Aspen Elementary (18)/Interim Report

Interim Report
http://mode.lanl.k12.nm.us/get_interim1112.php?team_id=18

Definition of the Problem
Heavily trafficked classical cross-intersections (stoplight) present some disadvantages, for example: they are prone to generate accidents between cars or also involving pedestrians which often have a high rate of fatal injuries; cross-intersections with heavy traffic can generate delays; the stoplights interrupt the traffic flow, and as a consequence of this cars use extra gas while stopping, waiting, and then reaccelerating. Roundabouts represent an interesting alternative to try to address most of these problems. The focus of our project is to try to establish, by means of computer simulations, the roundabout as an improved alternative to the classical cross-intersection.

Plans for solving the problem computationally
We are using the program Netlogo, recommended by one of our teammates for the simulator programming environment. The reason for choosing this tool as it is one of the easiest programming environments to use for this kind of simulation. We are planning to attempt the roundabout alternative first, and then the classical cross-section if there is enough time. We think this program is appropriate for this type of problem because it is an agent-based language. In addition, some of our teammates are already familiar with it.

Progress up to date
In the first part of the project we did some research on the benefits of existing roundabouts across the U.S. and also in other countries. Some of these benefits are: a) roundabouts have proven to be safer than intersections; b) the introduction of roundabouts results in a reduction of overall collisions; c) injury resulting collisions are less frequent; d) the fatality collisions are greatly reduced, as well as the pedestrian collisions (Ref. 1-12).

Our work will concentrate on simulating the intersection of Trinity and Diamond roads in Los Alamos, NM. In order to obtain a rough estimate of how many cars go through this intersection during peak hours we performed a car counting exercise for approximately 30 minutes. The data obtained from this survey will be used as an input for our simulations. We did some research in a reference book on urban planning. We went to state sponsored websites and news articles to research roundabouts.

In an effort to gather more information about the intersection under study we interviewed a Los Alamos County traffic engineer. He talked about traffic at peak hours and how to obtain a rough estimate of how many cars go through a given intersection. He also showed us a computer simulation of an intersection and he explained the most commonly used car counting devices ranging from the simplest ones to the most advanced. He gave us different results from the different devices and the data that the car counting machines obtained for the same intersection that we surveyed by hand. We studied this data and will be using it as raw data to base our agent based simulation code on.

Expected Results
The results that we are expecting to obtain are that roundabouts are better than classical cross-sections for these reasons: Roundabout's long term costs are cheaper than cross-sections because cross-section have maintenance costs associated them. In addition roundabouts have a life span that is about two times longer than classical cross-sections. Also, studies have proven that roundabouts decrease the number of crashes, favor the traffic flow significantly, and as a result cars use less gas. So we hope to see a better traffic flow and less amount of gas used for the roundabout alternative.

Introduction
Hello Team 17! My name is Ruben Guadiana and I've been involved with the Supercomputing Challenge for about ten years. I started out in High School as a participant for two years and then helped with mentoring and judge through college and now that I've started my career. I currently work in Albuquerque for the US District Court as a Network Administrator.

Progress
I think your project idea is really great. It's awesome that you've taken a local problem and aim to solve it using your community's unique resource! Your research so far is superb, more than most teams I've looked at and that's a huge asset for your project. If possible be sure and type of your conversations with the different professionals for use as a reference. Come final judging, the judges will appreciate the time you took to meet people and gather your statistics by hand! Now, it's time to get busy on the programming itself. You'll want to have as much completed by the face to face review scheduled for later this month.

Mentors
It's always a good idea to have mentors for the SC Challenge. Your report doesn't specifically mention any mentors but it does sound like you've met a few professionals in this field that you could enlist as mentors. The person that you've interviewed would probably make a great mentor especially once you get your program running, you'll be able to discuss it's effectiveness with someone who looks at these things for a living. If you need some help with the math model or programming piece of the project, take a look at the list at http://www.supercomputingchallenge.org/resources/mentors.shtml and see if you can find someone to help.

Model
Part of this evaluation focuses on your math model and actual source code. With all the research you've done I'm sure you'll be able to come up with a great model. In each phase of presentation it's important to communicate everything you've accomplished. Make sure by to have a math model created and hopefully as much code as possible for display at the face to face evaluation for the best possible feedback from the judges. This can help you prepare for the final presentation as well.

Face to Face Evaluation
Later if February you've got another review coming up and it will be important to have actual source code to show to the judges. Work hard for the next few weeks to have as much as possible to present to the judges for the best feedback and direction as possible.

To better prepare you for the upcoming review I wanted to share with you the judging rubric which will be used at the final judging. Review it throughout your preparation to understand what the judges will be looking for. http://www.supercomputingchallenge.org/archive/05-06/expo/poster_session_rubric.html Presenting your project as if it were final in the upcoming interim review will better prepare you for the final judging in April. If you have any questions or comments please email me at rubane@gmail.com

Very well done so far, I look forward to seeing your final product!