Traffic Engineering

Bay Area Traffic Decoded: What cell phone and GPS data reveals about traffic patterns

580 and I80 Traffic Jam

A new article from researchers at MIT and UC Berkeley uses data from cell phones and GPS to track traffic patterns. "Understanding Road Usage Patterns in Urban Areas" from December's Scientific Reports assess how drivers from certain areas effect the whole network. 

 We find that the major usage of each road segment can be traced to its own - surprisingly few - driver sources. Based on this finding we propose a network of road usage by defining a bipartite network framework, demonstrating that in contrast to traditional approaches, which define road importance solely by topological measures, the role of a road segment depends on both: its betweeness and its degree in the road usage network. Moreover, our ability to pinpoint the few driver sources contributing to the major traffic flow allows us to create a strategy that achieves a significant reduction of the travel time across the entire road system, compared to a benchmark approach.

Drivers from Sanjose, Hayward, Dublin, San Rafel and San Ramon often find themselves stuck in the worst traffic. Could more metering be the answer?

HCM 2010 now available online!


Need to reference something in the most recent Highway Capacity Manual but can't make it over to the library? Well now UC Berkeley students, faculty and staff can access it online through Knovel. If you're off campus you'll need to use the proxy server to access it.

Friday Seminar - Robert Campbell on Failure to Yield: A Framework for Evaluation of Compliance Measures

Traffic Circle Trails

This week's Friday TRANSOC Seminar features Robert Campbell, Ph.D. candidate, University of California, Berkeley, presenting "Failure to Yield: A Framework for Evaluation of Compliance Measures."

In this presentation, we explore the safety problem of drivers violating yield signs in a freeway context. Drivers violating traffic controls is a common problem, and agencies have a collection of strategies that they often use to address the problem. These include increasing the size of the traffic control, adding an advanced warning upstream, installing pavement markings, or using LEDs to capture drivers' attention. As common as these (and other) compliance measures are, however, no research has been done to properly evaluate how they compare in terms of effectiveness.

In cooperation with Caltrans, two different compliance strategies—increasing the size of the control, and adding supplemental pavement markings—were implemented on Interstate 10 in Los Angeles so that the effectiveness of each could be measured in the field. Although our experiments involve yield sign violations, the insights obtained can be adapted to other contexts as well, such as pedestrian crosswalks or turn prohibitions at intersections.

We will explore the outcomes of these two experiments and, using a method we have developed to allow for responsible comparisons of effectiveness, will come to conclusions about the performance of each one. Our analysis offers insights into the mechanisms behind the observed behavioral response that occurs in drivers over time after a strategy is implemented (including what happens in the often-ignored "unstable" or "novelty" phase), which we can then use to inform our assessments of each compliance strategy. Our results will reveal flaws with the conventional before-and-after approach used to evaluate compliance measures, and will show how such errors can be avoided or corrected.

 The seminar will be held in 534 Davis Hall from 4:00-5:00 on Friday, April 27. Please join us for a TRANSOC-sponsored Cookie Hour in the ITS Library, 412 McLaughlin, from 3:30-4:00.


Friday Seminar: Anurag Pande presents "Traffic crash patterns: What can we learn from retailers?"


This week's TRANSOC Friday Seminar features Anurag Pande presenting, "Traffic crash patterns: What can we learn from retailers?"

 Data mining applications are becoming increasingly popular for many applications across a set of very divergent fields. Analysis of crash data is no exception. Association analysis or market basket analysis is used by retailers all over the world to determine which items are purchased together by consumers. It is then applied to stock items (e.g., Salsa and Chips) close to each other. In traffic safety research based on association rule mining, crashes are analyzed as supermarket transactions to detect interdependence among crash characteristics. The results from the analysis include simple rules that indicate which crash characteristics are associated with each other. Results will be presented from two of research articles in which this application is demonstrated using crash data from the state of Florida.

The seminar will be at 4:00pm in 534 Davis Hall as usual. Don't miss Cookie Hour in the library at 3:30!


