Engineering 4.0: Big data for smart cities – challenges and opportunities

Data is everywhere, constantly trickling in and out of the smartphones in our pockets, and through cheap sensors monitoring various aspects of city living.

“If we can connect these things together and make use of new opportunities like real-time monitoring, we’ll have a lot of useful information,” says Professor Ajith Abraham of the Department of Computer Science, Faculty of Engineering, Built Environment, and Information Technology (EBIT).

“Cities are growing and populations are urbanising rapidly, which can lead to high levels of pollution, traffic and crime,” Abraham says, explaining that harnessing data in a better way would enable smart technological solutions for modern cities.

That is why EBIT researchers are using computer models and diplomacy to bridge gaps between those who hold the data and those who need it, by making it easy for everyone to talk to each other.

Modern artificial intelligence (AI) uses processes inspired by Mother Nature, like evolutionary algorithms, machine learning and artificial neural networks. These mimic the way the human brain works by connecting different individual data sets, or nodes, in the same way synapses connect neuron cells in the brain. These systems learn from data to find meaningful information more quickly and efficiently than a human or a traditional computer programme could.

Professor Nelishia Pillay, AI specialist at EBIT, agrees that Big Data is how smart cities will tackle humanity’s biggest threats, from inequality, to poor health, to climate change. “Going into the Fourth Industrial Revolution, all these aspects should link with each other nicely,” she says.

Pillay says academia, the private enterprise sector, and the government at all levels from cities to national, have massive amounts of important data that would be better managed through a repository, with a dedicated governing body. Right now, however, these collections of data are disconnected and disordered.

Through her research, she is creating models to help mismatched datasets work together to solve problems using artificial intelligence. “If you have models of the problem you can show what kind of data is needed to deal with that problem, and we can start building from that.”

Pillay hopes her work will result in a tool for African cities to streamline infrastructure maintenance by combining data from sanitation, transportation and other departments, for instance.  

This necessary cross-talk between all the players involved in running a city is also why EBIT’s smart cities and big data research takes place in the context of other projects around transportation, materials and minerals beneficiation, energy, and the environment.

“We get really excited working with different companies and entities to solve complex problems, but when it comes to doing a project, the biggest bottleneck is data,” says Pillay. “The problem is getting the actual data that you need in order to use the artificial intelligence that will get the smartness out of it.”

Abraham will lead the Big Data cause in his new role as director of the Data Science Institute. “My focus for the next six months will be to build a strong interdisciplinary team within the University of Pretoria and beyond.”

When efforts and data from academia, the government, and the private enterprise sector come together to make use of artificial intelligence for smart cities, Abraham says, “we can do wonders”.

Published by Sipho Mphurpi

Copyright © University of Pretoria 2020. All rights reserved.