Tag Archives: decision support systems

Geothink at the 2016 Annual Meeting of the American Association of Geographers

By Drew Bush

From March 29 to April 2, 2016, Geothink’s students, co-applicants, and collaborators presented their research and met with colleagues at the now concluded 2016 Association of American Geographers (AAG) Annual Meeting in San Francisco, CA. Over the week, Geothinkers gave 11 presentations, organized six sessions, chaired five sessions, and were panellists on four sessions. See who attended here.

“This year’s AAG provided a great opportunity to get geographically diverse Geothinkers together,” Victoria Fast, a recently graduated doctoral student in Ryerson University’s Department of Geography and Environmental Studies, wrote in an e-mail to Geothink.ca. “I can’t think of a better place for a meeting about a special journal issue on open data; there are so many fresh, uncensored ideas flying around the conference, both inside and outside of sessions.”

Of particular note for Fast was Panel Session 1475 Gender & GIScience (see her Geothink.ca guest post here). Panelists in the session included Geothink Head Renee Sieber, associate professor in McGill University’s Department of Geography and School of Environment; And, Geothink collaborator Sarah Elwood, a professor in University of Washington’s Department of Geography.

Others agreed.

“A panel on gender and GIScience was refreshing and enlightening,” Geothink Co-Applicant Scott Bell, a professor of Geography and Planning at University of Saskatchewan, wrote to Geothink.ca.

“My presentation was in a day long symposium on human dynamism,” he added. “It summarized a recently published Geothink aligned paper on human mobility tracking and active transportation (published in the International Journal of Geographical Information Science). It seemed to go over pretty well, I’m glad I was in the day-long event as the room was packed most of the day.”

For others, the high cost of the location meant they couldn’t stay for a full week or attend every single session. Still they reported good turnout by members of the Geothink team.

“This year we did not organize a specific panel or panels, or specific sessions to showcase Geothink work,” wrote Geothink Co-Applicant Teresa Scassa, Canada Research Chair in Information Law and professor in the Faculty of Law at the University of Ottawa. “This meant that our presentations were dispersed across a variety of different sessions, on different days of the week.”

Many Geothinkers were also intimately involved in running parts of the conference.

“This was a standout AAG for me,” wrote Geothink researcher Alexander Aylett, a professor and researcher at the Institut national de la recherche scientifique, who ran three sessions (Find an overview of what Aylette’s sessions did at www.smartgreencities.org). In collaboration with Andrés Lluque-Ayla from Durham University we ran a full day of sessions on the overlap between “Smart” and “Sustainable” cities.   We had some excellent presentations—including one from fellow Geothinker Pamela Robinson—and a strong turn out throughout the whole day. (Even at 8 AM, which was a shock to me!).”

For some students, it was the first time they had attended the meeting or presented their own research.

“This was my first time at the AAG,” said Geothink Newsletter Editor, Suthee Sangiambut, a maser’s student in McGill University’s Department of Geography with Sieber. “I was quite excited to be at the event and was able to meet all kinds of geographers, all of whom had different ideas on what geography exactly is.”

“It was great to see how global events of the past years were shaping our discussions on the Geoweb, privacy, surveillance, national identity, immigration, and more,” he added. “Those at the Disrupt Geo session were able to hear perspectives from private sector and civil society sides, which was quite refreshing and is something I would like to see more of in the future.”

The AAG annual meeting has been held every year since the association’s founding in 1904. This year’s conference included more than 9,000 attendees.

If you have thoughts or questions about this article, get in touch with Drew Bush, Geothink’s digital journalist, at drew.bush@mail.mcgill.ca. We also want to thank Victoria Fast for her willingness to share photos from the 2016 AAG Annual Meeting.

Please find an abstract for the presentation mentioned in this article below.

Leveraging Sensor Networks to Study Human Spatial Behavior

In the past decade society has entered a technological period characterized by mobile and smart computing that supports input and processing from users, services, and numerous sensors. The smartphones that most of us carry in our pockets offer the ability to integrate input from sensors monitoring various external and internal sources (e.g., accelerometer, magnetometer, microphone, GPS, wireless internet, Bluetooth). These relatively raw inputs are processed on the phones to provide us with a seemingly unlimited number of applications. Furthermore, these raw inputs can be integrated and processed in ways that can offer novel representations of human behavior, both dissagregate and aggregate. As a result, new opportunities to examine and better understand human spatial behaviour are available. An application we report here involved monitoring of a group of people over an extended period of time. Monitoring is timed at relatively tightly spaced intervals (every 2 minutes). Such a research setting lends itself to both planned and natural experiments; the later of which emerge as a result of the regular and on going nature of data collection. We will report on both a natural experiment  and planned observations resulting from 3 separate implementations of our smartphone based observations. The natural experiment that emerged in the context of our most recent month-long monitoring study of 28 participants using mobile phone-based ubiquitous sensor monitoring will be our focus, but will be contextualized with related patterns from earlier studies. The implications for public health and transportation planning are discussed.

