Infinite Analytics

Stories

How Covid has warranted a next-generation approach to nowcasting

by Team IA, June 10, 2021

Instantaneous forecasting can be a great tool for planning, decision-making, and saving lives, especially in the Covid era. Until now nowcasting has been immensely helpful in predicting the weather and managing traffic wherein instant decisions can be taken to plan evacuations or re-route respectively. Precipitation nowcasting, for example, is done using Machine Learning and it focuses on 0-6 hour forecasts. The resultant forecasts can have a resolution of 1km with a latency of only five to ten minutes. This is including data collection delays, outperforming traditional models, even though the model is still in its nascent stages of development. State-of-the-art nowcasting techniques when combined with new technology such as electronic message boards or smartphones as an effective tool in minimizing injuries or loss of lives, especially during fog or dust storms which cause multiple-vehicle collisions.

But Covid has taught us that new generation or next-generation of nowcasting applications should be developed for a wide variety of specific needs of the society. The approach to next-generation nowcasting works at reducing the number of variables to arrive at more accurate analysis. Industry and country targeted approaches must be used to reduce the variables and arrive at the necessary few, pertaining to a particular sector. With the now specific selection, the variables will be relatively stable in relation to each other compared to the earlier traditional model, even during a major fluctuation such as the pandemic or other crises. This model will remain more adaptable and will give timely and unique insights into the economy. 

Why is next-generation nowcasting needed?

According to a McKinsey report, current nowcasting models need to be revisited as they gave unreliable results which led to misleading analysis of the economy. Though this is not uncommon during a period of major economic stress, this situation must be tackled for better results. The pandemic brought along a structural break with an unprecedented shutting of businesses across the world. This resulted in many variables being broken, and the resulting inability to get the impact of the ongoing events on the economy. While the demand for consumer goods such as fitness material and equipment, healthcare products etc reached new heights, the F&B industry came to a grinding halt. Variables which were used to predict about the other businesses, were thus rendered useless.

The next step in nowcasting would be an exciting range of smartphone-based apps for nowcasting. In the pandemic era it could mean knowing beforehand. For example, planning your financial/non-financial activities on the basis of the prediction about any imminent danger from new covid variants in your area and more such data. On a lighter note, such apps will also be able to tell you about a storm headed in your directions, route-planning for biking enthusiasts, garden-keeping using precipitation data and more.

If you want to know how AI can help your business, write to us at contactus@infiniteanalytics.com

Subscribe to our newsletter

* indicates required



   


Avatar

by Team IA

Leave a comment

Your email address will not be published. Required fields are marked *