Urban Congestion

Data Skeptic

b'## Urban Congestion\r\n\r\nSegment Intro: Data Skeptic features interviews with experts on topics related to data science, all through the eye of scientific skepticism. \r\n\r\nHOST: Lewis Lee is a Ph.D. candidate in civil and environmental engineering with a focus on transportation engineering.\r\nHis data visualization work has helped people understand complex topics like why public transportation buses seem to arrive in clumps, and his published research spans topics such as land use planning and congestion pricing. It\'s the topic of congestion pricing that I specifically invited him on to discuss.\r\nLewis, welcome to Data Skeptic.\r\n\r\nLEWIS: Thanks for having me, Kyle.\r\n\r\nHOST: Maybe to start with, could you give us a definition of what exactly is congestion?\r\n\r\nLEWIS: It\'s kind of a cluster concept, I think. It\'s kind of more like you know when you see it, but I guess in the most general way it\'s conditions where a road is much below its free-flow speed while the cars are actually impeding each other\'s speed.\r\n\r\nWithin congestion, there is kind of two types of congestion, or regimes. One is light congestion \x97 or what some engineers just call congestion \x97 which is the density, the number of cars per, let\'s say, kilometer as it rises. The speed falls, of course, but the overall flow, which is what you\'ve measured are going past you, increases as the density of cars increases. So that would be like the roads are just getting more clouded but more people are moving. And you\'ve got another type of congestion, which is sometimes called heavy congestion, or hyper congestion, which denotes that it\'s more extreme where actually less cars are moving the more crowded it gets.\r\nAnd hyper congestion is what you see usually in networks, like in a downtown. There are more cars on the road, but less people are getting where they are going.\r\n\r\nHOST: Is there a well-understood science for a why that\'s the case? I feel like everyone\'s been in a traffic jam, and maybe we all have our pet

Next Episodes

Data Skeptic

Heteroskedasticity @ Data Skeptic

📆 2016-09-09 02:00


Data Skeptic

Front End and API Overview @ Data Skeptic

📆 2016-09-02 02:00


Data Skeptic

Music21 @ Data Skeptic

📆 2016-09-02 02:00


Data Skeptic

Paxos @ Data Skeptic

📆 2016-08-26 02:00