Opinion: Can AI prevent buildings from collapsing, asks UJ’s Prof Tshilidzi Marwala


Opinion: Can AI prevent buildings from collapsing, asks UJ’s Prof Tshilidzi Marwala


Publishing Date: 5/28/2018 9:00 AM

Many African people feel alienated from science because they cannot see themselves reflected in modern science. It is therefore important to bring forward concepts, which have been part of our culture, into the mainstream science to de-alienate ourselves from science, writes Professor Tshilidzi Marwala.

The Vice-Chancellor and Principal of the University of Johannesburg (UJ) as well as the author of the books Conditioning Monitoring Using Computational Intelligence and Artificial Intelligence for Rational Decision MakingProf Marwala recently penned an opinion piece, 'Organic Intellectual Notion – Knowing concepts without theory, published by the Sunday Independent, 27 May 2018.

'Organic' intellects abound in Africa – abbreviated version

My grandmother was my first engineering teacher. Vho-Tshianeo Marwala did not go to university in the US to study mechanical engineering as I did. She did not go to the University of Cambridge in England to study for a doctorate in artificial intelligence as I did. She never even left South Africa.

Despite these limitations, she taught me engineering at an early age. She taught me how to predict the failure of buildings in advance. Last month, three children died when an abandoned building collapsed in Johannesburg. The mayor said the tragedy could have been avoided. In March, three people died when a building in Durban collapsed. Again, it could have been prevented. In 2015, Grayston Drive Bridge over the M1 highway collapsed, killing two people.

To predict or prevent buildings collapsing, let us go back to the engineering lessons my grandmother gave me.

She made clay pots, an art rich in lessons for engineering, such as supply-chain management, metallurgy, applied mathematics, thermodynamics and artificial intelligence (AI).

First, one needs to identify good clay that requires knowledge of material science. The clay is delivered to the manufacturing venue to be processed and formed into pots, which involves three-dimensional visualisation and the ability to form shapes. The pots dry in the sun and are baked in a furnace. This requires knowledge of thermodynamics. The fire is put out and the pots cool slowly.  This is annealing, which is learned in metallurgical engineering. If the pots are cooled fast, they crack.

Annealing is such a powerful concept that an artificial intelligence algorithm called simulated annealing was created from it.

My grandmother was able to apply the annealing process without knowing the Boltzmann equation invented by the Austrian scientist, Ludwig Boltzmann.

To understand simulated annealing, it is important to understand his equation. My grandmother did not, yet she understood the practical side of annealing.

This notion of knowing a concept without knowing the theory of it is what Italian philosopher Antonio Gramsci called being an organic intellectual. This means education is important in its totality.

It means we should incorporate indigenous knowledge systems into our curriculum, research and innovation agenda. It means that Africans should not be alienated from science.

My grandmother would tap each pot and listen to the sound and this told her whether it was strong or weak. Using sound to test quality, is what the engineers call non-destructive testing.

It's applied in aerospace engineering to assess whether planes have body cracks.

In our book, Condition Monitoring Using Computational Intelligence, we take the concept of listening to objects into the fourth industrial revolution, an era in which machines are becoming more intelligent.

The sound my grandmother was listening to is what we call vibration data in engineering.

Vibration data is processed by artificially intelligent machines in the fourth industrial revolution.

The older Vho-Tshianeo got, the more good pots she disposed of because her hearing was deteriorating. Artificially intelligent machines do not have hearing loss.

At the University of Johannesburg, we offer courses in vibration analysis, signals and systems, thermodynamics and artificial intelligence. These are necessary to take the framework I was taught by my grandmother into the fourth industrial revolution.

This framework can be used to monitor the safety of buildings and bridges.

Data acquisition devices are embedded on buildings and bridges and data is relayed to an artificially intelligent machine, which analyses it and decides if there is any danger of a collapse. In case of imminent danger, automated messaging can be used.

The views expressed in this article are that of the author/s and do not necessarily reflect that of the University of Johannesburg.



 Professor Tshilidzi Marwala