The Urban Demand-Response Challenge

When everything is said and done, the smart cities conundrum boils down to addressing the challenge of reliably and continuously supplying services and resources to very high concentrations of citizens and enterprises amid huge and often unpredictable spatial and temporal variations in demand for mobility, energy, accommodation, education, healthcare, security, and transportation.

This combination of urban densification and huge swings between low and peak demand represents a double demand-response challenge for cities:

• How to build infrastructure capable of addressing very localized and sudden peak demands while maintaining affordability of services—cost of living is a key enabler of economic development and social inclusion
• How to do this in a sustainable way without compromising the environment

While demand-response is most often seen in the context of smart girds, the term is used more generally here, referring to any system delivering output matching demand.

Three key paradigms and technologies are expected to fundamentally help solve the challenge described above:

• Artificial Intelligence– Deep learning algorithms will be instrumental to not only predicting demand levels in time and space but also intelligently rerouting and dynamically reallocating assets, as well as reorganizing service networks to optimize resource availability where and when it is needed. This can even include sophisticated dynamic pricing schemes to influence (read: flatten) demand, as already implemented in mobility with examples including Uber’s surge pricing and Road User Charging (RUC). Switzerland-based energy services provider and electricity producer Alpiq has been harnessing AI for energy optimization and dynamic charging for electromobility for many years (Energy AI). Flexitricity pioneered AI-based demand-response in the United Kingdom. AI allows smart cities infrastructure to be made both intelligent and efficient.

• The Sharing Economy– The sharing economy allows the looping of privately held assets like cars, energy generation systems, and housing into the wider public smart cities environment. Car and ridesharing (Uber), home sharing (Airbnb), office space sharing and co-working (WeWork), smart home surveillance and healthcare platforms, and micro-grids are allowing cities to accommodate peak demand in times of stress. The distributed nature of the sharing economy goes a long way in making resources available when and where they are needed. At the same time, it allows cities to reduce investments in centralized infrastructure such as public transformation, for example, by shaving off some of the peak demand requirements. It does require close cooperation between city governments and the private industry. 

• Cross-Vertical Holistic Approaches– Optimization across industries, as opposed to within them, allows yet another level of rationalization. For example, the aggregation and integration of centralized power networks, distributed microgrids, and electric vehicles allows bi-directional exchange of energy (V2G, V2H, H2G) in a dynamic way, yet again removing pressure from centralized infrastructure.

In all of the above cases, it turns out densification itself, while being part of the challenge, is at the same time part of the solution. Neither the sharing economy nor AI-enabled smart city infrastructure can effectively and efficiently operate in non-dense environments. AI is very much dependent on being able to leverage the power of the statistics of big numbers. Similarly, the sharing can only operate efficiently in environments where a large number of assets are shared among a large number of users in geofenced footprints. Urbanization and densification are intrinsically linked to many if not all of the smart cites approaches and technologies.

Vendors and suppliers positioning their solutions as enablers of one or more of the paradigms mentioned above will emerge as the leading smart cities agents. For example, it is critical that IoT platforms targeted at the smart cities vertical provide embedded support for cross-industry use cases, AI applications, and as a service business model.

Addressing the demand-response problem in a sustainable way is an even bigger urban challenge. Interestingly, the very same paradigms that help address the demand-response challenge also contribute to sustainability as unexpected allies, with distributed renewable energy just being one obvious example. Ultimately, these structural sustainability approaches should allow governments to avoid having to issue symptom-fighting legislation such as odd/even number plate driving rules.

While cities have long been regarded as the main culprits of pollution and its overall negative impact on the environment, they might well turn out to be the very agents through which long term sustainability can be achieved. To put it more bluntly, urbanization moving forward is the only way to achieve true sustainability. Through near 100% utilization rates of resources and infrastructure via extreme forms of sharing and optimization only possible in urban contexts, the environmental footprint per citizen can be dramatically reduced or, indeed, made zero via carbon-neutral or even carbon-negative cities (i.e., self-sufficient green cities with vertical gardens, renewable energy generation, carbon-neutral buildings, etc.).

However, there is one big caveat (or risk) to all this: cities’ dense populations are also very vulnerable to disasters, a risk which needs to be mitigated through advanced resilience strategies (see the “Urban Resilience: A New Smart Cities Objective” foresight and upcoming report on smart cities resilience). In a way, resilience is about solving the ultimate demand-response challenge in the face of catastrophes, representing the ultimate form of peak demand amid compromised infrastructure.