|
Where the rubber hits the road This gets a little complex: First, we add up the populations in each zone, dividing each building's capacity by the square of its distance from the center of the Student Greenhouse Field. These zone sums are assigned to the corresponding nodes. The traffic potential of each possible path -- in essence, its usefulness as a walkway -- is the product of its two node endpoints. To estimate actual traffic, we take these values and modify them with coefficients representing both path surface and path outlook. For path surface, tougher terrain merits a lower coefficient. For path outlook, we ask the question: As a pedestrian looks down a path, do they see another path continuing in the same direction? So, does our model actually work? In the graph linked to the left, our theoretical calculations are displayed alongside actual observations from November 27 and 29, 2001. The fit may be exaggerated because the empirical data was actually used to calibrate the coefficients for path terrain and path outlooks. Thus, the match between theory and reality on the Student Greenhouse Field is striking, but to have any real confidence in our model, we need another test. Any ideas? >> |