Irena Itova

UniverCities – Co-founder, PhD candidate University of Westminster

United Kingdom

irenaitova@univercities.ai

Mining Urban Sustainability
Smart tools potentials for the development of sustainable cities

Connecting land-use data and material flows to economic sectors, enables parallel analysis of resource intensities and land intensities related to economic activities. However, the lack of existing empirical study addressing the question of the spatial distribution of material flows and the implication changes in the metabolic profile of regions for regional land use changes still remains a great challenge. Smart technologies related to distributed computing, Ambient Intelligence (AmI) and Deep Learning (DL) have the potential in addressing this specific challenge. Their application has made a great contribution in developing and providing solutions to many of the most complex scientific discoveries, including the Human Genome Project.

Problems to be addressed by smart technologies:
This powerful application of advanced cutting-edge technologies can help address some of the most pressing urban sustainability-related problems that future mega-cities are expected to deal with. Global air and water pollution and agricultural soil depletion represent worldwide pressure, directly associated with meeting the ever-growing global urbanization high demands- bound to host 75% of the world’s population by 2050. The birth of these global mega-cities and the collective effort of national and international governments to provide their current and future residents good quality of living, in the same time represents an opportunity to collectively organize and address the common urban challenges.
Implementation-state of the art:
The new wave of urban computing is about the omnipresence of invisible technology in urban environments and thus citizens’ everyday life. Mining of urban sustainability includes:

dedicated powerful software application to log urban infrastructure based on blockchain technology
spatial organizations and interactions based on the network theory
mobility and travel behaviour
ecosystems and public services

By activating the use of Deep Learning (DL) and Ambient Intelligence (AmI) technologies in assisting better-informed planning, urban and territorial plans can become powerful “democratic and transparent dynamic tool” for local, regional, national and international governments to jointly respond and deliver human-centred planning that fosters collective, well-balanced prosperity. Both methods are at the infancy of their application in real-world issues and the global scientific community largely acknowledges their possible contribution. My research is positioned at the frontier of technology led urban studies, addressing the current and future urban challenges, highly acknowledging their interdisciplinary nature. It fosters novel experimentation and application of AmI and DL technologies in providing more informed insights and answers, supporting novel scientific discoveries in urban studies.
Limits of the implementation of smart technology:
There is insufficient research in the area of big data analytics and context-aware computing middleware components and prototypes focused on the context awareness in relation to the urban planning domain, and more specifically related to:
large-scale application in the context of smart sustainable cities (SSC)
modelling and management of context information in distributed pervasive applications and in open and dynamic pervasive environments
My research is focused on enabling large-scale application of AmI and DL using a mega-city metabolism-based model called Urban Metabolic Networks. The model trains artificial intelligence (AI) on million(s) anonymized data from various consumption trends, collected by residents from diverse demographics at the level of a mega-city. We are searching for and testing the application of the most suitable existing pattern recognition algorithm in addressing series of pre-defined urban challenges. The research has a goal to teach AI how to spot the earliest signs of pre-defined urban flows anomalies and respond with the prompt diagnosis.