DEVELOPMENT OF A DEEP LEARNING MODEL FOR FORECASTING AND OPTIMIZING RIDE-SHARING ROUTES

Authors

DOI:

https://doi.org/10.26577/jpcsit20253106

Keywords:

Trip Forecasting, Ride-Sharing, Machine Learning, Deep Learning, SimpleRNN, Long Short-Term Memory, Gated Recurrent Unit

Abstract

The study investigates the potential use of machine learning (ML) technologies, including Recurrent Neural Networks (RNNs), in ride-sharing and urban mobility optimization. Advanced deep learning (DL) models can solve growing challenges in urban areas, such as road safety, environmental pollution, and traffic congestion. Three different RNN architectures (SimpleRNN, LSTM, GRU) are compared to predict trips with their pickup and drop-off points. According to the assessment metrics, GRU shows better results in terms of Mean Haversine Distance (6.450 km) than SimpleRNN (7.156 km) and LSTM (6.569 km). Moreover, the GRU model surpasses other models in other indicators, such as MSE (0.0010) and MAE (0.0211). In addition, OSRM API is used to build routes between predicted pickup and drop-off points, as well as to optimize ride-sharing routes using real-time geographic data. The study highlights that ML approaches, in particular DL, can be used to solve problems related to urban mobility by improving transport efficiency and reducing traffic. The study results provide recommendations for developing urban transport systems using data-driven approaches to enhance ride-sharing opportunities.

Downloads

Download data is not yet available.

Author Biographies

Nurbolat Amilbek, Astana IT University, Astana, Kazakhstan

Nurbolat Amilbek is a promising junior scientist and graduate student at Astana IT University, specializing in data science, machine learning, and artificial intelligence. As a dedicated researcher, Nurbolat is actively involved in various projects that apply advanced technologies to solve real-world challenges. He is currently pursuing his graduate studies, where his academic focus includes intelligent systems, predictive modeling, and data-driven decision-making. Despite being early in his academic career, Nurbolat has already demonstrated a strong commitment to advancing research in smart technologies and their application to urban development. Through his work at Astana IT University, he aspires to contribute to the future of smart cities and innovative solutions for urban infrastructure.

Beibut Amirgaliyev, Astana IT University, Astana, Kazakhstan

Beibut Amirgaliyev is a distinguished researcher at Astana IT University and is recognized for his contributions to academia and industry. He holds a PhD in Computer Science and serves as a Professor at Astana IT University, focusing on research areas such as machine learning and computer vision. Dr. Amirgaliyev has published numerous papers on automatic number plate recognition and solar collector systems, with his work cited by over 200 researchers.

Didar Yedilkhan, Astana IT University, Astana, Kazakhstan

Didar Yedilkhan is a distinguished researcher at Astana IT University, recognized for his extensive experience in industry, research, and higher education. He serves as the Director of the Smart City Research Center and is a Senior Researcher at Astana IT University, focusing on data science, machine learning, and deep learning. Dr. Yedilkhan has a robust academic background, holding degrees from institutions such as the Kazakh National University named after al-Farabi and the University College London. His professional roles include being a Lead Researcher and Project Manager at Astana IT University, where he leads projects on intelligent IT systems for urban infrastructure. His projects aim to enhance city safety and convenience through smart technologies.

Zharasbek Baishemirov, Astana IT University, Astana, Kazakhstan

Zharasbek Baishemirov is a lead researcher at Astana IT University and Kazakh British Technical University, recognized for his extensive experience in the fields of industry, research, and higher education. He holds a key role as a Lead Researcher at Astana IT University, where his work primarily focuses on mathematical modeling, mathematics, machine learning, and artificial intelligence. Dr. Baishemirov has a strong academic foundation, having completed his studies at leading institutions such as the Abai Kazakh National Pedagogical University and other renowned universities.

        2 0

Downloads

How to Cite

Amilbek, N., Amirgaliyev, B., Yedilkhan, D., & Baishemirov, Z. (2025). DEVELOPMENT OF A DEEP LEARNING MODEL FOR FORECASTING AND OPTIMIZING RIDE-SHARING ROUTES. Journal of Problems in Computer Science and Information Technologies, 3(1), 56–71. https://doi.org/10.26577/jpcsit20253106

Issue

Vol. 3 No. 1 (2025): Journal of problems in computer science and information technologies

Section

Articles