COVID-19 - the road to normal...
Published on June 15, 2021
Article by Alastair Kitson
Hi and welcome to this article, it’s been a while since I wrote one and I hope you’ll find it interesting and not just more COVID analysis! Over the last couple of years I have been working with real time modelling which has changed the way I think about mobility and the annual averages, neutral months etc that we all may be used to.
There is a lot of discussion about COVID recovery and how mobility is back to pre-pandemic levels. I think that is too broad brush, and there are many things that can be learned (or indeed as modellers, assumed!) which may improve our understanding of activities and behaviour with a little deeper delving.
The general headlines we hear are…
- Traffic is up …. or down!
- The peak period has changed, the peak has gone, the peak is back!
- Post COVID will be a different world / exactly as before!
Are these sensible? Is it still too early to tell? Is mobility that simple?
This isn’t a thesis - I have only taken weekday TRIS data from a few locations but I am confident that the lumps bumps and trends will be similar in your UTMC output, vision-based cams etc. As we are now all conditioned to rolling averages, this is also what I have used.
Let’s get started.
Plotting the timeline of one site (or locality), shows exactly what we all know and expect; lockdowns make traffic less. From here you might go in to make some basic assumptions about our travel needs, with around 55% of travel being essential. Perhaps a further 15-20% is education, and something like 20% which at no point has ever recovered is “remote office staff”. This leaves only about 10% for “leisure” traffic in a usual weekday.
However, what this covers up is that heavy goods vehicles - HGV’s (as may be expected) recovered much quicker, but not across the board.
Is it all just COVID? Perhaps we are seeing Brexit effects at the start of 2021?
With the HGV’s out of the analysis, you can really see how much light traffic dropped – as low as 15% near Swindon, and that isn’t taking consideration of the new delivery vans that are likely to be in the mix!
If it isn’t obvious, different regions have different dependencies on traffic and mobility. While mobility somewhere semi-rural like M40 Oxfordshire may be in the middle, demand on the M4 inside the M25 has remained less, while more rural locations are “normal” and as expected with seasonality, some positively booming!
Finally, that shift in peaks has happened – or has it? Again, basing against “early 2020” seasonality, the impact is very variable across all the differing types of trips, as may be expected, rural peaks back at 100% and more urban areas with more office dependency are not.
What does this mean? For me - lots to question.
First, data is useful! If you have monitoring sites, get recording and configured to categorize vehicle bins!
Second, as we know traffic is clearly complicated, regionally and there are a lot of factors in play - do we understand and represent it adequately?
Finally, is AI and data on its own useful, or do you need human interpretation to get meaning? Both - as practitioners, for example we can fall into a trap of assuming that as we work in offices, so must everyone else. The numbers suggest that the places that you may most wish to deploy MAAS actually are the places where travel is now most affected and “new normal” could change that for ever.
My thoughts might confirm what you expect or be radically different ramblings! Let’s hear what you think either way! Would love to hear that seaside towns have been doing great as the A64 data might show, and it isn’t a broken counter!
I have been wondering about the environmental impact of traffic vs NO2. I’m not quite brave enough to publish dabblings in a different field, but I would love to hear from you if you have facts and figures on it!
If you think this analysis may be helpful for you locally, give me a shout and I’ll try to help. email@example.com