Friday Seminar: Dr. Venky Shankar on Considerations Regarding Design Variations on Safety

4th & Howard

Today's Friday Seminar from TRANSOC features Dr Venky Shanker P.E. presenting "Considerations Regarding Design Variations on Safety."

This talk will present some thoughts on the consideration of design variations in the analysis of network safety, with a focus on strategic guidance on probing the effects of design variations in depth. Strategic guidance will be described through example models estimated empirically using statistical methods. The talk reflects ongoing research activity and integrates insights from the speaker's prior experience in decision-making in design policy matters at state governmental levels.

The seminar will be held at 4:00pm in 534 Davis Hall. Don't forget about Cookie Hour in the library at 3:30! See you here.

More on PPPs and Road Financing

PA Turnpike tilt-shift

Yesterday we talked about Britain's proposed privatization of their transport infrastructre and made an error when we said the Pennsylvania Turnpike was leased to a private company. In 2007 bidding was opened on the Turnpike and in 2008 the highest bid was received from Spanish firm Abertis Infraestructuras, but ultimately the plan failed. Currently the Turnpike is managed by a state-operated commision that "receives no state or federal taxes to operate and maintain its toll road system." The Pew report, Driven by Dollars, outlines several of the problems that were a part of the leasing proposal and calls for a more open process and transparency.

As funding sources dry up, such as the transportation bill now stuck in gridlock on Capitol Hill or depleted state budgets, transportation agencies will have to come up with new methods of financing which has an increasing interest on private money. The 2011 book Road to Renewal examines private investment in transportation projects from around the world, outlining what works and what doesnt for PPPs as well as how to protect public interests. Louise Nelson Dyble has a recently published article "Tolls and Control: The Chicago Skyway and the Pennsylvania Turnpike" which compares the two plans and raises questions about impact on future transportation planning policy.



Privatizing Infrastructure: Leasing Toll Roads

M6 J7 Fireworks

This week Britian's Prime Minister David Cameron gave a speech on infrastructure. He touched upon many different industries and modes, but this is what he said about highways:

Now, road tolling is one option, but we are only considering this for new, not existing capacity.  For example, we’re looking at how improvements to the A14 could be part-funded through tolling.  But we now need to be more ambitious.  We should be asking ourselves, ‘Why is it that other infrastructure’ — for example, water — ‘is funded by private sector capital through privately owned, independently regulated utilities, but roads in Britain still call on the public finances for funding?’ We need to look urgently at the options for getting large-scale private investment into the national roads network; from sovereign wealth funds, from pension funds, from other investors.  That is why I’ve asked the Department for Transport and the Treasury to carry out a feasibility study of new ownership and financing models for the national roads system and to report progress to me in the autumn.  Let me be clear: this is not about mass tolling and, as I’ve said, we’re not tolling existing roads; it’s about getting more out of the money that motorists already pay.

People are already panicking about China owning the motorways, though the BBC does have a nice Q&A about public private partnerships and toll roads. There is also a focus on "shovel-ready" projects, which is apparently concept in the UK. The most famous example of a privitized toll road in Britian currently is the M6 north of Birmingham, which opened in 2003, was proposed by John Major when he was Prime Minister, and is regarded as a mixed success.

These sorts of public private partnerships (PPPs) are quite common in the US. Edit: There was an attempt to lease the Pennsylvania Turnpike, but that failed with lots of lessons learned. The Pew Center on the States wrote an overview of what states and agencies should consider when entering these PPPs. It will be interesting to see how this plays out.

Friday Seminar – Eleni Christofa on Traffic Signal Optimization with Transit Priority

light rail tracks and wires, October 23, 2005

This Friday’s TRANSOC Seminar has Eleni Christofa, Ph.D. Candidate, University of California, Berkeley, presenting "Traffic Signal Optimization with Transit Priority."

Traffic responsive control with transit signal priority (TSP) is a strategy that is increasingly used to improve transit operations in urban networks. However, none of the existing real-time signal control systems has explicitly incorporated the passenger occupancy of transit vehicles in granting priority, or has effectively address issues such as the provision of priority to transit vehicles traveling in conflicting directions at signalized intersections. A person-based traffic responsive signal control system with TSP is presented that provides priority to transit vehicles while minimizing the negative impacts on the auto traffic even when transit vehicles travel in conflicting directions. The objective is to minimize the total person delay at the intersection by explicitly considering the vehicles’ occupancy and schedule adherence. Such a system has been made feasible by advanced technologies which provide real-time information such as Automated Vehicle Location systems and passenger counters.

