A year ago I didn’t know what a digital twin was, now it seems everything has one. I have been working with a digital twin called S2AM, the Siemens Sheffield Advanced Multimodal Simulator. A mouthful, but this is our name for a new system that simulates how people move through a transport network.
We have created S2AM with colleagues at the University of Sheffield, exploiting the graphical processor units we all have in our computers. GPUs usually make PC games run smoothly because they’re brilliant at doing lots of maths really quickly.
They use loads of processing cores, each living their own parallel lives. In S2AM that power is used to work out how individuals would move around a city or region. Some people always elbow their way onto a train however full it is. Others step back if the next bus looks busy. Some have limited mobility or vision. S2AM allows us to track how each of those people move through a transport network, ‘plugging together’ existing simulations of railway traffic, pedestrian movement, buses or trams.
We can change anything on the digital twin in a way we can’t in the real world – we can add another train each hour and see what happens to demand for buses. We could (virtually) increase congestion charging, estimate the reduction in road traffic, evaluate the improvement in air quality, and work out how many more people would travel by train.
We can change anything on the digital twin in a way we can’t in the real world
We could work out how all those travellers, with their quirks, behave if we gave them more accurate, tailored information to improve their journeys. Best of all we can use ‘big data’, information already available in today’s connected world, to see how accurate our model is and continuously improve it.
I’m not sure I’ll ever find my digital twin, but I know that S2AM is a twin with a future.