Data science-backed rowers win race
After 35 days of rowing, the On Shoulders of Giants (OSOG) team won the ‘fours’ race class in the Talisker Whiskey Atlantic Challenge. Ahead of the race, OSOG partnered with Arca Blanca data scientists to help plan the optimal route.
In December we wrote about On Shoulders of Giants, the team of four friends from London who set out from the Canary Islands to row across the Atlantic in the Talisker Whiskey Atlantic Challenge. 35 days and 3,000 miles later, Rauri, Dan, Justin and Mark reached the finish line in Antigua and won the fours race class.
The four friends were motivated to support two causes close to their hearts – motor neurone disease research (through the My Name’5 Doddie Foundation) and supporting adults with learning difficulties (through the Pursuing Independent Paths charity).
“The four of us looked after each other,” said Dan about their teamwork. OSOG cemented their friendship and built incredible resilience by braving the constant two-hours-on/two-hours-off rowing shifts; the confinement to an 8-metre-long boat; and the challenging weather conditions encountered at sea.
Data science maps route
Arca Blanca’s data science team helped them to select a path that would win the race. While northerly routes are more direct, southerly routes typically have more favourable equatorial winds. The best solution is therefore a compromise between rowing a shorter route and leveraging the trade winds on longer routes.
Traditionally rowers use gut feel to find the best route. OSOG, in conjunction with Arca Blanca, shifted the paradigm on route planning using a data-led approach. Using data on crossings from six boats from the last three years, along with weather data from the last 20 years, the scientists created a framework to plot the quickest path to the finish line. The scarcity of crossing data required them to adapt methods usually applied to big data to work with smaller data sets.
George Cevora PhD, Chief Data Scientist said:
“Guiding a rowing boat across the Atlantic is quite a different challenge to what we usually do, especially because of the limited data availability. The exceptional versatility of my team was the main driver in maximising the potential of OSOG.”
After the team crossed the finishing line, the model was analysed for accuracy against the newly-logged data. OSOG arrived within 17 hours of the time than forecasted by Arca Blanca – which represents 3% of the overall time. This is high accuracy for a race in which boats routinely finish a few days apart.
A model like this could be developed for several different business applications where past data can help create insights for the future. For example, house sales forecasts, pricing estimates and consumer trends analysis. This specific use case also demonstrates how a smart use of data science can avoid the common pitfall of insufficient data – all too familiar in areas such as macroeconomics where key data are often updated only once a year, despite being crucial for strategic decision making of any business.
We congratulate the team again for their incredible achievement and wish them a well-deserved rest upon their return home!
A deeper look at the data science techniques used to help the OSOG will be published in the coming weeks.
About the causes
Pursuing Independent Paths (PiP) supports adults with learning disabilities to achieve their potential focussing on promoting choice, participation and independence.
The My Name’5 Doddie Foundation aims to make a difference to the lives of those coping and battling with Motor Neurone Disease (MND) through funding research into its causes and potential cures.
About Arca Blanca
Arca Blanca is an integrated consulting and data company, resolving complex growth and profitability challenges. We help businesses of all types thrive in the face of unprecedented change and upheaval driven by new technologies.
More in insights...