Analysing weather data to improve transport management
In a previous blog post I’ve already discussed the benefits of letting artificial intelligence (AI) loose on historical weather data in the context of the energy sector. More specifically, I explained how this allows for a better forecasting of green energy production. But what if we applied the same concept (if only as a thought experiment) to the transport sector? Would it yield similar results?
The underlying idea or principle is basically the same: you analyse historical weather data in order to predict the use of specific modes of transport, so you can adapt your transport policy and services accordingly. And to a certain extent the mode of transport people tend or prefer to use does indeed depend on the weather conditions. If the weather is nice, people are more likely to walk or cycle, for instance. If the weather is bad, they are more likely to drive or use public transport.
A far from binary model
When we look at it more closely, though, it remains highly debatable whether the weather is as strong a determinant of transport mode as it is of green energy production. If the sun is shining, you can definitely count on a certain amount of solar energy being produced: there’s little to no doubt about that. But even if the sun is shining, you might still see a lot of people driving around in their cars or taking a bus instead of walking or cycling. So, in the case of the transport sector, this seems like a far less binary model. In other words: on top of the weather, there are probably a lot of other parameters at play here. Not the least of which is a cultural mindset or mentality that remains, in Belgium and Luxembourg at least, thoroughly fixed on private and business car use, as opposed to public transport or other alternative means of transport.
Also, adapting your transport policy and services to the weather is far less easy than adapting your energy policy and services. This is especially true in the case of public transport, where you have little flexibility, since you are working with fixed timetables and routes. The fact that the number of buses, trams, trains and coaches rarely matches the volume of people to be transported, also speaks volumes in that respect.
The ideal world of Smart Cities
In an ideal world, where you have real-time management (RTM) of public transport, you would be able to suddenly and flexibly increase or decrease your transport capacity depending, among others, on the weather conditions. In that ideal world, you would also have demand-responsive transport (DRT), which basically means that the mode of transport is auto-adapted to people’s needs – which could equally be influenced or determined by the weather. Not only could this new transport concept lead to an increase in multimodal people transport, but ideally it would also entail vehicles automatically altering their routes based on a particular transport demand rather than using a fixed route or timetable.
This ideal world may not have been (fully) realised yet, but that hasn’t stopped us from giving a name to it: the Smart City. In our DigiLabs we are looking into all kinds of new technologies, ideas and concepts that can help us turn today’s ideals, such as the Smart City, into future realities. In Belgium and Luxembourg that research is focused, among others, on artificial intelligence (AI). By investigating the impact of the weather on a number of different industries and services from an AI perspective, we hope to help our customers obtain a better view on their day-to-day (service) business and - why not? - create profitable new business opportunities for the future.