Here’s how to build better public transport after COVID-19

This article is brought to you thanks to the collaboration of The European Sting with the World Economic Forum.

Author: Vinod Bijlani, AI and IoT Practice Lead, Hewlett Packard Enterprise

• Commuters are wary of returning to public transport in the wake of the COVID-19 pandemic;
• Technologies such as AI and the Internet of Things can help detect an control the risk of transmission during journeys on public transport;
• Contactless systems, automated sanitization and big data can all contribute to a better public transport system that prioritizes health and safety.

COVID-19 may have put the world on a one-way street to change, but it has not stopped the traffic. As resilient people across the globe prepare to return to a new kind of normal, commuting and public health will take centre stage.

Using public transport will be inevitable in this near normalcy and transport agencies are gearing up with smarter strategies to support commuter health and wellbeing. Robust public transport systems are being reinforced with artificial intelligence (AI) and the Internet of Things (IoT) to detect and proactively control the risk of transmission during commutes. Pandemic preparedness is the keyword here and a holistic view of mass transit the pressing priority. Advanced technology and meticulous planning coupled with comprehensive public health policies are set to drive transportation in the years to come.

Innovation goals in public transportation have shifted radically in the wake of the pandemic. From developing fuel-efficient, greener and more reliable ways of mass transit, the focus is now on a safe commuter experience through uncrowded surroundings. Passengers expect a safer travel environment with precise temperature and mask screening, and better transit and station hygiene.

A recent global survey reveals that up to a third of people in some cities have been wary of returning to public transport. However, technology-enabled systems to adapt to the demands of social distancing and staggered peak hours could turn the tide for the better. With train frequencies specifically expected to be more flexible based on better managed real-time platform density data, travelling safely can get back on track. Such a transformation involves processing massive amounts of data and implementing incredibly complex analytics, but it is a significant step towards securing public confidence.

Using computer vision at station entrances

Among the primary measures for infection containment at the entry points of public transit stations are touchless fever detection and mask-detection technologies. Fever and mask detection combines thermal imaging with intelligent video analytics and AI techniques to not only identify individuals but also determine their body temperature and the presence of masks.

This solution has been implemented on the Argentina Trains network, which combines facial recognition and ultra-fast temperature checks with thermal and light imaging, while also validating ticket reservations. The access control system is intelligently connected to the Argentinian train database which checks for QR Codes and allows only essential workers to access the Buenos Aires Metropolitan Area trains during rush hour. Similarly, South Korean capital Seoul has installed glass-panelled bus shelters fitted with external thermal cameras and internal ultraviolet (UV) sterilizers. Installed along major routes, these shelters work to curb infection transmission by checking for fever and creating a sterile space in which to wait.

Speech recognition to reduce touch

Crowded stations equipped with speech recognition and voice-enabled ticket vending machines will minimize contact with high-touch surfaces and prevent cross infections. Shanghai Metro’s automated ticket vending machine uses voice and facial recognition technologies to dispense tickets after recommending ideal routes based on the destination.

Voice-activated buttons to call an elevator from outside or communicate the floor number from inside should be implemented at transit stations. For instance, the elevators designed by thyssenkrupp Elevator Italia have voice assistance that works on Wi-Fi or 4G to initiate the desired commands. Alternatively, buttons could be activated by hand proximity and motion detection without physical contact, such as the holographic buttons implemented in elevators.

Automated sanitization of escalators

Escalator handrails heavily used by commuters will need constant sanitizing to keep them germ-free. Automated escalator handrail cleaning systems that use UV light to disinfect the rubber belt should be deployed to ensure infection control despite contact. Transport for London has implemented these on London’s tube network after a six-week trial at the station serving Heathrow Airport.

AI-led optimization to monitor crowds and automate response

Challenges from crowds at stations and during transit will need to be managed with staggered work hours and other health safety regulations imposed by employment agencies to avoid rush-hour stress on the transportation network. Such regulatory changes are bound to impact Origin-Destination matrices which hold data representing movement through a city from an origin to a destination on transportation networks.

Transport planners will need to access data-driven simulation tools to build from actual-day reconstructions. They can then revisit schedules and provide insights and recommendations on policy implementation. Critical to this will be a holistic view of platforms, stations and commuters in transit. While real-time data of trains and buses is available, obtaining instantaneous, accurate commuter data can be a challenge.

AI and IoT techniques can help retrieve the information required by transportation planners for more efficient crowd control. Wi-Fi data from commuters’ phones can determine crowd densities both at the stations and in transit using an anonymous, non-participatory method. Transport agencies can employ probabilistic models and big data analytics to extricate crowd density information from platforms and a station’s vicinity.

Applying context-aware fusion analytics to the collected data can provide information on crowd surges and their causes, as well as inadequate supply-demand levels. Alert thresholds for crowd levels, along with Standard Operating Procedures (SOPs), can be defined based on the safe distancing protocols. The SOPs, automated using agile simulation systems, can then activate AI-based real-time simulations that provide optimized response plans, such as introducing additional trains and buses.

With advances in high-performance computing and big data, agencies will be able to make use of simulations in real-time to solve crowding problems. This system can also be used to improve response times during all types of incidents, track failures and signalling systems issues. Singapore’s Land Transport Authority has implemented a similar data-driven analytics system to manage commuter traffic surges and emergencies.

On the right track

Investment in advanced tech-enabled systems may focus on the present-day needs for infection containment. However, they should add value by creating new services and capabilities. Transport planners and agencies need to employ data-driven analytics to better understand locations and commuter behaviour for the future. Greater initiatives to evolve transportation systems should not only safeguard public health and safety but also take passenger experience to a new high. The upshot could be a reduction in commute-induced stress and fatigue, a boost to mental wellness and improved social wellbeing.