1 00:00:02,040 --> 00:00:07,080 Speaker 1: Welcome to brain Stuff from How Stuff Works, Hey, brain Stuff. 2 00:00:07,120 --> 00:00:11,920 Speaker 1: Lauren Vogel bomb here in Los Angeles completed a thirty year, 3 00:00:12,160 --> 00:00:15,840 Speaker 1: four hundred million dollar project to synchronize four thousand, five 4 00:00:15,880 --> 00:00:19,400 Speaker 1: hundred traffic lights across the city. By deploying a vast 5 00:00:19,440 --> 00:00:23,360 Speaker 1: network of magnetic sensors installed under roadways plus hundreds of cameras, 6 00:00:23,520 --> 00:00:26,720 Speaker 1: the city can now capture real time traffic data using 7 00:00:26,760 --> 00:00:30,280 Speaker 1: a centralized computer system to synchronize green lights in order 8 00:00:30,320 --> 00:00:34,800 Speaker 1: to reduce LA's notorious congestion and rush hour gridlock. Soon 9 00:00:34,880 --> 00:00:38,320 Speaker 1: after the Los Angeles system was completed, city engineers cheerfully 10 00:00:38,320 --> 00:00:40,879 Speaker 1: reported that the average speed of traffic was up sixteen 11 00:00:40,960 --> 00:00:43,080 Speaker 1: percent and that the time it took to drive five 12 00:00:43,120 --> 00:00:45,720 Speaker 1: miles that's about eight kilometers on l A city streets 13 00:00:45,920 --> 00:00:48,840 Speaker 1: was down from twenty minutes to seventeen point two minutes. 14 00:00:49,440 --> 00:00:53,360 Speaker 1: And yet in l A still held the number one 15 00:00:53,400 --> 00:00:56,640 Speaker 1: spot for most traffic congested city in the United States. 16 00:00:57,640 --> 00:01:00,279 Speaker 1: So why didn't sinking the lights make more a dent 17 00:01:00,360 --> 00:01:03,680 Speaker 1: in the congestion problem? We spoke with Tim Lomax, a 18 00:01:03,760 --> 00:01:06,680 Speaker 1: research fellow with the Texas and m Transportation Institute and 19 00:01:06,760 --> 00:01:10,319 Speaker 1: a thirty five year veteran of traffic analysis. He says 20 00:01:10,360 --> 00:01:13,520 Speaker 1: that adaptive traffic signals are absolutely a good thing, but 21 00:01:13,560 --> 00:01:16,880 Speaker 1: they are not the quick, fixed panacea that frustrated commuters 22 00:01:16,880 --> 00:01:20,320 Speaker 1: believe them to be. Sinct traffic signals work best when 23 00:01:20,360 --> 00:01:22,600 Speaker 1: there is a clear and predictable flow of traffic in 24 00:01:22,680 --> 00:01:26,720 Speaker 1: one direction, explains lomax. I. Think of the traditional commuter scenario, 25 00:01:26,800 --> 00:01:29,440 Speaker 1: in which traffic flows primarily from the suburbs into the 26 00:01:29,480 --> 00:01:32,560 Speaker 1: city center in the morning and reverses course in the evening. 27 00:01:33,400 --> 00:01:35,600 Speaker 1: La Max said that in that kind of scenario, it's 28 00:01:35,600 --> 00:01:38,720 Speaker 1: easy to design a fairly well functioning traffic signal system 29 00:01:38,720 --> 00:01:41,000 Speaker 1: because it's clear which side of the road should get 30 00:01:41,040 --> 00:01:44,039 Speaker 1: longer green lights, and those lights can be timed with 31 00:01:44,080 --> 00:01:47,360 Speaker 1: the flow of traffic to keep cars moving. The problem 32 00:01:47,480 --> 00:01:50,120 Speaker 1: is that traffic patterns in the biggest US cities aren't 33 00:01:50,120 --> 00:01:53,280 Speaker 1: so cut and dried. La Max said, if you look 34 00:01:53,280 --> 00:01:56,240 Speaker 1: at the economic landscape now, there are jobs and populations 35 00:01:56,280 --> 00:01:59,680 Speaker 1: spread all over most metropolitan areas, and the highest volume 36 00:01:59,720 --> 00:02:02,320 Speaker 1: of geting could be from one suburb to another suburb. 37 00:02:02,760 --> 00:02:06,040 Speaker 1: All of your critical bottleneck intersections have heavy traffic coming 38 00:02:06,080 --> 00:02:10,079 Speaker 1: at them in both directions. So how does traffic signal 39 00:02:10,120 --> 00:02:13,000 Speaker 1: sinking work in the first place. A California's city of 40 00:02:13,040 --> 00:02:16,799 Speaker 1: Irvine explains the process like this. The city's traffic control 41 00:02:16,840 --> 00:02:19,360 Speaker 1: center calculates the arrival time for a group of cars 42 00:02:19,400 --> 00:02:21,880 Speaker 1: at each intersection, assuming the cars are traveling at a 43 00:02:21,919 --> 00:02:25,160 Speaker 1: certain speed, and then times the traffic signal to turn 44 00:02:25,240 --> 00:02:28,680 Speaker 1: green just as this group will hit the intersection. More 45 00:02:28,919 --> 00:02:31,560 Speaker 1: green time is given to a main street with greater 46 00:02:31,560 --> 00:02:35,600 Speaker 1: traffic volume than a side street with less volume. Synchronized 47 00:02:35,600 --> 00:02:37,919 Speaker 1: traffic lights don't mean that a driver