(Originally posted in the 2016 INFORMS blog).
A lot happens before the conference officially starts. Among those things, there is the combined colloquia, which are currently 3: for doctoral students, for young faculty, and on teaching effectiveness. I had such a great time last year at the doctoral students colloquium that I came back for the teaching effectiveness one this time.
Attendees of all colloquia are invited for a dinner on Friday, day -2. I had a great time talking to a cohort of PhD students and young faculty from US, Italy, Israel, and Hong Kong:
Had a great time at the combined colloquia dinner. Looking fwd to the teaching colloquium @INFORMS2016 #informs2016 pic.twitter.com/asZvjR0wZs
— Thiago Serra (@thserra) November 12, 2016
This morning, Jeffrey Camm was the first TEC speaker. He made interesting considerations about teaching students to explain why a solution is optimal to their clients. He also made the case for sub-optimal solutions, which sometimes are preferable because their worse value is more than compensated if they disrupt less how things are currently done.
Camm’s pitch for sub-optimality: you may be losing just a nickel to get a solution with less disruption #informs2016 pic.twitter.com/XzUUiLCMJ2
— Thiago Serra (@thserra) November 12, 2016
Fredrik Odegaard talked about his experience using and creating analytics-related cases. He mentioned good resources for cases and explained how we works with them through his 80-minute lectures.
Odegaard: there are many sources for #orms cases, including @INFORMS Transactions on Education #informs2016 pic.twitter.com/PoxdGOCbUC
— Thiago Serra (@thserra) November 12, 2016
Odegaard: typical time split to cover a case in a 80-minute class. Case-based exams take 4 hours! #TEC #informs2016 pic.twitter.com/QRqaBCoaJs
— Thiago Serra (@thserra) November 12, 2016
Matt Bailey was next. He pitched for ambiguity and unstructured problem solving, which is what will truly prepare students for real situations. That requires discipline not to delve in the data without a goal in mind. He worked out the probability for an, apparently, unusual case: a waitress at a small town in Ohio was given her stolen ID by an underage customer trying to buy alcohol.
Bailey: teach students to work backwards from the question they want to answer, not by staring at the initial data! #TEC #informs2016
— Thiago Serra (@thserra) November 12, 2016
Bailey estimates the probability of the “impossible”: a waitress given her own stolen ID at a bar #TEC #informs2016 pic.twitter.com/Vr7oNvZ7hl
— Thiago Serra (@thserra) November 12, 2016
Prakash Mirchandani discussed active learning and counterposed case-based lectures with games, which he prefers. He discussed when and for how to long to use different types of games along a course.
Prakash Mirchandani makes the “push-pull” case for active learning in the #orms classroom #TEC #informs2016 pic.twitter.com/hT2fWMpF7l
— Thiago Serra (@thserra) November 12, 2016
The next speaker was Amy Cohn, whom I have been following on Twitter for a while and seen her enthusiasm when talking about the Center for Healthcare Engineering and Patient Safety (CHEPS) at the University of Michigan. Her enthusiasm seems even bigger in person! She gave some interesting examples of engaging students by letting them choose project topics that they care about, and by replacing those tons of blending problems from linear programming textbooks with the real-world problems that she is currently solving.
First tweep at #TEC: @ProfAmyCohn talks about teaching outside of the classroom #informs2016 pic.twitter.com/yViX3CP4yS
— Thiago Serra (@thserra) November 12, 2016
.@ProfAmyCohn is telling great experiences from letting students do #orms projects in what they care about! #TEC #informs2016
— Thiago Serra (@thserra) November 12, 2016
Finally, James Cochran gave a whole lecture on engaging bits that can make students interested in the class. The most memorable part was when he explained how to use legos to explain linear programming and how to smash them when you want to explain shadow prices. He also run his version of “Who wants to be a millionaire”!
Cochran at #TEC: can you guess which of these Georges is considered the father of linear programming? #informs2016 pic.twitter.com/Lu4VZ9Hnk9
— Thiago Serra (@thserra) November 12, 2016
Kudos to Mihai Banciu for curating such an inspiring crowd of lecturers. The best way to teach is by showing, which all of them did!
Thiago Serra 