Projects

SECTION ONE

(Academic Projects )

DATA DRIVEN MODELING ACROSS SCIENTIFIC DISCIPLINES

Intersection of machine learning and complex urban and spatial systems

1- Planetary urban modeling: Modeling different urban phenomena at the scale of planet (Ongoing)

Project website: https://github.com/sevamoo/urban-modeling-at-scalehttps://sevamoo.github.io/3d/
Planetary urban modeling (Link)
Planetary urban modeling: Global road networks from Open Street Maps (OSM) (Link)

2- Urban morphology meets deep learning: A comparative study of urban forms in 1.1 million cities across the planet

Project website: https://sevamoo.github.io/cityastext/Related publication: Vahid Moosavi, Urban morphology meets deep learning: Exploring urban forms in one million cities, town and villages across the planet, https://arxiv.org/abs/1709.02939.
Urban forms in 1.1M cities, towns and villages
Urban morphology meets deep learning (Download the large version (550 MB!) of this map here.)
Finding similar forms to a selected urban form
Topological Data Analysis in 1.1M urban forms
Clusters of similar urban forms from all over the world. (More samples)

3- Learning Physics: Fast and scalable urban flood risk estimation, Learning to emulate slow physics based simulation engines (Ongoing)

Related publications:João P. Leitão, Mohamed Zaghloul and Vahid Moosavi, Modelling overland flow from local inflows in “almost no-time” using Self-Organizing Maps, Proceedings of 11th International Conference on Urban Drainage Modelling, Palermo Italy, 2018.Guo Zifeng, João P. Leitão, Vahid Moosavi, Learning Physics: Speeding up physics based urban flood simulations using deep convolutional neural networks, (Under review at the journal of Water Research) Collaborators Dr. Joao Paulo Leitao, research scientist, systems engineering and intelligent network operations, department of urban water management, EAWAG.Guo Zifeng, PhD student, CAAD, ITA, ETH ZurichDr. Mohamed Zaghloul, CAAD, ITA, ETH ZurichProject website: https://github.com/guozifeng91/urban_flood
Guo Zifeng, JoĂŁo P. LeitĂŁo, Vahid Moosavi, Learning Physics: Speeding up physics based urban flood simulations using deep convolutional neural networks
Multi-modal convolutional neural network architecture

4- Machine learning and structural design (Ongoing)

CollaboratorsChair of Structural Design at ETH Zurich Polyhedral Structures Laboratory, School of Design, University of Pennsylvania Block Research Group (BRG) at ETH ZurichRelated publicationsLukas Fuhrimann, Vahid Moosavi, Patrick Ole Ohlbrock, Pierluigi D’acunto, Data-Driven Design: Exploring new Structural Forms using Machine Learning and Graphic Statics, Proceedings of the IASS Symposium 2018 Creativity in Structural Design July 16-20, 2018, MIT, Boston, USA[1] (The Best Master Thesis Award at Department of Civil Engineering ETH Zurich)Liew, A., Avelino, R., Vahid Moosavi, Van Mele, T. and Block, P., Optimising the load-path of compression-only thrust networks through independent sets, (Accepted at Structural and Multidisciplinary Optimization)Karla Saldana Ochoa, Patrick Ole Ohlbrock, Pierluigi D’acunto, Vahid Moosavi, Beyond typologies, beyond optimization, Submitted to IASS Annual Symposium 2019 - Structural Membranes 2019, Form and Force, 7–10 October 2019, Barcelona, SpainHao Zheng, Vahid Moosavi, Masoud Akbarzadeh, Machine Learning Assisted Evaluations in 3D Graphic Statics, Submitted to IASS Annual Symposium 2019 - Structural Membranes 2019, Form and Force, 7–10 October 2019, Barcelona, Spain
Using deep convolutional neural nets as a surrogate to predict the behavior of a truss layout optimization algorithm.
Use of Self Organizing Maps to analyze the effect of different design variables on lots of design options, generated based on graphic statics.
Use of Higher Order Statistics (HOS) to cluster the generated design options for a stadium roof, by Combinatorial Equilibrium Modelling (CEM), a method of structurally informed form generation based on 3D graphic statics. For further info please refer to: Karla Saldana Ochoa, Patrick Ole Ohlbrock, Pierluigi D’acunto, Vahid Moosavi, Beyond typologies, beyond optimization, Submitted to IASS Annual Symposium 2019 - Structural Membranes 2019, Form and Force, 7–10 October 2019, Barcelona, Spain. Image generated from the data set from the master thesis, by Giancarlo Casutt.
Subdivision of the force diagram leading to different geometric forms. via 3D Graphic Statics
Hao Zheng, Vahid Moosavi, Masoud Akbarzadeh, Machine Learning Assisted Evaluations in 3D Graphic Statics
Hao Zheng, Vahid Moosavi, Masoud Akbarzadeh, Machine Learning Assisted Evaluations in 3D Graphic Statics

