I’ve often wondered why tube stations weren’t designed with platforms on one side of the train to let passengers exit before the doors open on the opposite side to let passengers board the train, like the shuttle between Gatwick North and South terminals. Wouldn’t that vastly reduce congestion, make the Underground more pleasant and speed up the tube?
Why wasn’t it designed that way? Was it too expensive, was it simply not considered or did someone crunch the numbers and decide it wouldn’t work? I suspect the answer is a mixture of the first two answers - after all, the tube was built a long time ago to service a much smaller population.
This short-term approach to infrastructure is seen all around the world, particularly in the UK, where double-decker, longer and faster trains are precluded by too many low road bridges, short platforms and ageing tracks. Many UK cities are blighted by poorly-designed, post-War council estates that are not only eyesores, but have been a detriment on their residents by reducing access to job opportunities, community engagement and even, in some cases, put their safety at risk.
Of course, both the UK’s transport infrastructure and council estates were designed with the best of intentions. Council estates were a marked improvement on inner-city slums (or bomb sites) and the train network was the world’s first. With the advent of smart cities, we now face a similar challenge of creating a lot of new infrastructure. However, this time, the data that fuels a smart city will be instrumental in its long-term planning.
By analysing the behaviour, movement and interactions of the population of a city and combining this with long-term demographic trends and Census data, the infrastructure of a smart city can be better planned in the short- and long-term. This information can influence both macro policies, such as the provision of municipal services and urban development, and granular issues, such as the design of buildings and the timing of trains.
For example, by deploying beacons in a skyscraper or shopping centre, architects can use real data on how people react to fire alarms, how they use services such as lifts and toilets, and how they interact with surrounding city infrastructure. This data, properly analysed in contrast to existing computer models based on best guesses, should usher in an era of building design that is empirically perfect for a city and its residents.
Imagine this level of attention to detail deployed on a city-wide scale in perpetuity. It naturally makes you wonder how well-planned and efficient a city could become. In 20 years, we may look back at the congested, concrete jungles of today’s cities the same way we recoil at the thought of the smog-draped, slum-filled metropolises of the 19th Century.
Of course, the theory of wide-scale data science-driven planning is naturally bound to what is financially and politically feasible. In a city like London or Manchester, any “smart” urban renewal is going to be piecemeal and likely to focus first on peripheral or dilapidated areas. A pure, entirely smart-designed city is likely to be built from scratch in countries like China, India or the UAE.
This may be no bad thing for the UK. The industrial revolution meant that it was first with a lot of new infrastructure and city designs, creating a scenario where subsequent developments in other countries were able to learn from (and leapfrog) the UK’s mistakes. The design of smart cities could take some time to perfect, but the benefits can scarcely be imagined.