IBM Broadens Push in China, India to Predict Pollution Patterns

  • Indian capital to study link between traffic patterns and smog
  • Cognitive computing applied to policy responses to dirty air

International Business Machines Corp., which has been developing artificial intelligence to help fight Beijing’s toxic air pollution, plans to work with other municipalities in China and India on similar projects to manage air quality.

IBM Research, a division of the company, will help an advisory body known as the Delhi Dialogue Commission to understand the link between traffic patterns and smog in the Indian capital so that decisions related to improving air quality can be backed with modeling, the Armonk, New York-based company said on Wednesday.

The announcement comes as smog continues to enshroud Beijing this week, prompting city officials to issue a red alert for the first time since a four-tier warning system was put in place in 2013. Air quality in the Chinese capital is expected to remain at hazardous levels until at least Thursday morning, according to the U.S. embassy in Beijing.

The undertaking is an offshoot of a similar initiative started last year to address air pollution through a partnership with Beijing’s municipal government. In collaboration with Beijing, IBM has been testing a computer system capable of learning to predict pollution patterns 72 hours in advance, along with pollution trends as many as 10 days into the future.

IBM Research will also assist the city of Johannesburg and South Africa’s Council of Scientific and Industrial Research in modeling air pollution trends and quantifying the effectiveness of the city’s actions to tackle the issue, according to the statement.

Computer Modelling

The company is also working with local authorities in China’s northern cities of Baoding, Zhangjiakou, and in the Xinjiang region.

IBM’s model is able to estimate where pollution is coming from, where it will go and then analyze possible government responses, said Dong Jin, associate director of IBM’s China research unit. The systems apply IBM’s cognitive computing, which is capable of taking unstructured data, understand it, learn from mistakes and offer improvements, said Dong.

The company uses data drawn from air monitoring stations, meteorological and environmental satellites, emissions, land use, traffic patterns, social media and the macro-economy, said Zhang Meng, chief meteorologist at the research unit.

IBM is also using similar technologies to help forecast the availability of renewable energy in the U.K., the U.S., Japan and China, according to the statement.

— With assistance by Feifei Shen

    Before it's here, it's on the Bloomberg Terminal. LEARN MORE