Research Bites: “Inferring Gas Consumption and Pollution Emission of Vehicles throughout a City”
As part of our "Research Bites" series, in which we ask data science researchers to spend five minutes telling us in their own words about their work, and opportunities for practical applicability in the context of sustainable development or humanitarian action, today we hear from researcher Yu Zheng from the Beijing lab of Microsoft Research. Yu is a co-author of this work along with researchers Jingbo Shang, Wenzhu Tong, Eric Chang and Yong Yu.
Inferring Gas Consumption and Pollution Emission of Vehicles throughout a City
Congestion and traffic jams cause drivers to waste gas, which on a mass scale causes greater air pollution that impacts on the environment and people's health. This study inferred the gas consumption and pollution emission of vehicles traveling on a city’s road network, using GPS trajectories generated by a sample of 32,000 taxicabs in Beijing over a period of two months.
The full research paper can be accessed here.
1. Tell us about your research paper in two sentences
Our research allows us to instantly infer the gas consumption and pollution emission of vehicles traveling on a city’s road network. Through machine learning and data mining techniques, we used GPS trajectories from 32,000 Beijing taxi cabs first to compute travel speed and traffic volume, and then to estimate gas consumption and vehicle emission for specific time slots and road segments.
2. Why do your findings matter?
Energy consumption and pollution are two major issues for cities. Urban traffic is connected to both. A better understanding of the energy consumption and emissions from traffic can help us to save energy and reduce pollution, which is critical to cities’ sustainability and people’s health.
3. How could this research be put into practice?
Here are a couple examples: Our research can enable energy-efficient, time-dependent traffic routing – suggesting the best driving route with the minimum gas consumption between origin and destination within a city’s road network. Real-time estimation of vehicle emissions can be put into practice through fine-grained pollution alerts and pollution controls. In the long run, our research findings can help diagnose the sources of air pollutants and inform governments’ decision-making on tackling air pollution. For instance, we can answer questions like “What is the percentage of PM2.5 generated by vehicle emissions?” Our research can also identify the specific roads where energy is being most significantly wasted. These kinds of knowledge can improve the decision-making on how to improve a city’s transportation infrastructure.
4. Why did you select this topic to research?
I live in Beijing, a metropolis that consumes a large amount of gas every day and is suffering from air pollution. There is a rising concern about air pollution and gas consumption from citizens and governmental officials. As a data scientist who is passionate about urban big data, I saw that the taxicabs traversing Beijing are mobile sensors continuously probing traffic patterns on road surfaces. I believe the trajectory data of taxi cabs, combined with my big data and machine learning expertise, can help tackle pressing urban challenges. Since 2008, we have been working on urban computing, which is a process of acquisition, integration, and analysis of big and heterogeneous data generated in urban spaces aimed at solving the major problems cities face. Urban computing connects unobtrusive and ubiquitous sensing technologies, advanced data management and analytics models, and novel visualization methods, to create win-win-win solutions that improve urban environments, human life quality, and city operation systems. Urban computing can help us understand the nature of urban phenomena and even predict the future of cities.
The researchers will be on hand to answer questions in the comments section below, so we invite practitioners from the development or humanitarian sectors to join the discussion!
Microsoft Research @MSFTResearch