In this article, we profile Nana Abankwa, PhD Student at the University of Southampton and winner of the Rolls-Royce Data Storytelling Challenge Graduate Competition.
What is your background and area of study?
I began my undergraduate degree in Mechanical Engineering in 2010 at the University of Southampton. Following completion in 2013, I began a PhD programme at the Institute for Complex Systems Simulation (an EPSRC Centre for Doctoral Training at the University of Southampton). The first year was the taught component of the programme, equivalent to an MSc. degree. Afterwards, I joined the Computational Engineering and Design Group of the Faculty of Engineering and the Environment, for the research component of the programme. I am currently in the final stages of my PhD, where my research is focused on improving fishing vessel safety at sea using cost-effective computing technologies.
What attracted you to that area of study?
Long before I started university, I undertook online tutorials about programming, which I really enjoyed, and this kickstarted my initial interest in this area of study. During my undergraduate studies I found the related modules really interesting, and when it came to choosing optional modules I made sure I could choose as many related to computer programming. As my skills in programming developed I began to consider how you could use it to interact with the real world. I started looking into how you could program small devices which can actually measure tangible things in the real world, and my interest grew from there.
In terms of my PhD, it is slightly linked to a project during my undergraduate studies involving small electronic devices like Microsoft’s .NET gadgeteer for rapid prototyping. This project was under the supervision of my current superviser and developed into the idea for the PhD. After completing the taught year academic of the doctoral programme in 2014, we were allowed to pick the direction we would like our research to go in. Taking advantage of my experience and passion, I chose to look into programming which interacted with the real world through sensors using the Raspberry Pi, a low-cost single-board computer. With fishing being one of the most dangerous jobs in the world, I decided to focus on using single-board computers to improve safety at sea by developing a cost-effective safety monitoring solution.
You recently took part, and won the Rolls-Royce Data Storytelling Competition, congratulations! What made you apply for the competition?
I was immediately interested in the competition when I saw the advert for it because of my background in mechanical engineering, and experience from a previous internship during which I used data to develop a preventative maintenance programme. I was particularly interested in exploring and learning more about data that Rolls-Royce uses to make business decisions. During my undergraduate studies, I learnt about engines but did not have as much exposure, or access to the amount of data that was available to us in the challenge.
Additionally, I thought it would be useful for me to apply some of the data analysis and visualisation techniques I have developed throughout my PhD to the competition. Ultimately, I knew it would enhance my skills and the experience would be useful in my future career.
Can you summarise your winning entry?
In order to visualise the data presented in the challenge, my solution was to create a web application using the Python programming language, Dash (a python framework for building analytical web applications), and libraries including pandas and numpy. I aimed for the visualisation to be descriptive, diagnostic, predictive, and prescriptive, using a top-down approach. It was also designed to allow further exploratory analysis by stakeholders. With a good visualisation, whether the user is an operator, commercial officer, workshop manager, spares planner, or supply chain manufacturer, the visualisation should provide relevant and actionable insights.
The end product was an interactive website with which people could interact with the data. In addition to the main summary page, this had different pages for each sub-area of data, enabling users to drill down and focus on data they were interested in. I decided to use a web application because nowadays most people have quick access to web browsers through their phone, tablet, or PC. I believed my entry would be far more accessible and user-friendly compared to a solution that required additional software or licences from the user. Throughout the process, I reminded myself that ultimately, the goal is to turn data into information, and information into insight.
I was surprised but extremely happy when I was announced as the winner, as I had no idea about approaches used by the other entries and was aware that approximately 500 people had registered for the competition.
Do you have any other achievements you are particularly proud of?
Towards the end of my first research year of the doctoral programme, I was on the organising committee of the 5th Student Conference on Complexity Science in Grenada. This interdisciplinary conference was the UK’s largest conference for early-career researchers in complexity science, and attracted approximately 200 participants from all over the world.
Another achievement I am proud of, is being a member of the team that came 1st place in the Micro-Sailboat Class of the 2016 World Robotic Sailing Championship in Portugal. This competition enabled me to apply aspects of my PhD as it also used a Raspberry Pi. After winning this competition, the team was featured on the University of Southampton’s website.
Work during my PhD has also been presented at conferences and published as:
Abankwa, N. O., Johnston, S. J., Scott, M., and Cox, S. J. “Ship motion measurement using an inertial measurement unit.” In Internet of Things (WF-IoT), 2015 IEEE 2nd World Forum on, pages 375 - 380, Milan, Italy, 14 - 16 Dec 2015. https://doi.org/10.1109/WF-IoT.2015.7389083
Abankwa, N. O., Squicciarini, G., Johnston, S. J., Scott, M., and Cox, S. J. “An evaluation of the use of low-cost accelerometers in assessing fishing vessel stability through period of heave motion.” In 2016 International Conference for Students on Applied Engineering (ICSAE), pages 59 - 63, Newcastle upon Tyne, UK, 20 - 21 Oct 2016. https://doi.org/10.1109/ICSAE.2016.7810161
Some of my work has also been submitted to:
The IEEE Sensors Journal with the title, “Estimating the Longitudinal Centre of Flotation of a Vessel in Waves using Accelerometer Measurements.” Author List: Nana O. Abankwa, James Bowker, Steven J. Johnston, Mark Scott, and Simon J. Cox.
What are your career aspirations?
When I finish my PhD I aim to apply some of the skills I used during the Rolls-Royce competition to a data science career in either industry or research. I am hoping whatever I do has practical applications in the world of either smart devices or the Internet of Things, to either generate data or use data to develop actionable insights. As part of my prize for winning the competition, I am starting a 10-week placement with Rolls-Royce. I am really looking forward to this for two reasons. Firstly, the internship will allow me to implement my solution within Rolls-Royce and secondly, to see their approach to handling this sort of data. Even though some data was available during the competition, it will be interesting to learn how Rolls-Royce specifically manages and utilises data like this. During the internship, I will learn a lot more and hopefully, this will be an invaluable experience as a data scientist.
Who do you most aspire to and why?
That’s a tough one! I would have to say my supervisor, Professor Simon Cox, is definitely one of them, mainly because of his sheer passion for all things small device related - it is infectious! I am also inspired by my parents for all their support and teaching me the value of hard work and perseverance. Other people who inspire me are “data-storytellers” like Hans Rosling and Cole Knaflick whose book I referred to when tackling the Rolls-Royce Data Storytelling Challenge.