When I was traveling through Beijing earlier this month on a GeekWire reporting trip, my throat became itchy with mucus and my eyes dried up because of the pollution. But locals told me that the week prior was much worse, so I was actually a bit lucky.
Fast forward to today, though, and the air quality in Beijing is reaching extremely hazardous levels, with schools canceling outdoor activities and factories reducing production this week as the government works to reduce pollutants. It’s the worst pollution Beijing has seen in nearly two years.
There are a number of ways technology is being used to combat this pressing problem and while visiting Beijing, we had a chance to stop by Microsoft Research Asia to see how Microsoft is assessing and predicting air quality data throughout China.
Yu Zheng is a lead researcher with Microsoft Research Asia, which is Microsoft’s main research facility in the Asia Pacific region that houses more than 250 researchers and developers, along with thousands of other employees.
Zheng and his colleagues focus on using big data and machine learning to tackle urban problems. One project they’ve worked on is called Urban Air, an interactive map that lets users see air quality levels across 72 cities in China.
The Chinese government already has air quality monitoring stations to inform its citizens of pollution levels, and many developers have built smartphone apps using this information. However, Zheng explained that these stations are expensive and can’t assess exact air quality measurements of every block of every city in China, which is problematic given that air quality readings can be dramatically different even between two nearby locations.
To produce a more precise measurement, Microsoft researchers in Beijing infer the real-time and fine-grained air quality information throughout a city based on historical and real-time data from the existing monitor stations, in addition to other sources like temperature, humidity, wind speed, weather, traffic flow, human mobility, structure of road networks, and more.
“We can assess the air quality of a region by using this model, even if there is no monitoring station built,” Zheng said.
Zheng added that people can use this information to inform decision-making for when to go jogging, or when to close windows, for example. The Urban Air mobile app is used three million times per day by users in China, Zheng said.
Taking it one step further, Microsoft has also developed a way to help consumers and government officials predict pollution levels over the next 48 hours. This technology, which crunches data from the monitoring stations, meteorological information, and a handful of predictive algorithms, is used by the Chinese Ministry of Environmental Protection, providing it with 48-hour fine-grained air quality forecasts for four major Chinese cities every hour.
Zheng said that something like Urban Air can help the government figure out exactly when and where to invest money to reduce traffic or factory production when trying to control pollution.
“This can inform decision-making for both everyday people and governments,” Zheng said.
You can check out Urban Air here.