Weather data isn’t just about predicting rain
The weather is perhaps the one universal that every human can relate to, it is the universal connector between everyone regardless of where you are from. Most people have seen broadly the same kinds of weather, despite geographical differences meaning that others see more extreme versions than others.
It is also one of the most important elements of any economy, after all, with any kind of excess of one type of weather, there is likely to be significant disruption, whether this is excessive precipitation destroying vital infrastructure or over abundance of dry weather causing droughts. It also has a profound impact on society as a whole, with some even saying that weather has played a part in starting the Syrian civil war, with John Kerry, then Secretary of State saying ‘I’m not telling you that the crisis in Syria was caused by climate change, but the devastating drought clearly made a bad situation a lot worse.’
It is therefore not simply a case of looking at the weather forecast to ascertain what to wear that day, companies need to be able to use long-term planning to predict how the weather in several weeks may impact them.
UK based supermarket chain Tesco have become well known for their use of weather data to help predict sales and stock requirements. For instance, if they can predict that there will be a heatwave in 3 weeks time, they can then make sure that they have increased stocks of disposable barbecues, sun cream and cold drinks to cater for the increased demand. When this was widely reported in 2013 it was said that the chain had managed to save £6m ($7.5m) per year and seen a cutting out-of-stock by 30% on special offers.
However, since 2013, when Tesco’s use of this data became more publicly known, the world’s weather has become considerably more complex. 2014, 2015, and 2016 have all been the hottest years on record, which has caused more unpredictability in our weather and more disruption to companies. This can be anything from supply chains being disrupted due to infrastructure damage through to shortage of stock because farmers couldn’t grow a specific crop due to difficult conditions. It means that the use of data is essential not only in the prediction of these events, but in the planning in case they occur.
There are a huge and diverse number of data points that can impact changes in the weather and trying to input every one and then predict with any degree of certainty down to localities is incredibly difficult and requires considerable power. It is why the NOAA invested millions in a Cray supercomputer that processes 3 quadrillion calculations per second. The huge variations in conditions that go into weather conditions requires this kind of power and is something that more and more companies are needing to look at to help run their businesses effectively.
Everybody knows that running a business effectively is as much about awareness of external influences as it is about what you do internally, and weather is the ultimate example of that. Every company can be impacted and the use of data to predict and react to it is essential.