CTwalk Map is an interactive web tool that seeks to highlight the social cohesion potential of neighbourhoods while unmasking local access. Utilizing population, location, and pedestrian network data from open sources, it estimates the number of individuals from different age groups who can reach city destinations within a 5 or 15-minute walk, highlighting opportunities for social cohesion and access inequities.
Transparency is a key feature of CTwalk Map. It openly acknowledges limitations in the available data, encouraging further data collection and refinement. Users navigate the map through a user-friendly interface and explore different destinations. They are encouraged to identify anomalies in the map and scrutinize the completeness of the used data sources.
Furthermore, it serves an educational purpose by enhancing public understanding of the relationship between AI, urban design, accessibility, and social cohesion. It simplifies complex concepts and provides in-depth insights into how street networks shape accessibility, empowering citizens to actively participate in urban planning discussions.
Vasileios Milias
(Βασίλης Μηλιάς)
My research centers around the intersection of urban accessibility, spatial data science, and machine learning. In my work, I combine computer science knowledge with participatory approaches to inform the design of healthy, sustainable, and inclusive living environments.
My primary focus is on developing computational methods, tools, and indicators for evaluating the degree to which cities are accessible, inclusive, sustainable, safe, and attractive for all. Through my work, I aim to support the UN's SDG 11: "Make cities and human settlements inclusive, safe, resilient and sustainable".
News
Received the first prize demo award for CTwalk MAp at the NWO ICT.OPEN conference.
Gave an invited talk at TNO on "CTwalk and CTstreets: Exploring Walkability and Pedestrian (Co)Accessibility through Interactive Maps".
Lead the workshop on “What Factors Impact Access to Public Spaces and for Whom?“, realized in the City Hall of Helsinki (Finland) with Equal-Life's researchers and stakeholders.
CTwalk Map featured in Rafagas Links "#2103: cities, biodiversity, minefield". Read HERE.
Published and presented a paper in the AGILE2023 conference, in Delft, Netherlands. Read HERE.
Published and presented a paper in the CUPUM2023 conference, in Montreal, Canada. Read HERE.
Lead the workshop on “What Factors Impact Access to Public Spaces and for Whom?“, with the support of the American Planning Association. Read HERE.
Presented CTwalk Map and how to map walking accessibility of different population groups at the National Program for Sustainable Digital Information Management Demo Day. Read HERE
Gave an invited talk at TU Delft’s CityAI Lab on computational methods for assessing accessibility.
Short CV
Work Experience
> April-June 2015: Project assistant (MIT, U.S.)
Education
> 2010-2016: BSc & MSc in Electrical and Computer Engineering (NTUA, Greece)
Relevant Conferences & Venues
> 2024: NWO ICT.OPEN2024 conference (Utrecht, Netherlands)
> 2023: 18th International Conference on Computational Urban Planning and Urban Management - CUPUM (Montreal, Canada)
> 2023: 26th AGILE conference (Association of Geographic Information Laboratories in Europe) - Spatial data for design (Delft, Netherlands)
> 2023: European Human Exposome Network scientific meeting (Leuven, Belgium)
> 2023: NWO ICT.OPEN2023 conference (Utrecht, Netherlands)
> 2022: European Human Exposome Network scientific meeting (Barcelona, Spain)
> Computers, Environment and Urban Systems | Journal
> Environment and Planning B: Urban Analytics and City Science | Journal
My top picks
> PostgreSQL
> QGIs
> Javascript (only when needed)
Relevant Projects
Here are some of the projects I have thoroughly enjoyed working on during the past years.
CTstreets Map
Interactive Map
Focus: Walkability
(2023-Ongoing)
CTstreets Map is an interactive web tool that seeks to highlight how walkable are the streets of Amsterdam.
It uses granular population, location, and pedestrian network data from open sources to estimate how walkable each street is.
It has been designed through a participatory approach involving urban experts who live and work in Amsterdam.
It estimates the degree of walkability for the 5 and 15-minute walking areas from each housing block in Amsterdam, revealing neighborhoods that encourage walking and exposing disparities in walkability.
CTwalk Map
Interactive Web Map
Mapping Pedestrian (co)Accessibility
(2023-Ongoing)
CTnetwork
Python Library
Focus: Accessibility &
Street Centrality Measures
(2023-Ongoing)
A Python library for calculating various street network centrality and accessibility metrics, as well as the size and layout of surrounding pedestrian and bicucly infrastructure.
(Co)accessibility
A Python library for calculating prospective social encounters between persons from different demographic groups
Streetception
Python and Javascript toolkit
Capturing perceived qualities along city streets
(2023)
A Python and Javascript toolkit for calculating perceived safety and attractiveness of city streets through crowd parcipation.
Social Distancing Dashboard
The Social Distancing Dashboard offers interactive city maps that show at street and neighborhood level whether people moving through public space can keep to the distance rules. It provides an overview of various aspects – such as the width of footpaths and the location of bus stops – that influence the extent to which we can actually comply with the distance rules. In this way, it can contribute to raising awareness of the limitations that the design of public space entails and to decision-making about COVID-19-related interventions in the city.
Equal Life
The EU-funded Equal-Life project will develop and utilise the exposome concept in an integrated study of the external exposome and its social aspects and of measurable internal physiological factors and link those to a child's development and life course mental health. This will be done using a novel approach combining exposure data to characterise, measure, model and understand influences at different developmental stages. The goal is to propose the best supportive environments for all children.
And here are some less relevant to my research but very fun projects I have also been a part of.
Dataslip
As we navigate the physical and digital world, we unknowingly leave behind an immense trail of data. We are informed about this via short statements (e.g., cookie popups) or lengthy documents (e.g., privacy policies). However, even when we know that data is collected, we remain largely unaware of its nature.
Dataslip is an interactive installation where the abstract notion of personal data is translated into a material and tangible representation in the form of a receipt. The receipt contains detailed information and illustrative examples of the data generated from our interactions with five different categories of products and services. Its length is proportional to the amount of data collected about us.
Kaggle's Competition
The goal of this competition was to identify from which camera models, images were taken. My team’s
solution was based on deep learning techniques using Python (Keras library).
Sentimental Airbnb
Collected data from Airbnb and Twitter and made a visualization that shows color-coded clusters of areas with similar prices, the sentiment of the tweets that have been sent in these areas and a wordcloud containing the word-of-mouth as represented by these tweets.