Using Data to Revolutionize How We Make Decisions

Community members taking part in a planning process as part of Robert Goodspeed's doctorl work in Dripping Springs, Texas.

Robert Goodspeed, assistant professor of Urban Planning at the University of Michigan’s Taubman College of Architecture and Urban Planning, examined how decision support systems could be applied to urban planning processes during his doctoral work. This photo is of one such process in Dripping Springs, Texas.

By Drew Bush

The decision support market, a segment of the healthcare industry, made financial headlines when estimation of its global value by 2019 reached USD 239 billion, a jump of almost 38 billion since 2012. According to a new report, major players in the industry have poured money into new technologies that can take advantage of big data.

Digital health initiatives like those led by Canada Health Infoway have led to the creation of a network of systems that securely connect and share health information. Decision Support Systems like this one utilize computer-based data to aid in individual decision-making by supplying a massive bank of previous cases that aid in choosing the most likely answer or predicting trends. Most consist of interactive computer-based systems that utilize data and models to solve problems requiring geographically or temporally dispersed information.

In healthcare, IBM’s Watson system has been leading the trend to improve decisions made by doctors. “Watson knows what tests are relevant to further characterize a particular patient condition and what tests are not,” the report states. “It is a great help to physicians to have an assistant that is able to have read the latest journal articles and is loaded with medical information to recommend what tests may be relevant in a particular situation.”

An estimated 30 percent of all costs incurred for healthcare delivery come from tests that are either of little value in a patient’s case or sometimes outright wrong, according to some reports. Like platforms offered in other industries, the decision support system engineered by IBM offers the promise of more nuanced testing to enable better decisions on which medical tests can be best applied to specific patient conditions.

Robert Goodspeed, assistant professor of Urban Planning at the University of Michigan’s Taubman College of Architecture and Urban Planning, studies decision support systems.

Robert Goodspeed, assistant professor of Urban Planning at the University of Michigan’s Taubman College of Architecture and Urban Planning, studies decision support systems.

Using decision support systems to analyze data and make better decisions has helped to improve processes in many industries. Geothink 2015 Summer Institute Instructor Robert Goodspeed, assistant professor of Urban Planning at the University of Michigan’s Taubman College of Architecture and Urban Planning, has studied this trend.

Although Goodspeed doesn’t work in healthcare, his research examines what he refers to as “planning support systems.” His work has looked at how we can use information technology to improve processes that engage community members in urban planning decisions. During his doctoral work, he created a process that allowed individuals to access information about their neighborhood and city to improve discussions.

This research involved community members placing stickers on maps to categorize specific areas for different land uses. This data was then transferred to digital form with one person entering the data as it was called out. Interactions such as this ensured entering the data could be reviewed by the group as a whole and reflected the ideas that they had discussed.

“The participants reported learning quite a bit and I could observe their plans evolving,” Goodspeed said. “So that’s just one example of the sorts of tools and practices that I think or feel we need. Especially as we’re facing issues like climate change where we want to quantify things and create indicators, and know how the plans we are creating are going to do or how they’ll perform against these different indicators.”

The Varied Uses of Decision/Planning Support Systems

In more recent research, Goodspeed has taken his work with planning support systems and applied it to improve environmental-decision-making processes surrounding North America’s Great Lakes ecosystems. Work he’s done as part of the Great Lakes Aquatic Habitat Framework project have used GIS datasets to examine aquatic habitats such as streams, rivers, and lakes in the region. The process also supplies a “big pile of data” for decision-makers in the fisheries and environmental management departments in Canada and the United States.

Unlike in planning where professional tasks follow a somewhat structured process, ecosystem-based management systems must consider a whole variety of information and tasks, Goodspeed said. Work in the project has included leading participatory design workshops for professionals north and south of the border to aid in the development of a tool that will one day allow easy digital examination off all the information on the Great Lakes collected for the project.

Community participation in planning processes that help to envision the possible future often result in a final product that’s inherently more understandable, Goodspeed added.

“And really it requires that kind of combination of creativity but being specific about what you think will happen and what you think will work,” he said of his work with decision-support systems. As big data is increasingly used to inform decision-making, this trend will only continue to grow beyond the industries of healthcare and environmental planning.

Tweet him @rgoodspeed.

If you have thoughts or questions about this article, get in touch with Drew Bush, Geothink’s digital journalist, at drew.bush@mail.mcgill.ca.