Article - Issue 49, December 2011

Thinking Smarter

Dr Scott Steedman CBE FREng

Download the article (50 KB)

Dr Scott Steedman FREng

Dr Scott Steedman CBE FREng

Smart cars, smart buildings, smart cities: over the last few years, the ‘smart’ word has moved into the mainstream. The Academy is currently preparing a report on smart infrastructure, to be launched in January, which is aiming to set down some principles. So what does smart mean in this context? What do we need to do to ensure that a fusion of information and communications technology with traditional engineering solutions brings real innovation in the efficiency, safety and comfort of our urban environments and infrastructure systems?

At a time when we need to squeeze more out of less, building smart seems attractive. The value of smart systems is that feedback from the system itself provides opportunities to increase performance that traditional engineering cannot deliver alone. Properly designed and operated, engineered systems in smart cities will work more effectively, delivering a better and more sustainable outcome for everyone.

A smart system monitors, measures, analyses, communicates and acts. It uses information captured from sensors to deduce changes in demand or availability and then adjusts the way the system functions – either by providing information so that humans can intervene or through an automated response.

A simple example of a smart system is a flashing warning sign on a motorway. Cameras ahead detect an incident and the system delivers messages to alert drivers approaching the area so they can decide to act, perhaps by changing their route, to avoid possible delays.

The concept of using information from sensors to improve the performance of engineered systems is not new. What is new is the idea that this same approach could be applied to systems on a grand scale, such as at the urban or city level. Smart cities would not only have better transport systems, they could also reap many other benefits, from safety and security to energy consumption to public health and wellbeing.

Our challenge as engineers is to reach beyond the obvious – traffic lights that sense traffic flows, or warning signs on motorways – and to see new opportunities where information, delivered in the right form and at the right time, can yield real benefits. The essence of a smart system is timing. Different systems operate with different response times. Whereas a driver on a motorway has seconds to act, decisions to intervene at a city scale could benefit from data collection over much longer timescales – minutes for traffic jams, hours for floods, and weeks or months for many social issues, such as making streets safer at night.

Engineering smart cities will require new standards for the reliability and availability of data. Reliability will depend on the users’ expectations and the consequences of a wrong decision based on that data. At a simple level, if traffic alerts are frequently wrong, then drivers ignore them. Sophisticated modelling can take uncertainty into account, but if the outcome of a wrong decision means serious consequences, then the input data needs to be highly reliable.

Availability is a second key challenge. The built environment is already awash with information that could inform decision-making – from security cameras, utility meters, mobile phones, computers and satellite imagery. Gaining access to this data, even in a selective manner, raises questions of privacy and security that need to be resolved. But there is no need to instrument every kettle in every house to monitor electricity demand, or to track every mobile phone to measure speed on motorways. Selective sampling and careful fusion of data and their interpretation through robust mathematical modelling would provide highly reliable decision-making tools to benefit individuals, organisations and governments alike.

Smart cities promise shorter journey times, more productive working environments and a safer, healthier built environment. But as we operate closer and closer to the limits of any system, the consequences of failure become ever more serious. Smart engineering needs to be resilient and to use as many independent sources of information as possible to guide and direct the user, whether human or machine, to a better outcome.

However, defining exactly what people want from their built environment and what society is prepared to accept, balancing privacy against safety and performance, will require engineers to step outside their usual domain. Smart engineering will require thoughtful interaction with the audiences that will make use of our technology. Public engagement must be an essential element of all smart cities engineering.

Dr Scott Steedman CBE FREng

[Top of the page]