Covering Disruptive Technology Powering Business in the Digital Age

Home > Archives > News > Interview with Brad Gammons How IBM is Using Data and Deep Learning to Clean Our Air in Cities Across the Globe
image
Interview with Brad Gammons How IBM is Using Data and Deep Learning to Clean Our Air in Cities Across the Globe
image
September 24, 2018 News

 

IBM’s Global Managing Director for Energy, Environment and Utility Industry Brad Gammons was in town for CEPSI 2018.  It’s not an event we are familiar with, but we have it on good authority that it is the most important event in the region each year for electrical supply industry. So, it’s appropriate that Brad would be in town to attend.

As a man who has had to testify on environmental issues in the United States House and Senate, we assumed that spending some time talking with DSA would be “water off a duck’s back” for Brad and that turned out to be the case. He shared information which helped us understand just how and why IBM is a leader in the energy and utility space, and how they are using data analytics with deep learning techniques to clean the world’s air one city at a time.

A look at Brad’s profile showed that he is IBM’s business leader for the Green Horizons initiative, so we used the chance to speak with Brad to get a little better educated about the initiative.

Brad explained that Green Horizons grew out a collaboration between IBM’s China Research Laboratory and the Environmental Protection Bureau of Beijing, to provide one of the world’s most advanced air quality forecasting and decision support systems, able to generate high-resolution 1km-by-1km pollution forecasts 72 hours in advance and pollution trend predictions up to 10 days into the future.

According to Brad, they are able models and predict the effects of weather on the flow and dispersal of pollutants, as well as the airborne chemical reactions between weather and pollutant particles. They did this by mixing and blending chemical, particle and weather models then using deep machine learning to continuously improve prediction accuracy. The scope of the project grew to incorporate “source identification” which involved implementing high-density sensor networks to identify where pollutant sources are emanating from, in turn allowing the Chinese government to act on this information taking appropriate action where needed.

From a technical perspective, Brad pointed out how this is a great example of how deep learning can be used to successfully solve highly complex issues with many “moving parts”.

When questioned on whether IBM has successfully used what they learned from their Green Horizon’s engagements for the benefit of smaller companies and organisations, Brad confirmed that this was the case. He explained that the “trickle down” has not just been limited to the energy and utility sector for which he is responsible. The benefits can be seen by anyone that has downloaded the weather company’s app which displays air quality indicators such as a breathing index which leverages the intelligence learned from the Green Horizon’s initiative.

The Green Horizon’s initiative itself has grown from strength to strength. Brad told us that the experience has moved from the starting point of air pollution to areas including soil and water with projects now rolling out with numerous other cities in China and across the globe. Examples include working with the Delhi Dialogue Commission to understand the correlation between traffic patterns and air pollution in India’s capital, clean air projects in China with the Environmental Protection Bureau in Baoding (one of China’s most polluted cities) to support the city’s environmental transformation. Also in Japan, working with Toyo Engineering Corporation and renewable energy company Setouchi Future Creations LLC to implement monitoring systems that will help Setouchi manage and control energy from the plant’s 890,000 solar panels.

We pointed out to Brad that China and South-East Asia do not have the best reputation for being ecologically and environmentally conscientious and wondered if he felt that the work that IBM has been doing in this region on improving air quality, reducing pollutants and improving sustainability were compatible with the demands of emerging economies.

Brad was aware of the “image” of Asian economies, but his experience over the last few years suggests that things are changing fast. In the case of South East Asia, Brad feels the pattern is similar to other parts of the world, with citizens becoming increasingly informed and aware about the dangers of air quality and they are increasingly getting easily available access to independent air pollutant monitoring services. As a result, citizens have become far more vocal and active in communicating to their governments that action needs to be taken. These concerns are leading to governments prioritising pollution and other environmental issues.

In the case of China, there were many issues to overcome. It was coal driven, with heavy industry including aluminium smelting and cement plants all contributing to air quality issues. However, you can hear the pride when Brad explains how he personally now sees China as a “poster boy” for what can be achieved and more importantly for demonstrating how it is possible to remediate these issues.

Brad spoke a lot about machine learning and deep learning, but we wondered what role AI was having in the work in which he and his teams engage. On raising this subject with Brad, he had the “wry look” of someone that was trying to pick between substance and hype. He knows that there is a lot of hype around AI and having an accurate definition of this technology is important. Historically Machine learning and deep learning were considered to be part of AI. Today in Brad’s view when we talk about AI we are referring to cognitive capabilities, increasingly in respect of machine and human interaction.

Given the potential for risk and danger to citizens if “things” go wrong in utilities such as power stations we wondered if Brad saw ethical dangers or risks of relying too heavily on AI capabilities in this space. According to Brad, IBM’s view in this respect is clearly defined. They do not use AI to replace human decision making it is used to augment and support it. “we don’t take the human out of the equation” instead AI is used to support human decision making.

Brad concedes that for very automated tasks like common Q&A, then AI-powered chatbots can handle queries without human collaboration. What Brad is referring to is where AI is used for deep complex operational issues, IBM’s approach is clear, it’s about augmenting and supporting the person doing the job. He gave a real-world example of a project that IBM carried out with a company called Fluor, one of the world’s leading engineering, construction and project management conglomerates. IBM implemented AI project management assistants that focused on making sure that projects stayed on track and identified areas where projects can be taken off its planned path. Brad’s point is that AI is not being used to replace project managers but to make them more effective.

The scope and scale of the projects that Brad describes show how companies like IBM really propel the whole analytics eco-system forward. They have the capability to work on projects of a scale that very few companies have the resource or experience to undertake. They work at the cusp of where data and deep learning technology can take us. In doing so contribute to the entire Big Data and Deep Learning eco-system. The cutting-edge work they do ultimately feed its way into “the data analysis eco-system at large”.

 

(0)(0)