Since Prime Minister Mark Rutte addressed the nation on 12 March with a clear message to stay at home, public transport agencies were tasked with a unique set of challenges. During the ADS and ADMS webinar ‘COVID-19 and Capacity Modelling in Public Transport’, representatives from Netherlands Railways, or shortly NS, Dennis Huisman and Menno de Bruyn explained how the organisation used data science to adapt its services and predict its future.
Like most others, the public transport sector was immediately and powerfully impacted by the coronavirus outbreak and subsequent “intelligent lockdown”. Calculating the timetable which would enable essential workers to keep the country running – including healthcare staff on the frontlines of the pandemic and food industry workers keeping us fed – was a massive undertaking to set in motion immediately.
Dennis Huisman, expertise manager logistic processes at NS and professor of public transport optimisation at Erasmus University Rotterdam, admitted such a task “would normally take about a year”.
The importance of public transport data modelling in COVID-19
The webinar held on 2 July 2020 was the latest in a series of events hosted by Amsterdam Data Science (ADS) and Amsterdam Medical Data Science (AMDS) in collaboration with Elsevier and Google that explores ‘The Power and the Weakness of Data and Modelling in COVID-19’.
This edition took the conversation to public transport – namely, how can train operators continue to provide a necessary service, keep staff and passengers safe and overhaul nationwide timetables as responsively as possible? And what will be the impact on long term travel behaviour?
See full recording of this webinar here on this YouTube channel
A new train of thought
As many commuters’ sole means of travelling to work, train services had to continue to provide this service, while protecting their own staff (keeping drivers and crew to a minimum), operating enough capacity, with a reduced demand of approximately 85%, and discouraging the public from taking non-essential trips (the exact opposite of the organisation’s former strategy).
To upgrade the schedule changes were modelled using an algorithm known as a ‘line planning model’ to determine which lines had the biggest impact on capacity and travel time reduction. While lines between Amsterdam and Rotterdam (via The Hague) as well as those between Rotterdam and Utrecht are typically the most in-demand, the schedule had to account for a fair distribution throughout the country.
The various scenarios also had to adapt to changing regulations, for instance, there was a possibility to increase capacity after masks were required on public transport.
Looking towards the future of public transport
Menno de Bruyn, transportation researcher at NS, spoke about predicting the future demand for train travel. After all, the pandemic has drastically changed the way we live, work and socialise, and returning to ‘the way things were’ post-COVID-19 may not be realistic.
In search of answers, NS conducted a survey which targeted 47,500 regular train travellers. The data identified three major implications that threaten the future of NS. Firstly, we’ve figured out how to do things digitally. Home offices have become the norm and everything from family gatherings to concerts and Friday night drinks are taking place online, meaning the demand for train travel is at an all-time low for an unforeseeable length of time.
In addition, other modes of transport are gaining popularity. Travellers with access to a car use it as a safer, socially distanced alternative. Then there’s the question of retaining happy customers. Before COVID-19, NS received an 89% customer satisfaction rate for their offering. Now with a drastically reduced schedule and the potential to come in contact with the virus, the general public are unsurprisingly less inclined to feel as relaxed about train travel.
On the other hand there is also some good news: when people are resetting the way they travel, this may come with a decrease of peak demand when travellers choose to travel at different times.
Impacts beyond COVID-19
Quick responsiveness throughout the organisational structure is the most significant change for NS since COVID-19. The ability to pre-empt government decisions and implement schedules and procedures quickly relied on agile calculations of data, most of which is collected from OV-chipkaart use. Technological developments meant they can now process travel statistics on the same day, when information previously took two to three days to gather.
When will Dutch public transport return to pre-COVID-19 levels? As with most discussion around this virus, De Bruyn acknowledges “that’s the million-dollar question. We may be in a recession for years. It will take a reasonable amount of time, and it may even never happen.”
AMDS & Smart Health Amsterdam
The Amsterdam Medical Data Science Group (AMDS) is supported by The Right Data Right Now consortium, which includes Amsterdam UMC, OLVG, Vrije Universiteit, Pacmed, Amsterdam Economic Board and Smarthealthamsterdam.com
Smart Health Amsterdam is the network for data- and AI-driven innovation in Amsterdam’s life sciences and health sector. Join our community here