Blending AI and Travel Planning
The use of Artificial Intelligence (AI) in transport has been highlighted as an area of opportunity by both the Department for Transport (DfT) and the Chartered Institution of Highways and Transportation (CIHT). Whilst there are several areas where it can assist with quantitative data and analysis, it’s role in supporting mode shift and behaviour change is relatively unexplored.
This project sought to bring together innovative approaches to data collection and AI for understanding travel behaviours, creating a blueprint for highly targeted, personalised prompts for behaviour change. Through FTVG’s collaboration workshops, it unified Dr Felicity Heathcote-Marcz, an ethnographic expert from Bare Analysis, with Andy Martin and Joshua Hooley from the London Luton Airport (LLA) Surface Access team to test this approach using LLA as its incubator.
“We’re always on the lookout for innovative ways of encouraging more sustainable behaviours for our staff”, says Andy, “An airport functions like a 24/7 miniature city with its own unique transport networks, peak periods and demands, so this felt like a great opportunity to test out something new.”
The Bare Analysis team brought together experienced freelancers, including Dr Nima Dadashzadeh from the University of Huddersfield and Andrew Longman, an AI Consultant.
People First
The project commenced with over 100 ethnographic interviews with LLA staff, aimed at understanding everyday staff journeys and experiences. Felicity, who led the ethnographic research explained;
“In my opinion, nothing can replace gaining open and honest feedback from the people who are taking these trips every day. Speaking to people face-to-face really helped us to understand the principal influences behind their travel behaviours, as well as barriers to travelling more sustainably.”
The feedback revealed a number of themes that influenced travel choices; including the reliability of public transport, safety concerns among women and cyclists, the convenience of driving and lack of incentives for other forms of travel.
The team also utilised LLA’s Staff Travel Survey to support their findings, combining responses on mode choice with motives for change.
“We do these travel surveys frequently as part of our regular surface access planning”, outlines Joshua, “But this felt like a great chance to cross-reference this data with real-life stories and experiences that we wouldn’t otherwise have gathered.”
The travel survey found that approximately 70% of staff travel to work by car, with just over 20% travelling by public transport. It also revealed some additional factors which could influence behaviour change.
“We conducted a series of statistical tests to isolate the main drivers for change”, Nima said, “From this we found that commuting costs reduced the odds of people taking up sustainable transport modes but schemes like staff travel cards and Cycle to Work have the potential to create modal shift for LLA staff.”
A key finding from both areas of investigation was “the driving trap”, where people found it very difficult to stop using their car once they started using it to travel to work.
After analysing and categorising these findings, the team developed a set of personas linked to different travel patterns and decision-making behaviours, which are summarised below.
Integrating AI
The use of the personas helped to identify more tailored solutions to encourage more sustainable travel behaviour. These included potential schemes that the LLA Surface Access teams could roll out, and a series of prompts and motivators to encourage people to change modes.
These formed the basis of a prototype “AI Travel Assistant” that matched user data to personas and developed personalised messaging to inform and influence future travel behaviours.
Andrew Longman outlined how this prototype could be the developed further;
“The personal travel assistant could be the tip of the iceberg; as well as supplying users with transport information and sustainable travel nudges, it could also be used to support ridesharing or demand responsive transport, as people’s travel patterns are better understood.”
The team hope that the project acts as a catalyst for change; unlocking opportunities to pilot more interventions for sustainable travel at LLA, as well as helping LLA continue their digital transformation journey at pace, embracing AI to achieve sustainable travel goals.
The findings from ethnography and travel survey analysis can be found here:
The analysis of personas and potential AI options can be found here:
The prototype AI Travel Planner tool is available to test here: