Professor Elfatih Eltahir hosts workshop on “the Future of the Nile Water”

May 11th, 20182018 News in Brief

Professor Elfatih Eltahir, the Breene M. Kerr Professor of Hydrology and Climate, hosted an international workshop at MIT on “the Future of the Nile Water,” on April 26 and 27. At the workshop, Eltahir presented a proposal for Sustainable, Smart, Equitable and Incremental resolution of the Nile’s water conflict to participants from Egypt, Ethiopia, and Sudan. The proposal is based on research conducted by Eltahir’s students. Read more on Eltahir’s website and on MIT News.

Professor Elfatih Eltahir, the Breene M. Kerr Professor of Hydrology and Climate, hosted an international workshop at MIT on “the Future of the Nile Water,” on April 26 and 27. At the workshop, Eltahir presented a proposal for Sustainable, Smart, Equitable and Incremental resolution of the Nile’s water conflict to participants from Egypt, Ethiopia, and Sudan. The proposal is based on research conducted by Eltahir’s students. Read more on Eltahir’s website and on MIT News.

+ More

Professor Lydia Bourouiba featured in Scientific American for her role in Breakthrough: Portraits of Women in Science

May 7th, 20182018 News in Brief

Assistant Professor Lydia Bourouiba was featured in Scientific American for her role in Breakthrough: Portraits of Women in Science, an anthology of short documentaries produced by Science Friday and the Howard Hughes Medical Institute. Breakthrough: Portraits of Women in Science highlights six distinguished scientists and their research. The aim of the anthology is to increase the public’s access to science and inspire and increase the numbers of minorities in STEM. Read more here.

Assistant Professor Lydia Bourouiba was featured in Scientific American for her role in Breakthrough: Portraits of Women in Science, an anthology of short documentaries produced by Science Friday and the Howard Hughes Medical Institute. Breakthrough: Portraits of Women in Science highlights six distinguished scientists and their research. The aim of the anthology is to increase the public’s access to science and inspire and increase the numbers of minorities in STEM. Read more here.

+ More

Professor Yossi Sheffi featured in The Boston Globe

May 5th, 20182018 News in Brief

Yossi Sheffi, the Elisha Gray II Professor of Engineering Systems and CEE, and Director of the Center for Transportation and Logistics, spoke with The Boston Globe about his book, “Balancing Green: When to Embrace Sustainability in a Business (and When Not To).’’ In the interview, Sheffi discusses how companies should approach sustainable practices and how it is integral to consider the entire supply chain process when evaluating a company’s impact on the environment. Read the article here.

Yossi Sheffi, the Elisha Gray II Professor of Engineering Systems and CEE, and Director of the Center for Transportation and Logistics, spoke with The Boston Globe about his book, “Balancing Green: When to Embrace Sustainability in a Business (and When Not To).’’ In the interview, Sheffi discusses how companies should approach sustainable practices and how it is integral to consider the entire supply chain process when evaluating a company’s impact on the environment. Read the article here.

+ More

CEE Research Night brings community together to foster new ideas

May 2nd, 20182018 News in Brief

CEE Research Night, held on April 24, brought the community together to showcase the wide range of research conducted in the department. Featuring over 30 electronic posters, the night was filled with research presentations and networking. Attendees also helped to determine the overall winners of the event. Read more on MIT News.

CEE Research Night, held on April 24, brought the community together to showcase the wide range of research conducted in the department. Featuring over 30 electronic posters, the night was filled with research presentations and networking. Attendees also helped to determine the overall winners of the event. Read more on MIT News.

+ More

Infrastructure monitoring in communications-constrained areas

May 2nd, 2018Undergraduate Student Life

By René Andrés García Franceschini ’19 During one of the first weeks of the fall semester of my junior year, I had a call with Andrew Weinert, a researcher at the Lincoln Laboratory, to try to narrow down what my SuperUROP project would be. Weinert works on the use of small Unmanned Aerial Systems (sUASs) to gather information after emergency situations. Emergency situations were in no short supply those days: Hurricanes Harvey and Irma had come and gone, and Hurricane Maria had just hit Dominica. Knowing that Maria was a day away from devastating my home in Puerto Rico, Weinert suggested that I take this experience as motivation for improving emergency response in these devastated islands. Having personally been affected by the collapse of the communications infrastructure back home (I did not hear from my family for almost a week), I saw the importance of having effective emergency response in communications-constrained areas. I thus decided to work with Professor Saurabh Amin on the use of sUASs for infrastructure monitoring in communications-constrained or -restricted areas, especially after emergency situations. Currently, sUASs are becoming more common in infrastructure inspections because of their increased levels of access and safety, compared to a human inspector. sUASs are currently being developed so that they can be controlled through GPS and 4G LTE signal, which opens the door to the possibility of automatically routed teams of drones carrying out tasks. However, because 4G LTE coverage can be variable after emergencies, my work tried to figure out [...]

By René Andrés García Franceschini ’19

During one of the first weeks of the fall semester of my junior year, I had a call with Andrew Weinert, a researcher at the Lincoln Laboratory, to try to narrow down what my SuperUROP project would be. Weinert works on the use of small Unmanned Aerial Systems (sUASs) to gather information after emergency situations. Emergency situations were in no short supply those days: Hurricanes Harvey and Irma had come and gone, and Hurricane Maria had just hit Dominica.

Knowing that Maria was a day away from devastating my home in Puerto Rico, Weinert suggested that I take this experience as motivation for improving emergency response in these devastated islands. Having personally been affected by the collapse of the communications infrastructure back home (I did not hear from my family for almost a week), I saw the importance of having effective emergency response in communications-constrained areas.

I thus decided to work with Professor Saurabh Amin on the use of sUASs for infrastructure monitoring in communications-constrained or -restricted areas, especially after emergency situations. Currently, sUASs are becoming more common in infrastructure inspections because of their increased levels of access and safety, compared to a human inspector. sUASs are currently being developed so that they can be controlled through GPS and 4G LTE signal, which opens the door to the possibility of automatically routed teams of drones carrying out tasks. However, because 4G LTE coverage can be variable after emergencies, my work tried to figure out better ways to route sUASs while accounting for this uncertainty.

To achieve this goal, I used an approach that combines tools for vehicle routing and geo-statistics. First, I adapted algorithms used for routing teams of agents that need to visit as many points of interest as possible to my scenario. Second, I introduced a kriging step, which attempts to predict the communications signal strength of unobserved locations based on all the locations that have been observed. This allowed the sUASs to simultaneously inspect nodes and learn about the communications-space of the map. In the end, this resulted in improved performance in terms of how many points of interest are inspected, how much computational time is required to make good solutions, and how many sUASs make it back to home base before time runs out.

Of course, conducting research is hardly meaningful if you cannot convey the information to others. Thankfully, the second half of SuperUROP is learning about different genres and means of communicating your work. Through SuperUROP, I got to engage with hundreds of other students carrying out research in fields that are opposite to my own, and I got to see what it was like to communicate research across academic fields. I also had the opportunity to network with amazing professors and industry leaders that are on the cutting edge of scientific research. And all of this while making $25 an hour.

Presenting my research during a SuperUROP poster session

Towards the end of the program, I was invited to record a short video about my experience. In the video, the interviewer asked me if I was doing this project because of Hurricane Maria. The reality was that I had discussed this project with Saurabh in May 2017, months before Maria even existed. However, through ridiculous serendipity my project evolved into the sole topic that I wanted to work on this year. In my work with Saurabh, I was able take a problem with real-world implications, model it, and use my Course 1 skillset to chip just the tiniest bit of it.

+ More

There’s a 30% Chance that it’s Already Raining: Inside a CEE Capstone Project

May 1st, 2018Undergraduate Student Life

By Rachel Galowich ’18 and Jill Dressler ’18 Your two favorite blogging seniors are back! In our last post, Jill gave a great introduction into the background and significance of our project. With our next two posts, we hope to dive a little deeper into the computer science going on behind the scenes. Jill and I are working with the same data set (a refresher: MassDOT traffic camera images, and the labels applied to each image by Google Cloud Vision API), but focusing on different objectives. My goal is to be able to predict the probability of future events given historical data. My first order of business was to sort through the thousands of labels Google produced. Most labels can be applied to any image, so they don’t provide new or significant information that can help MassDOT respond to an accident - “road”, “asphalt”, and “lane” are a few examples. There are a series of labels that appear to be random, and in some cases, are pretty comical, like “Arctic Monkeys” and “bodybuilding.” And finally, buried within the sea of all of these unhelpful labels, are the 43 that I am actually working with. They might indicate a reason to be concerned for driver safety (i.e. an event that might lead to an accident), the presence of an accident, or emergency response to an accident. To test whether the events represented by image labels can be predicted, we can introduce a lag in the data, or “shift” the data back [...]

By Rachel Galowich ’18 and Jill Dressler ’18

Your two favorite blogging seniors are back!

In our last post, Jill gave a great introduction into the background and significance of our project. With our next two posts, we hope to dive a little deeper into the computer science going on behind the scenes. Jill and I are working with the same data set (a refresher: MassDOT traffic camera images, and the labels applied to each image by Google Cloud Vision API), but focusing on different objectives. My goal is to be able to predict the probability of future events given historical data.

My first order of business was to sort through the thousands of labels Google produced. Most labels can be applied to any image, so they don’t provide new or significant information that can help MassDOT respond to an accident – “road”, “asphalt”, and “lane” are a few examples. There are a series of labels that appear to be random, and in some cases, are pretty comical, like “Arctic Monkeys” and “bodybuilding.” And finally, buried within the sea of all of these unhelpful labels, are the 43 that I am actually working with. They might indicate a reason to be concerned for driver safety (i.e. an event that might lead to an accident), the presence of an accident, or emergency response to an accident.

To test whether the events represented by image labels can be predicted, we can introduce a lag in the data, or “shift” the data back in time. Our data set includes images taken every three minutes, so shifting the data 20 times represents one hour of total lag. I have been using two tools to show the relationship between the data and its lag terms: covariance and partial-autocorrelation plots. Covariance plots reveal the labels that are strongly correlated with their lag terms, and partial-autocorrelation plots can show how the relationship changes with the number of lags. From this, I can learn what labels might be more easily predictable, and how far in the future they can be reasonably predicted.

The next step is to build a statistical model that can more concretely prove the possibility for prediction. I have been using logistic regression, which gives the conditional probability that an event will occur. In this case, the output is a binary variable, which takes the value of one if a label is applied to an image, and zero if it is not. The input is the series of lagged data for a label. I am still working towards refining this model, but am making steady guidance with Jeff’s guidance (and Jill’s support, of course). The eventual hope for this part of the project is to use the logistic regression model and another machine learning technique, hidden markov models, to infer the probability that one event, represented by a label (i.e. snowing outside → “snow”), might transition to another event (“snow” → “accident”).

Hope you enjoyed reading this post, and that Jill can live up to my blogging-prowess next week! Cheers until the semester finale!

Rachel Galowich ’18 and Jill Dressler ’18 are working together on a Senior Capstone project with Professor Saraubh Amin and Doctoral Candidate Jeffrey Liu.

+ More