From Lagos traffic to Global labs, Ogungbire is rethinking transport with artificial intelligence

In an era when traffic congestion, safety and climate concerns are converging on the world’s roads, a new generation of transportation researchers is turning to artificial intelligence for answers. One of them is Nigerian engineer Adedolapo Ogungbire, whose work focuses on applying advanced machine learning to real world transportation systems.

Ogungbire recently received the Tau Beta Pi Fellowship, a highly competitive award given by Tau Beta Pi, the oldest engineering honor society in the United States and the only one that represents the entire engineering profession. Each year, Tau Beta Pi awards about 30 fellowships of 10,000 dollars to support graduate engineers worldwide, a small number compared to the hundreds of thousands of members the society has initiated over its history. The fellowship programme, started in 1928, is widely described as one of the pioneering graduate award schemes among honor societies.

For Ogungbire, the fellowship is not only a personal milestone. It is also a signal that research emerging from African scholars is increasingly visible in global conversations about the future of mobility.

In this interview with THISDAY, he speaks about the award, his path into transportation research, and how his work on artificial intelligence, data and policy aims to shape safer and more efficient cities.

Congratulations on your Tau Beta Pi Fellowship. What does this recognition mean to you as a Nigerian transportation researcher?
Thank you. It means a lot on several levels. Tau Beta Pi is not a general scholarship program. It is an engineering honor society that selects its members from the top of their fields and then supports a small group of fellows each year based on academic excellence, research excellence and service.

So, when I saw my name on the fellowship list, I felt two things. First, gratitude that the work I am doing in transportation and artificial intelligence resonated with engineers outside my immediate circle. Second, a sense of responsibility. Coming from Nigeria, I know how many talented students never get global visibility. For me, this fellowship is a reminder that I have to use the opportunity well and keep the door open for others.

How did your journey into transportation and artificial intelligence begin? Was it something you always planned?

Growing up in Nigeria, you cannot ignore transportation. You feel it every day, in the time you spend in traffic, in the unpredictability of travel, in safety concerns when infrastructure fails. I did not have the language of “transportation systems” at that time, but I understood that how people and goods move shapes almost everything else.

In university, I was drawn to the analytical side of engineering. I enjoyed mathematical modeling and programming, but I also wanted to work on problems that touch daily life. Transportation engineering became the perfect intersection. Artificial intelligence came later, when I saw how data from phones, public cameras, sensors and vehicles could be used to understand and improve these systems in ways that traditional tools could not.

For a general reader, how would you explain the heart of your research in AI and transportation?

At a simple level, my research asks three questions.

First, how can we predict what will happen in a transportation network, whether it is traffic on a corridor, demand for buses, or patterns of crashes.

Second, how can we optimize decisions such as routing, scheduling and control in real time, using those predictions.
Third, how can we do this in a way that is fair and robust, so that the benefits of smart systems do not only go to people who already have the most resources.

Artificial intelligence and machine learning give us tools to learn patterns from large, messy datasets. I build models that can take information from multiple sources, learn how the system behaves, and then propose decisions that improve performance. The goal is not a fancy algorithm for its own sake. The goal is a safer, more reliable and more sustainable transportation system.

Many people use AI in transportation now. In your view, what is original about your contributions?

You are right that AI in transportation is becoming more common. What I try to do is tackle some of the gaps that appear when we move from theory to practice.

One part of my work focuses on robust methods adaptable to data scarcity and data quality in low- and middle-income settings. Many states of the art AI models assume rich, continuous data streams. City agencies in Nigeria or other African countries often have fragmented data at best. I have developed approaches that can still learn useful patterns from partial or noisy data and that explicitly quantify uncertainty, so decision makers know how much confidence to place in a prediction.
Also, I try to bridge research and implementation. It is not enough for a method to work in a paper. I work closely with practitioners so that the models I develop can be embedded in the tools they actually use and tested against their operational constraints.

Can you give an example of a project where you saw your research make an impact beyond the academic paper?

One example is my work on using machine learning to support planning and scheduling engineers for transportation agencies. States and Cities, across the world, and even more pronounced here in Africa, struggle with adequate transportation engineering strength to ensure project quality. The work was implemented and developed into a software tool used by a state transportation agency in the United States. It has also been used by a Lagos based startup within the framework of their SaaS tool.

The impact was not only in marginal improvements in efficiency. It changed the conversation. Managers could efficiently assign their workforce across cities and districts. Such impact, particularly when adopted by African agencies are important to me.