Everest and Animals and Everything Else

We all woke up bright and early Thursday morning for our backstage safari tour of Animal Kingdom.  The main function of the tour was to show us how Disney maintains all of the animals on property given that most of them in the wild would require miles of space to be able to live.  After slipping backstage, we saw how they manage each animal according to its species.   For example, giraffes have a specialized two-story building where they sleep at night, whereas monkeys have to live in a bunker-like structure because they are clever enough to escape the habitats of similar-size animals.  Each animal sleeps backstage in their enclosures rather than in the actual safari, and while they’re in the safari, Disney uses a combination of natural and artificial barriers to keep some animals separated from others.  For example, the lions are surrounded by a moat not visible to people on the Kilimanjaro Safari so that they are unable to reach the zebras.  Another such example is how they use cattle guards—pipes that require a dexterity level greater than that of cows in order to be crossed—to deter animals from leaving their section of the safari and entering that of another animal.

My (Jamey) personal highlight of the tour was being able to stand about five feet away from a 3,000 pound white rhino.  Probably the closest I’ll ever be to one (here’s hoping).  One thing I learned on the trip was that black rhinos have a problem with iron absorption.  They absorb so much of it that the iron concentration in their bloodstream can actually kill them when they are housed in zoos.  Disney circumvents this problem by giving them phlebotomies regularly, thus keeping their iron levels down and the rhinos healthy.  We also saw the skull of an American Lion, which I previously had not known to have existed, apparently the biggest of all the cats before going extinct approximately 11,000 years ago.

After the backstage tour, we began collecting data for ride output and average party size for Expedition Everest, and we also collected data on how many people Disney puts on each raft for the Kali River Rapids.  For Everest, what we’re doing is spacing ourselves out every ten people in the single rider line and then recording data on both the posted standby time, posted single-rider time, actual time in line, and then the average amount of people per party in the regular standby line, with hopes that we can eventually calculate the benefit of riding single-rider and also the throughput for Expedition Everest.  This, of course, required that we ride the roller coaster several times, but we put up with it in the name of statistical analysis.  We got lots of interesting looks from fellow riders as we craned our necks to see how many people were in each party and as we recorded information on our tablets and notebooks, and people inevitably asked us what we were doing.  Jordan got so tired of explaining the trip that he eventually resorted to telling people he was working for NASA or that he designed the ride.

For Kali River Rapids, we stood on the bridge overhanging the ride and counted how many people were seated in each raft so that we might eventually see how many people Disney is not fitting per ride in terms of capacity, then comparing the theoretical optimal output of the ride with the actual output which we observed.

We are making headway on the Magic Kingdom algorithm; before long, we should have it up and running and showing us the optimal touring plan for the park.  We have begun data collection on our new project of examining single rider lines and average throughput, but we still have (get) to collect data on Test Track, Rockin’ Roller Coaster, and Tower of Terror.  Today we made good progress; hopefully tomorrow we might notice some peculiar trend in the data and perhaps get some insight from it.  We’re waking up early tomorrow for our Epcot tour, tomorrow will be a busy day.  Oh yeah, and Stella almost died stepping out of the Backstage Safari van.

Jamey and Joey

(We apologize for this post being delayed!  Our internet WiFi went down yesterday and this morning and we had a full day at the parks!  This is the post from Thursday May 29.)

Deriving Solo

This is Hayden and Rahul for our final blog together!

Complicated Math

Today was probably our most intense purely Mathematics class thus far as Dr. Hutson gave us a statistics lesson. We started off in what was familiar territory for most, talking about linear regressions and lines of best fit. We got a little more technical as we talked about the interpretation of the slope-intercept form and how to make predictions using this information. And just like that we were using multivariable calculus to derive the formula.

Dr. Hutson did a great job of making the derivation very accessible to the entire class. We then moved onto an alternate derivation using Linear Algebra which was significantly more complicated (yes, what’s on the whiteboard is the easy version of the derivation). Although it was more convoluted, it had interesting implications on how to extend regression to non-linear functions.

Dr. Hutson derives the slope-intercept formula for Linear Regression

After this, we toned down the technical mathematics and moved back to the more applied math with a discussion on t-tests, a statistical hypothesis test. This led into our next side project: analyzing the FastPass+ system that Disney has recently put into place. The FastPass+ system was put into place ostensibly to reduce wait times both for guests who reserved FastPasses and for those who use the standby line by. Although clearly advantageous for users of the system, it remains up for debate if this actually reduces times for standby users.  Tomorrow we will see if this is indeed the case on two attractions in Magic Kingdom, Pirates of the Caribbean and Winnie the Pooh. We will split up in groups of two and record wait times for these ride at different times throughout the day tomorrow.

Project Update

The team in charge of measuring distances completed gathering and entering the data. The only hitch was Fantasyland, which having been redesigned recently did not have recent satellite images online to estimate distances. Tomorrow we will physically walk between attractions in this area of the park to gather this missing data. The algorithms team completed a program to find the shortest path between any two arbitrary attractions, given the above data. They also completed the design phase and began coding the Genetic Algorithm approach to solving Traveling Salesman Problem.

