“A lot of solutions for the smart city are going to be developed by looking at citizens data.” – Cuautemoc Anda, Future Cities Laboratory/Singapore-ETH Centre Speaker Q&A

Cuautemoc  Anda, PHD Researcher for the Future Cities Laboratory/ Singapore-ETH Centre will be speaking at the Smart Summit Asia on the 1st of December 2016. 

The Singapore-ETH Centre for Global Environmental Sustainability was established in Singapore in 2010 as a joint initiative between ETH Zurich and Singapore’s National Research Foundation (NRF), as part of the NRF’s CREATE campus. The Singapore-ETH Centre is an institution that frames a number of research programmes, the first of which is the Future Cities Laboratory (FCL), followed by the Future Resilient Systems (FRS).

Cuautemoc, who will be speaking at the Smart Cities Summit, completed our speaker Q&A:


1. Please provide us with some information about the projects you are working on to make cities smarter?

In the Future Cities Lab we develop a tool for Transport Planning based on Multi Agent Transport Simulations (MATSim). This tool allows the prediction of future scenarios such as testing new transportation policies, evaluating changes in the transportation infrastructure and assess the mobility issues of new urban developments (e.g. HDBs, shopping centres). We are currently working on synthesising a more accurate synthetic population for the model by using Big Data (e.g. smart card transit transactions, mobile phone data) and Machine Learning techniques (e.g. Bayesian Networks, Hidden Markov Models).  Big Data-driven MATSim will ultimately enable decision makers to plan for more efficient and sustainable cities.


2. How are you working to engage the citizens in smart city activity?

In our research team we have the responsibility to show the possibilities and results you can get for the transport planning domain when you have different big datasets in use. Generally, if the scientific community and business can show great improvements on citizens life by using data coming from them, then citizens will be more likely to generate and share their information through participatory sensing mechanisms. This new information can be used again to synthesised better models and applications. Some researchers call this phenomenon the data mining ecosystem. 


3. Do you think there needs to be stronger business and public collaboration?

Definitely! A lot of solutions for the smart city are going to be developed by looking at citizens data. There should be clear understandings between business, citizens and public agencies, in the data sharing mechanisms that can allow the generation of smart applications. Big data is insightful, but it gets more powerful when you cross-reference different big datasets.


4. Which are the key ways data is being used to enhance smart city capabilities?

Big data by itself can be use to derive insights by looking at hidden patterns by using clustering techniques (E.g. DBScan, PCA). Other interesting capabilities result when different datasets are cross-referenced to produce a better understanding of the current state of the world (E.g. Data Fusion). However, in the planning process of a city this is not enough! We also need to be able to generate what-if scenarios. Data-driven forecasting models are developed to explore different alternatives that can help us take inform-decisions.


5. How are security and privacy concerns being addressed?

Privacy is one of the biggest challenges of Big Data in the Smart City. Cases like the Netflix privacy breach (https://www.wired.com/2010/03/netflix-cancels-contest/) show us how even if a dataset is constructed to preserve privacy, by cross-referencing with another related dataset, it can be de-annonimised. Algorithmic efforts like Differential-Privacy (DP) are mathematical privacy guarantees that for some particular datasets can enable secure data sharing. However, DP for example can’t be use for human-mobility traces, since the traces itself are pseudo-identifiers. de Montjoye et al. (2013) found that by only having four spatio-temporal points at the resolution of mobile phone antennas, one can identify 95% of the individuals. As an alternative, in our research group, we use Probabilistic Generative Models to attenuate the privacy concerns. We use machine learning and real data to calibrate the parameters of the model. Then instead of feeding the real data in our simulation, we generate samples from the model, so to decrease the chances of being able to match an agent in the simulation with a real person.


De Montjoye, Y. A., Hidalgo, C. A., Verleysen, M., & Blondel, V. D. (2013). Unique in the crowd: The privacy bounds of human mobility. Scientific reports, 3.al technologies is transforming almost every aspect of modern life, and no other event covers Smart Home, Smart Cities and Industrial Internet in as much detail.


About Smart Summit Asia

Smart Summit is a 2 day conference and exhibition covering the Internet of Things (IoT) ecosystem and its impact on the digital society.

With 4 in-depth event tracks and over 80 leading speakers, no other IoT event covers the Smart Home, Smart Cities and Industrial Internet of Things in as much detail.

Make sure you are present in Singapore on the 30th November & 1st December for THE Smart event of 2016.

To register your place please visit: iotsmartsummitasia.com/register-now


For more information please call Georgia Deery on +44 (0) 330 335 3900 or email This email address is being protected from spambots. You need JavaScript enabled to view it.