Marina K.

Type
Private sector project: A research project resulted with a published article.

Description
In 2019, I had the privilege of being selected as one of the top 8 graduates and was subsequently awarded an Applied Data Science Fellowship. This fellowship provided me with the opportunity to conduct comprehensive research on the relationships between road crashes and urban features. My tasks included analyzing relevant literature, collecting and cleaning data, performing statistical analyses, and utilizing machine learning algorithms to gain insights into the complex interplay of road crashes and urban characteristics.

Following the OECD's recommendations for a modern road safety approach and employing casualty matrices, I systematically examined crash causation in 24 European cities, identifying cars as the predominant threat, with variations observed across different urban environments. Further analysis of urban features, including population density and infrastructure, revealed a noteworthy correlation: cities with higher pedestrian presence and lower speed limits demonstrated fewer casualties (see Fig 1.). While acknowledging potential reporting biases, my research emphasizes the need for an evidence-based road safety paradigm. This research advocates for urban decision-makers to prioritize walkability and address car-related hazards, fostering enhanced safety, public health benefits, and sustainable urban development.

Read more
You can read the scientific paper on EPJ Data Science. The source code of the project is stored on my Github.

Type
Personal project: Daily social mapping project happening every November.

Description
The idea of this online challenge is to create maps based around different themes each day of November using the hashtag #30DayMapChallenge. There are no restrictions on the tools, technologies or the data that can be used in the maps, the point is learning and sharing - that's is why I decided to give it a try and have some fun while doing so.
I mainly used the datasets that are publicly available, but I also played around with some web scraping (where possible) and the results are providing some useful insights, mainly for Italy - since that's where I live at the moment. You can see one of the maps I created here below:

Read more
The source code and other maps I created are hosted on my dedicated GitHub repo. Stay tuned for the 2024 edition of this super entertaining challenge!

Type
Public sector project: HORIZON EU (H2020) research and innovation action including 33 partners - INCIT-EV

Description
Since February 2023, I've been immersed in a compelling role as a Project Manager at a research foundation in Turin, contributing to a Horizon 2020 project called INCIT-EV and involving a vast network of 33 partners. This project aims to demonstrate an innovative set of charging infrastructures, technologies and its associated business models, ready to improve the EV users' experience.

My main responsibility has been orchestrating the development of a cutting-edge Decision Support System (DSS) alongside two internal developers and two external partners. This tool is tailored to assist Mobility Planners and Policy Makers in crafting customized action plans to promote the uptake of electric vehicles in cities. By facilitating seamless collaboration and communication, my aim has been to foster a sense of shared purpose among stakeholders, ultimately working towards a more sustainable future for urban mobility. Apart from the coordination activities, I also played a pivotal role in enhancing the tool's usability from the users' perspective.

Read more
You can read more about the project on the official website. The Decision Support System (DSS) tool will be accessible to the public in the near future here.

Type
Private sector project: Commissioned by a local Trade Union in Italy

Description
Between 2021 and 2023, I worked a lot with tourist arrivals data in Italy, performing various analyses and predictions. Two large-scale projects included the monthly analyses of mobile positioning data in one Italian region; and forecasting tourist arrivals for the Milano Cortina Olympic Games in 2026.

For the latter, I used a fixed-effect linear model in order to obtain the predictions of international arrivals in Lombardy region. Due to the history of an Olympic event in Piedmont region (2006), the model was first calibrated on the official international arrivals data in Piedmont region from 45 countries and during a period of 16 years, taking into consideration country-level GDP per capita, trade volumes and oil prices (including 2 dummy variables). This showed how an increase of arrivals during the Olympic Games is mainly due to international arrivals (see Fig.1).
After calibrating the model on Piedmont data, I applied and adjusted it to the Lombardy region arrivals data obtaining a result that predicted the peek of arrivals already in 2025, the year prior to the Olympic Games in Milano Cortina.

Read more
Due to privacy reasons, I cannot share more details about this project.

Type
Private sector project: Commissioned by a local Trade Union in Italy.

Description
In 2021, as a part of a large-scale project evolving around the prediction of mobility demand in Piedmont, I analyzed the mobility behaviours in the region during and after the COVID-19 pandemic thanks to open data from Apple, Google and Facebook.

The data taken into consideration included:

• general movements from Facebook (Movement Range Maps)
• requests for travel indications by car from Apple Maps (Mobility Trends Report)
• shopping areas and entertainment venues attendances; and office attendances from Google Maps (COVID-19 Community Mobility Reports)

Considering the period from March 2020 until June 2021, this analysis showed how mobility habits changed (at least temporarily), especially in the case of shopping, entertainment and office attendances where the % change was still below the baseline (-22% in the case of offices).

Read more
Due to privacy reasons, I cannot share more information about this project, but the data analysis code is available on my Github.

Type
Public sector project: HORIZON EU (H2020) research and innovation action including 69 partners - SHOW

Description
During late 2022, I was involved in a Horizon 2020 project called SHOW which aims to support the deployment of shared, connected and electrified automation in urban transport, all in order to advance sustainable urban mobility. During the project, real-life urban demonstrations taking place in 20 cities across Europe have seen the integration of fleets of automated vehicles in public transport, demand-responsive transport (DRT), Mobility a Service (MaaS) and Logistics as a Service (LaaS) schemes.

One of the demonstration sites also included the city of Turin, where 2 autonomous shuttles were operating in the hospitals area, along an authorized path of about 5 km. In order to drive operational improvements, I developed a Power BI dashboard for the shuttle tracking, leveraging expertise in data analysis and visualization.

Read more
I cannot share more details on the dashboard for autonomous shuttle tracking for privacy reasons, but you can read more about the project on the official website and also on the Turin pilot case.