will encounter an 48 00:02:37,960 --> 00:02:40,600 Speaker 1: unending number of green lights as she drives down a 49 00:02:40,600 --> 00:02:43,120 Speaker 1: major road, though. What it means is that all the 50 00:02:43,160 --> 00:02:45,480 Speaker 1: signals on a main road are set to run the 51 00:02:45,520 --> 00:02:48,160 Speaker 1: same cycle length. That is, the time that the signal 52 00:02:48,200 --> 00:02:50,400 Speaker 1: takes to go from green to yellow, to red and 53 00:02:50,480 --> 00:02:53,919 Speaker 1: back to green again. Ideally, the signal would turn green 54 00:02:54,000 --> 00:02:57,320 Speaker 1: again as the next group of cars arrives. Eli's traffic 55 00:02:57,320 --> 00:03:00,760 Speaker 1: system overhaul went further than most others by using cameras 56 00:03:00,760 --> 00:03:03,679 Speaker 1: and sensors to measure traffic flow and make real time 57 00:03:03,680 --> 00:03:07,640 Speaker 1: adjustments to keep traffic moving. Lomax says that poorly timed 58 00:03:07,639 --> 00:03:10,880 Speaker 1: traffic signals can cause significant delays and that cities can 59 00:03:10,919 --> 00:03:14,120 Speaker 1: make big games by updating signal timing every three years 60 00:03:14,160 --> 00:03:17,320 Speaker 1: to adjust for new traffic patterns. You might not completely 61 00:03:17,320 --> 00:03:19,960 Speaker 1: erase the difference between travel times during off hours and 62 00:03:20,040 --> 00:03:22,560 Speaker 1: peak hours, but you might reduce that difference by as 63 00:03:22,639 --> 00:03:28,120 Speaker 1: much as half. But retiming traffic lights costs money and manpower. 64 00:03:28,560 --> 00:03:31,560 Speaker 1: According to Texas A and M Research, cities should expect 65 00:03:31,560 --> 00:03:34,280 Speaker 1: to spend between three thousand, five hundred and four thousand 66 00:03:34,280 --> 00:03:37,800 Speaker 1: dollars per intersection and devote twenty to thirty work hours 67 00:03:37,840 --> 00:03:42,280 Speaker 1: for analyzing and re timing every signal. And jurisdiction issues 68 00:03:42,320 --> 00:03:44,880 Speaker 1: can complicate matters when these aim stretch of road passes 69 00:03:44,880 --> 00:03:48,200 Speaker 1: through different municipalities. What if one city or suburb doesn't 70 00:03:48,240 --> 00:03:50,480 Speaker 1: want to fork over money for the upgrade. That can 71 00:03:50,520 --> 00:03:54,040 Speaker 1: hold up re timing efforts for years. Here in Atlanta, 72 00:03:54,200 --> 00:03:57,160 Speaker 1: the fourth worst traffic city in the United States, the 73 00:03:57,240 --> 00:03:59,440 Speaker 1: last time that traffic lights were all SINCD was in 74 00:03:59,480 --> 00:04:02,280 Speaker 1: the midnight team seventies, when we had traffic lights at 75 00:04:02,280 --> 00:04:06,320 Speaker 1: only three hundred and twenty intersections citywide. As of eleven, 76 00:04:06,480 --> 00:04:08,600 Speaker 1: we had lights at a further nine hundred and forty 77 00:04:08,640 --> 00:04:13,040 Speaker 1: five intersections that were not included in the coordination system. 78 00:04:13,160 --> 00:04:16,040 Speaker 1: Retiming or sinking traffic lights is only one tool in 79 00:04:16,080 --> 00:04:20,320 Speaker 1: the traffic fighting toolbox. Lomax says that another effective strategy 80 00:04:20,400 --> 00:04:23,160 Speaker 1: is to clear away accidents and disabled cars more quickly. 81 00:04:23,600 --> 00:04:25,800 Speaker 1: He says the commuters aren't as miffed by long drive 82 00:04:25,839 --> 00:04:29,479 Speaker 1: times as they are by unpredictable drive times. What really 83 00:04:29,480 --> 00:04:31,880 Speaker 1: stresses people out is when the daily forty five minute 84 00:04:31,880 --> 00:04:35,560 Speaker 1: commute unexpectedly becomes an hour and a half commute, and 85 00:04:35,680 --> 00:04:38,840 Speaker 1: the most common causes of unexpected delays are accidents and 86 00:04:38,960 --> 00:04:43,080 Speaker 1: stalled cars. Oh and one more reason why synchronized traffic 87 00:04:43,160 --> 00:04:44,919 Speaker 1: lights may not help with traffic flow as much as 88 00:04:44,960 --> 00:04:48,520 Speaker 1: we'd like. Once people realize that congestion has improved on 89 00:04:48,560 --> 00:04:51,080 Speaker 1: a street, it encourages them to get in their cars 90 00:04:51,080 --> 00:04:53,760 Speaker 1: and drive on that street, thereby increasing the number of 91 00:04:53,800 --> 00:05:02,400 Speaker 1: cars on the road, which means more traffic. Today's episode 92 00:05:02,440 --> 00:05:04,640 Speaker 1: was written by Dave Ruse and produced by Tyler Clang. 93 00:05:04,920 --> 00:05:07,480 Speaker 1: For more on this and lots of other carefully timed topics, 94 00:05:07,600 --> 00:05:21,200 Speaker 1: visit our home planet, how Stuff Works dot com.