The proposed traffic responsive signal control system was first developed for isolated intersections, and extended to arterial signalized networks.  Evaluation tests for a wide range of traffic and transit operational characteristics show that the proposed system can achieve substantial reductions in transit delays with no significant increase in auto delays, and can outperform signal settings provided by commonly used optimization tools that minimize vehicle delays. The contribution of this research is the development of readily implementable strategies that take advantage of existing infrastructure to improve transit and traffic operations in congested metropolitan areas.

The seminar will be held Friday, March 16, from 4:00-5:00 p.m. in 534 Davis Hall. Please join us for a TRANSOC-sponsored Cookie Hour in the ITS Library from 3:30-4:00 p.m.

Friday Seminar - Vikash Gayah on The Aggregate Effect of Turns on Urban Traffic Networks

overland ave traffic

This week’s Friday TRANSOC Seminar has Vikash Gayah, Ph.D. candidate, University of California, Berkeley, presenting “The Aggregate Effect of Turns on Urban Traffic Networks.” 

This research creates and uses macroscopic traffic models to describe the aggregate behavior of vehicles on urban street networks. Insights gained from these models can then be used to design network-wide policies that may increase the ability of these networks to serve vehicle-trips. In particular, this work focuses on the turning maneuvers that exist in networks with multiple routes. The presence of multiple routes and turning maneuvers are found to have two effects on aggregate vehicle behavior: 1) they cause unstable and inefficient behavior when a network is congested; and, 2) they may reduce maximum vehicle flows across the network. Fortunately, this work finds that limiting the rate at which vehicles are allowed to enter a network and providing drivers with real-time information on current traffic conditions can help mitigate the first effect and allow the network to operate more efficiently. It is also found that the second effect may not always be harmful—lower network flows do not necessarily result in decreased network efficiency if the lower flows are accompanied by more direct vehicle routing. In fact, two-way networks, which accommodate conflicting left-turns and result in lower maximum vehicle flows than one-way networks, are found to serve trips at a higher rate because drivers travel shorter distances on average. Thus, in many cities, maximum network efficiency can be improved by converting one-way streets to two-way operation.

The seminar will take place at 4:00 PM in534 Davis Hall. Please join us for a TRANSOC-sponsored Cookie Hour in the ITS Library at 3:30 PM.

Friday Seminar - Lily Elefteriadou on Driver Behavior and Characteristics and Their Use in Traffic Modeling

Driver Experience @ Limeira

This week's Friday TRANSOC Seminar has Lily Elefteriadou, Professor and Director of the Transportation Research Center, School for Sustainable Infrastructure and the Environment, University of Florida, presenting “Driver Behavior and Characteristics and Their Use in Traffic Modeling.” 

Traffic modeling has frequently considered and accounted for variability in driver behavior and characteristics.  For example, microscopic traffic simulators have the capability to replicate vehicular movements (such as lane changing) considering driver characteristics to a significant level of detail.  Such traffic simulators can typically replicate traffic streams with several different driver types which are based on driver aggressiveness.  Vehicular movements (such as car following) are then determined based on the respective action of the particular driver type.  However, a limited amount of research has been reported to categorize driver types or to link particular driving actions with a set of driver types and their characteristics.  Car-following, lane changing, and gap acceptance algorithms have rarely been calibrated to match various driver types, and it is not always clear how micro-simulators incorporate driver behavior aspects into these algorithms.  This presentation will describe two approaches to collecting driver behavior and characteristics-related data so that they can be used to improve traffic micro-simulators.  The first approach is based on focus groups, while the second is based on in-vehicle field data collection with an instrumented vehicle.  The presentation will describe these two data collection approaches and will provide three example applications related to freeway merging, car-following, and arterial lane changing. 

Syndicate content