5- Health and place: Urban characteristics and urban air pollution at the street level (Initiated recently)

Key questions: 1- Can we predict the air quality measures at the street level and at the resolution of the buildings? 2-What are the nonlinear effects of different urban factors (e.g. types of activities, building geometry and density, etc.) on different air quality measures? 3- What are the main air quality regimes in a certain urban area?

Data from Google street view ACLIMA air pollution sensors


Building heights in San Francisco
Multi-channel spatial patches as inputs of a convolutional neural network, which predicts the air pollutions.
Multi-channel spatial patches as inputs of a convolutional neural network, which predicts the air pollutions for each spatial patch.

6- Data driven urban air pollution estimation: Finding candidate locations for air pollution monitoring stations in streets of Singapore

Related publication: Vahid Moosavi, Gideon Aschwanden and Erik Velasco, Finding candidate locations for aerosol pollution monitoring at street level using a data-driven methodology, the journal of Atmospheric Measurement Techniques 8, 3563-3575, 2015.Collaborators Dr. Erik Velasco, Center for Environmental Sensing and Modeling (CENSAM), MIT Singapore SMARTDr. Gideon Aschwanden, The University of Melbourne.Project website: http://goo.gl/TN3XdP

7- Deep learning applications for remotely sensed data sets

Collaborator: Dr. Thomas Esch, Head of Team of Smart Cities and Spatial Development, Earth Observation Center, German Remote Sensing Data Center, Land Surface, German Aerospace Center (DLR)
Data from acqin global sensors.

8- Exploratory city mining: Dimensionality Reduction and Geo-visualization of high-dimensional spatial patterns

Related publicationsVahid, Moosavi, "Contextual mapping: Visualization of high-dimensional spatial patterns in a single geo-map." Computers, Environment and Urban Systems 61 (2017): 1-12.Vahid, Moosavi, Computing with contextual numbers. arXiv preprint arXiv:1408.0889, 2014
Emergent clusters using the selected aspects. (Reference paper)

9- Markov Chains for data driven urban traffic simulation

Related publication: Vahid, Moosavi and Ludger Hovestadt, Modeling urban traffic dynamics in coexistence with urban data streams, Proceedings of the 2nd ACM SIGKDD International Workshop on Urban Computing. ACM, 2013.

10- Systemic risk in world economic networks

Related publication: Vahid Moosavi, Giulio Isacchini, A Markovian model of evolving world economic network, PLOS ONE | https://doi.org/10.1371/journal.pone.0186746, 2017.Supporting materials: https://sevamoo.github.io/Markovian_IO_SI_PLOSONE/
Perturbation Analysis: The effect of slowdown of electrical and optical equipment industry of China in other economic nodes (1995 and 2015).
Perturbation Analysis: The paradoxical effects of slowing down the activity of economic nodes on Kemeny constant of the network.
Systemic fitness of each country in comparison to its GDP share over time from 1995 to 2011.

