissatisfaction, and decreased ridership, particularly in transit-dependent areas. A major challenge lies in the discrepancy between actual and scheduled bus departure times, which disrupts timetables and impacts overall operational efficiency. To address these challenges, this paper presents a neural network-based approach for real-time bus departure time prediction tailored for smart IoT public transit applications. We leverage AI-driven models to enhance the accuracy of bus schedules by preprocessing data, engineering relevant features, and implementing a fully connected neural network that utilizes historical departure data to predict departure times at subsequent stops. In our case study analyzing bus data from Boston, we observed an average deviation of nearly 4 minutes from scheduled times. However, our model, evaluated across 151 bus routes, demonstrates a significant improvement, predicting departure time deviations with an accuracy of under 80 seconds. This advancement not only improves the reliability of bus transit schedules but also plays a crucial role in enabling smart bus systems and IoT applications within public transit networks. By providing more accurate real-time predictions, our approach can facilitate the integration of IoT devices, such as smart bus stops and passenger information systems, that rely on precise data for optimal performance.

Figures (9)

• Figure 1: MBTA monthly bus ridership since January 2020, showing the current decline of 61.75% in ridership, compared to pre-pandemic levels avisonyoung2024.
• Figure 2: Comprehensive IoT-enabled smart bus system: This diagram illustrates a public transit enhancement framework with three main components: data processing and dynamic scheduling (managed by a centralized cloud platform that processes real-time transit data and employs an AI-powered prediction model to adjust bus schedules dynamically), IoT-enabled buses (which continuously exchange data with the cloud to provide up-to-date information and improved service quality), and a mobile app for passengers to access the bus departure time information.
• Figure 3: An overview of the bus departure time prediction system: The system integrates three key data sources (transit operations data, meteorological data, and bus stop data). The preprocessing stage involves removing invalid entries and detecting outliers. Feature extraction generates both input and output features for the model. An FCNN with a one-bus stop lookback window is trained using the MSE loss function. In the prediction stage, the system provides real-time estimates of bus departure time deviations.
• Figure 4: A visualization of the geographical locations of 1,111 bus stops from the MBTA Bus Departure Times 2023 dataset. The map illustrates the distribution of these bus stops throughout the city of Boston.
• Figure 5: The distribution of trips across the routes over the three months ranges from a maximum of 290,545 trips to a minimum of 568 trips.