As the UK’s longest running and most prestigious prize for engineering innovation, the MacRobert Award continues to attract a high calibre of submissions each year. The winners receive a £50,000 prize, a gold medal, and the opportunity to sit amongst an illustrious cohort of previous recipients, including Rolls Royce, Raspberry Pi & Quanta Dialysis Technologies. Choosing the winner is a lengthy process, and the MacRobert judging panel has the unenviable task of selecting just three finalists to be publicly recognised for their extraordinary innovations.
Once an organisation reaches the finalist stage, the MacRobert judges take part in site visits to meet the nominated teams, explore their facilities, and ask searching questions to get under the skin of each innovation. In this blog post, the first of a series spotlighting finalists, we dive into the judges' visit to nPlan, who have been shortlisted for their revolutionary machine learning technology which is being used to accurately forecast risks within major construction projects including HS2, Crossrail and the Houses of Parliament.
nPlan
The first stop on the 2023 MacRobert Award judges' itinerary was to nPlan's headquarters in East London. After they were welcomed into the modern office space, the visit commenced with a presentation from Dev Amarita, nPlan’s Co-founder and CEO, who explained more about the origins of the company.
The judges heard that the inspiration for nPlan came from Dev’s previous experience working in industry. After witnessing several construction projects exceed timelines and budgets, he saw an opportunity for machine learning to be incorporated in construction management, in order to “build a world no longer limited by its appetite in risk”.
nPlan's innovative technology will help to avoid the major delays frequently witnessed within construction projects
Throughout Dev’s presentation, the case for nPlan’s technology became clear. Only one out of every seven large-scale construction projects finish on time - and the situation is worsening. Prior to the Covid-19 pandemic the median project delay was approximately 80 days. By the end of 2021 it was over 214 days - an increase of 167%. When learning of a carbon capture facility in Rotterdam which was unable to be developed due to the uncertainty surrounding the project’s timeline, Dev was determined to explore solutions.
The judges were impressed by the resulting innovation - technology that not only provides accurate assessments of when a project will be delivered, and at what cost, but can also identify any risks that may be encountered along the way.
Following Dev’s presentation, the judges engaged in a Q&A session with the nominated team members to further understand the innovation behind nPlan’s machine learning technology. The team explained nPlan’s individual forecasts are based on over 600,000 past project schedules, dating back to 1991, being fed through an AI algorithm – known as ‘deep learning’- which is then able to answer key questions every project manager needs to know. These might be, for example, how long will the project take, what might go wrong with my project and how can I reduce risk? The use of AI in this way is what makes nPlan innovative - deep learning technology supersedes the previous gold standard in project forecasting - quantitative schedule risk analysis (QSRA), which is rendered ineffective by inherent bias and the limited subjective opinions of the participants in forming their estimates.
Whilst it’s well known that judges mark each finalist against three distinct sets of criteria – outstanding innovation, proven commercial success and tangible societal benefit - the potential ‘wow factor’ of each nominee is also a serious consideration. Throughout the course of the visit, it was clear the judges were enthusiastic as to nPlan’s potential to bring substantial overall improvements to the productivity of the construction industry, which lags behind most other industry sectors.
Impressed by the quality of responses and the depth of knowledge displayed, the judges were keen to understand more about nPlan’s working culture. They viewed a presentation from Operations Director Sarah Grisman, confirming nPlan’s strong commitment to D&I: 37% of the organisation are female– above average for a tech start up- as are three out of four of the machine learning team. Sarah’s presentation resonated with the judges who also learnt more about nPlan’s commitment to working with local charities - including Build Force and Young Women’s Network - and providing free support to infrastructure projects based in Syria and Ukraine.
The judges were given the chance to reflect upon their visit to nPlan and were inspired by this young, diverse company’s work. Each judge commented on the technology’s immense potential to revolutionise the construction industry, transforming how projects are executed. As the visit drew to a close, it was clear the site visits to this year’s finalists were already off to a flying start. The visit to nPlan would be hard to beat – but could Paragraf, the second stop on the MacRobert judges’ journey, rise to the challenge?
The MacRobert Award finalist team from nPlan includes:
Dev Amratia, Co-founder and CEO
Natalia Culakova, Senior Data Scientist
Sarah Grisman, Operations Director
Vahan Hovhannisyan, Research Lead
Petra Ivascu, Machine Learning Engineer
Carlos Ledezma, Product Manager
Inneke Mayachita, Machine Learning Engineer
Alan Mosca, Co-founder and CTO
Dan Murphy, Senior Data Scientist
Peter Zachares, Senior Research Scientist
The winning team is presented with a £50,000 prize, a gold medal, national publicity, and a weekend away at Douneside House in the heart of the MacRobert Trusts estate in Aberdeenshire.
Nominations for the 2024 MacRobert Award will open in October 2023. Please contact Patrick Woodcock, MacRobert Award manager, on 02077 660 630 or [email protected] should you wish to discuss putting forward a nomination.