11- Clustering patterns of cyclones of Antarctic

Related publication: Hosking, J. Scott, Ryan Fogt, Elizabeth R. Thomas, Vahid Moosavi, Tony Phillips, Jack Coggins, and David Reusch. "Accumulation in coastal West Antarctic ice core records and the role of cyclone activity." Geophysical Research Letters, 2017.Collaborator: Dr. Scott Hosking, British Antarctic Survey, NERC, Cambridge, UK

SECTION TWO

(Commercial Applications)

DATA DRIVEN MODELING IN INDUSTRY

1- Developing A Google like search engine for spatial search: Multidimensional analysis of millions of locations for real estate development

Collaborator: Dr. Christian Kraft, Institute of real estate and financial services, Lucerne University of Applied Sciences, SwitzerlandProject website: https://github.com/sevamoo/reference_project

2- Real estate market dynamics: Developing a real estate portal by crawling publicly available data streams in Switzerland

Related publication: Vahid Moosavi, Urban data streams and machine learning: A case of Swiss real estate market (Technical report 2016). https://arxiv.org/abs/1704.04979Project website: http://www.keylead.ch
A Google like search engine for spatial search. A sample result for a specific site.
Real estate analytics, freely available to public: http://www.keylead.ch
Data streams of real estate advertisements, visualized using Kepler.gl. (Interactive)

Management and IT Consulting Projects (2004-2011)

From 2004 to 2011, as a systems engineer and management consultant, I was heavily involved in different planning and economic projects at different scales of business, sectoral and national levels in Iran. Below is a short list of the main projects I was involved in this time period.

1- Identifying the growth strategies and organizational re-design, 2010-2011, (project manager)

Industrial Management Institute of Iran (IMI), Deputy of research and consulting, Tehran, Iran

2- IT based Re-engineering from June 2008- December 2010, (project manager and partner)

Deputy of management and resources development at ministry of health, IranProject value: 193K USDThis deputy with more than 1200 employees is responsible for the management of budgets and human resources of 300,000 employees all over the country. We designed and initiated the Program Management Office (PMO) for physical resource management department that was supervising constructions and development of more than 4000 hospitals and medical universities across the country. I was at the age of 24 and won this contract and managed a team of 15 people. Related publications: Vahid Moosavi, Reza Jazemi and Abbas Seifi. (2010). BPR in the physical resource management department of health ministry of Iran, submitted to first international conference on business process management, Iran, Tehran, Spring 2010 (in Persian).Vahid Moosavi, Enayatallah Moallemi, Abbas Seifi (2010) Toward an alignment model between process maturity and business process management in the enterprise, submitted to first international conference on business process management, Iran, Tehran, Spring 2010 (in Persian).

3- Developing a systematic methodology for identification and analysis of corporate new investment options, 2008 (project partner, senior analyst and system designer)

Saipa Corp., Tehran, Iran.Saipa is the second largest automaker company in Iran, with US$ 7.1 billion revenue in 2011Related publication: Abbas Seifi, Vahid Moosavi and Ehsan Ardestani, A conceptual framework for evaluating new business opportunities for corporate diversification, The Journal of Enterprise Transformation, 2:2, 105-129, 2012.

4- Developing the R&D road map of ministry of energy of Iran (electricity division), 2007, (senior analyst)

Niroo research institute, Tehran, Iran. In this national project we engaged more than 400 experts in focused groups and expert panels and we developed the 5-year strategic road map of electricity sector of Iran.

5- Organizational assessment and implementation of continuous improvement system, 2006 (project manager)

KVC, Isfahan, Iran.

6- Advisor to the CEO for implementation of continuous improvement system 2006, (project manager)

A chemical production company (Farzin-shimi), Isfahan, Iran,.

7- Organizational diagnosis, 2005, (junior Analyst)

A software provider company, Yekan co., Isfahan, Iran.

8- Organizational diagnosis 2005, (junior Analyst)

A contractor company, Sanaye hararati co. Isfahan, Iran.

9- Providing a methodology for stakeholder satisfaction analysis based on EFQM model, 2005, (junior analyst).

Isfahan steel co., Isfahan, Iran.