Reported network coding issues - Chicago Regional model


Wolfgang Scherr from PTV America:

1. There seem to be discrepancies between "speed" and "free flow travel time" in certain links of the Chicago Regional network. For example, the "Chicago Skyway" (nodes: 8158, 7714, 8184, 8190, 8196, 8246, 8208, ...,2049, 2387) has "speed" of 27MPH while the free flow travel times correspond to 45MPH. As indicated above, the results reported here rely on free flow travel times and not on speeds.

2. There are some bi-directional links which in reality represent mono-directional ramps. The current bi-directional coding creates parallel alternatives to get on or off the freeway/expressway, which leads to a kind of local randomness of the link flow solution at these places. I would recommend coding all ramps as one-ways. Here are a couple of such places, defined by key node numbers:

A) 6652, 2430, 2429
B) 8559, 8563
C) 2196, 2301
D) 7138, 8994
E) 10265, 10253, 11946