Animal Kingdom

Real Dung Beetle

This afternoon we went to the Animal Kingdom for our second time. As luck would have it, as soon as we entered the park it began to pour! We decided to make the trek to the It’s Tough to be a Bug show and wait out the rain there. After we were sufficiently scared by giant bugs and the feeling of insects crawling across our seats, the rain had stopped and we made our way to some of the outdoor attractions. We had to repeat favorites from our last visit, like Expedition Everest and the Kilimanjaro Safari. This time after we finished the Safari, we continued our animal encounter on the Pangani Forest Exploration Trail. Here we got an up-close view of Western Lowland Gorillas, Naked Mole Rats, Okapis, and many other animals native to Africa. Luckily the Dung Beetles in the Research Station were much smaller than those in It’s Tough to be a Bug!

Activity patterns of gorillas

We finished out our day at Animal Kingdom with a ride on the Kali River Rapids. After our last time on the ride some members of our class developed a theory that the heavier side of the raft is more likely to get wet… Well, after our ride today the theory seems to be true, and I will remember to always try and sit by an empty seat. On our way back from the park, the students decided they would try to play some of their own music, but Dr. Hutson didn’t seem to enjoy this. After about fifteen seconds of Miley Cyrus, he turned off the stereo. Next we tried “Ridin’ Solo” by Jason Derulo to which Dr. Hutson replied he wished he was “riding solo.” We weren’t able to please him until Jamey brought out some Pearl Jam.

Birthday Party

Today was William’s 2nd Birthday (Dr. Bouzarth’s son)! And we all know what birthdays mean -- birthday cake! Dr. Bouzarth was kind enough to share birthday cake with the class, which was greatly appreciated.

Happy Birthday William!

Need a Laugh?

While eating cake, we learned that in high school Joey made a Friday parody for his calculus class. You won’t regret watching this video, featuring our own Joey Iannetta rapping about derivatives:

https://www.youtube.com/watch?v=RAOVZ4zBiCU.

100% Animal Kingdom

Saturday, May 17

Today we visited Animal Kingdom, arriving at the park just at before it opened. We opened with Expedition Everest, then went to Expedition Everest, then tried Expedition Everest again. Needless to say, definitely a good intro to the day. After a wakeup call from the yeti, we migrated over to the Safari where we saw some of the multitude of animals Disney maintains at the park, including hippos, rhinos, giraffes, crocodiles (Stella's favorite), and some lively flamingos, all of which were 100% real.

We then went to see Finding Nemo the Musical, a fantastic retelling of the movie considering it was only an hourlong show, then split up in the park for the rest if the day until three. My group ended up riding Dinosaur then Kali River Rapids, both of which were a lot of fun despite the long wait time at the river rapids. Lessons learned from the day trip included that flamingos turn pink from the shrimp they eat, black rhino's horns are made out of keratin like our finger nails, elephants' pregnancies last up to 23 months, and that deforestation is a major cause of flash flooding.

 After we all had a chance to catch a quick nap and freshen up, we met in the professors' room for a continuation on the lectures started on Thursday.  We were first introduced to the Knapsack problem, a complex problem that very much relates to what we've been talking about with Graphs.  Here is an example of the knapsack problem.  Imagine you are packing for a long trip, say six months, but you are flying.  The airline you are using will only allow 2 bags.  You have to look at the weight and size of each item and decide which items you will bring.  It might sound pretty easy, but coming up with an algorithm to achieve this task has yet to be done. 

Dr. Hutson expanded upon his lecture on Linear programming by showing us the beginning basics of how to solve these problems.  He did a really good job of wrapping up a few courses in vectors, matrices, and operations research in half a day!  It was really interesting to see a computational way to  see how to get the answers that the computer generates for you.  These answers that he showed us gave very precise solutions which can be good, but not always the ideal answer.  For example, if you were looking to get a precise answer on how many of those items you could fit in your bag, you wouldn't want the answer to be 12.243.  It would have to be an integer solution.  This is where things got a little more technical.  This, however, was really cool for a lot of students because we now got to see application of concepts we learned about in previous math classes.  For example, in MTH-160, we learned about null spaces and today we learned how those can apply to solve solutions to our linear programming problems.  This transitioned very nicely into continuing Dr. Harris' lecture where we began to take an in-depth look at the very famous Traveling Salesman Problem.

This problem is really the focus of our studies right now.  The traveling salesman problem (TSP) asks the following question: given a list of cities and the distances between each pair of cities, what is the shortest possible route that visits each city exactly once and returns to the original city?  We applied this problem to Epcot yesterday when we competed to go to the maximum number of rides in the shortest amount of time. Today we learned some algorithms to help assist solving TSP, but ultimately we discovered there is no algorithm to solve this problem. 

Tomorrow looks like we'll finally get a little rest from all the parks.