warren@ihnss.UUCP (07/16/83)
Since everyone had so much fun with the physics of freeway traffic, here are three more little mysteries of auto traffic. This time the question is why these systems are designed the way they are, which in each case would seem to lead to more congestion and confusion: 1) Parking lot funnels. I have noticed increasingly that parking lots for large stores and shopping centers seem to be designed to make it difficult to park without first passing in front of the store (between the parking area and the entrance.) In some in illinois, you must pass through this area at least twice, once to get in and once to get out. This design maximizes the chance for vehicle-pedestrian collisions, encourage parking illegally near the store, and generally increase congestion. Why not instead have no roadway in front of the stores, only turn-around space for going from one parking aisle to the next? 2) Advanced green lights. I used to see a fair number of traffic lights with cycles that gave left-turn cycles to cars waiting at the end of the normal green cycle. Now, I rarely see these things that do not give dedicated left-turn cycles if a car is waiting at the beginning of the cycle. The difference is that if traffic conditions are light enough to permit the left-turners to get through the normal cycle, then no special cycle is needed. The new lights, however, give them one anyway, since the decision is made at the start of the cycle. Putting the left-turn cycle at the end of the normal green cycle would seem to me to reduce congestion (by increasing the proportion of time given to through traffic) and would avoid the problem caused by several cars trying to make left turns after the light turns I don't see why these lights are designed the way they are. 3) Tripped lights. In several areas of my town, 2-way stop signs have been replaced by tripped traffic lights. The effect has not been positive. On these streets, during most of the day traffic on the side street could escape within 30 seconds. With the tripped lights, traffic must wait for a light cycle, frequently a long time, which then stops all traffic on the main street for a considerable period of time. Why doesn't anyone set these things as flashing red for the side street, permitting cars to exit, and trip to stop the traffic on the main street only after a timeout of 30 seconds or so? It seems like a much better solution. Feel free to ignore these questions, but I find them more interesting than yet another bags vs belts debate, more on radar detectors, the 55mph limit, big vs small cars, and any of the numerous other constantly recycled topics. -- Warren Montgomery ihnss!warren IH x2494
crc@clyde.UUCP (07/18/83)
I agree that sensing lights would be better than the mindless ones we usually encounter. Some lights in New Jersey (Dioxin State), are this way. Some of them are magnetic As one would expect, they are designed for 4000 lb american cars. Some are so bad they don't switch for motor cycles. One, in Wayne NJ doesn't switch for my TVR. There are only 800 lbs of iron in the car. (Wayne doesn't plow in the winter until there are two inches of snow on the ground.) Jerks can ruin anything. yours for intelligence in government and other fantasies. Charles Colbert
chris@umcp-cs.UUCP (07/20/83)
Concerning lights with malfunctioning magnetic sensors: Right On!!
There's this one particular light, at the main entrance to the College
Park campus of the U of MD, that seems to work exactly backwards. I
often leave here at 2 or 3 in the morning. I get to the light; it's
red. There's no one nearby on the main street, Route 1. (However,
there're headlights far down the road so I don't dare make a
left-on-red.) So I sit at this red light for a while. Eventually,
someone comes down Route 1. Just as they approach the light, it turns
red for Rt. 1 traffic and green for me! If it let me through 60
seconds ago no one would have had to even slow down!
- Chris
--
In-Real-Life: Chris Torek, Univ of MD Comp Sci
UUCP: {seismo,allegra,brl-bmd}!umcp-cs!chris
CSNet: chris@umcp-cs
ARPA: chris.umcp-cs@UDel-Relay