This code estimates the generalized continuous-discrete model with a t-distributed error kernel. It also replicates the Monte Carlo study of this paper. If you use any part of this code in any form, please cite this paper.
R code for producing the results in "Asymptotic Approximations of Transient Behaviour for Day-to-Day Traffic Models" by Watling & Hazelton. The file code.r contains functions; examples.r instances of their use. The other files are used solely for plotting the Leicester network.
This is no data, but the matlab code (and it's not that much) with which all experiments in the paper have been conducted. Insofar possible, it is reasonably well annotated. Feel free to use in whatever way you want (replay some of the experiments, try new things, etc) for research and educational purposes. If you see a way to make money with it let me know (I could use a PhD working on the subject ;-)
Contributors:Peng Sun, Mike Hewitt, Lucas Veelenturf, T. Van Woensel
This dataset contains 32 instances of the time dependent pick up and delivery problem with 10 to 45 customer request (20 to 90 nodes).
the first row is number of vehicles
the second row is vehicle's capacity
the third row is planning time horizon
idx; x-coordinate; y-coordinate; 0; 0;
idx; x-coordinate; y-coordinate; 0; load; profit; earliest time window; latest time window.
idx; x-coordinate; y-coordinate; 0; load; 20; earliest time window; latest time window.
Five different speed profile;
[speed choose matrix]
For each links, it has a speed profile. For instances, the link (0,4) is assigned with speed profile 1, which is given by [Speed 1].