Abstract: The development and implementation of an ongoing HOV lane monitoring and evaluation program is important to measure progress toward meeting the goals and objectives of the project. Monitoring programs provide information required to make operational changes, plan other facilities, and compare the local experience with national facilities. The goal of the study “Developing an HOV Evaluation Framework for UDOT” is to enable UDOT to perform its own HOV lanes evaluation. A comprehensive manual and a thorough training program on evaluating the Salt Lake Valley HOV lanes are the major deliverables of this study. The effectiveness of the HOV lanes extension in the Salt Lake Valley will also be reviewed. The study’s research objectives are to briefly review of the effectiveness of the Salt Lake Valley HOV lanes extension, develop a comprehensive manual specifically designed to evaluate the HOV lanes in the Salt Lake Valley, and train UDOT professionals.

EVALUATION OF UDOT’S ATIS TECHNOLOGIES

Team: Dr Peter T Martin (Principal Investigator), Aleksandar Stevanovic, Dhruvajyoti Lahon (R.A), Kyle Cook (R.A.)

Abstract: The Advanced Traveler Information System (ATIS) uses communication and information technologies to collect, process, and distribute useful information to the traveling public. ATIS provides timely information on traffic, construction, road conditions, travel times, expected delays, alternative routes, and weather, thereby giving travelers the opportunity to make informed decisions on when to go, what transportation mode to use, and which route to take. The goal of this study is to evaluate UDOT’s ATISs. This study will assess the effectiveness of four ATIS systems in Utah: Variable Message Signs (VMS), Highway Advisory Radio (HAR), 511 toll-free phone service, and CommuterLink online service. This study’s research objectives are to determine the level of public awareness of ATIS in Utah, assess usefulness of UDOT’s ATIS, investigate the effect of ATIS on driving habits, and determine the influence of demographic factors on ATIS use. The public surveys will be conducted randomly in the Salt Lake Valley. Two-hundred licensed drivers will be interviewed personally. The responses will be statistically analyzed. Additionally, respondents will be asked open-ended questions to obtain their views on and suggestions for ATIS.

High Occupancy Vehicle Lanes Evaluation II

Team: Dr Peter T Martin (Principal Investigator), Dhruvajyoti Lahon (R.A.), Aleksandar Stevanovic (R.A.)

Abstract: High Occupancy Vehicle (HOV) lanes opened on May 14, 2001 on the reconstructed Interstate 15 (I-15) in the Salt Lake Valley . Periodic performance evaluations of the HOV facilities help determine whether the lanes are serving their intended purpose. This study evaluates the effectiveness of the HOV lanes in the Salt Lake Valley in their current third year of operation. Vehicle volumes, speeds, and levels of service were collected by video detection using the video feeds provided by the Utah Department of Transportation. The “floating car” technique was used to obtain travel time. Vehicle occupancy, person throughput and violation rates were collected manually. Almost all of the National Cooperative Highway Research Program HOV Systems Manual standards were met with respect to vehicle volume, person throughput, and travel time. Based on the accident data analysis, HOV lanes did not appear to be inherently unsafe. A public opinion survey was conducted and the results indicate that there is strong support for the HOV lanes from both users and non-users. It was concluded from this study that the HOV lanes are performing more effectively than they did during their first year of operation. The findings indicate that the HOV lanes prove their value during more congested periods. As congestion in the Salt Lake Valley increases, the benefits of the HOV lanes are also expected to increase. There is a need for continued monitoring to identify and keep track of these benefits.

Video Detection

Team: Dr Peter T Martin (Principal Investigator), Gayathri Dharmavaram (R.A.), Aleksandar Stevanovic (R.A.)

Abstract: This study evaluates the performance of video detection systems installed in the Salt Lake Valley . {Video detection is a non-intrusive type of detection technology that consists of cameras positioned on mast-arms to capture images of passing vehicles. A vision processor analyzes these images by using application specific algorithms. But the accuracy of video detection is still not known under different environmental and light conditions}. The performance of vendors such as Traficon, Iteris, Autoscope and Peek were evaluated to measure the accuracy of video detection systems. Data collected under different test conditions were viewed for the discrepant calls. The sources of the discrepant calls were noted. These sources were coded and rated according to their level of importance to analyze which types of discrepant calls were important in the red and green phases. The results showed that Traficon emerged as the winner, producing 97% correct detection under all test conditions, followed by Autoscope and Iteris. Video detection performed well in day and dusk conditions (87.2%). But the performance deteriorated in inclement weather conditions (81.3%) and night conditions (73.4%). The main reason for the deterioration in the performance of the system was the vision processor technology. Fourteen percent (14%) of the important discrepant calls could potentially affect the signal timing.

Dilemma Zone II

Team: Dr Peter T Martin (Principal Investigator), Vikram C Kalyani (R.A.), Aleksandar Stevanovic (R.A.)

Abstract: High-speed signalized intersections present risks to driver safety. These risks heighten when an intersection is far away from the previous intersection. Advance-warning signals (AWS) are provided on such roads to advise drivers that they are approaching an intersection. They also warn drivers of impending signal changes at the intersection. This paper evaluates two AWS sites in Utah, one in Brigham City and the other in St. George. It compares the proportion of vehicles in the dilemma zones (DZ) of the sites to a control site without an AWS. The study concludes that AWS do not reduce the number of vehicles in the DZ. Therefore, the existing AWS in Utah should be modified. The AWS in Brigham City could be improved if the distance from the AWS to the intersection were increased. The time period between actuation of the AWS and the signal change to yellow at the intersection could also be decreased. The AWS in St. George could be improved by redesigning the traffic controller operation at the intersection.

Evaluate Effectiveness of Dilemma Zone Advanced Signal Warning

Team: Dr Peter T Martin (Principal Investigator), Vikram C. Kalyani (R.A)

Abstract: High-speed signalized intersections present risks to driver safety. These risks are compounded when an intersection is far away from the previous intersection. Advance-warning signals (AWSs) are provided on such roads to advise drivers that they are approaching an intersection. They also warn drivers of impending signal changes at the intersection. This paper evaluates two AWS sites in Utah, one in Brigham City and the other in St. George. It compares the proportion of vehicles in the dilemma zones (DZs) of the sites to a control site without an AWS. The study concludes that AWSs do not reduce the number of vehicles in the DZ. Therefore, the existing AWSs in Utah should be modified. The AWS in Brigham City could be improved if the distance from the AWS to the intersection were increased. The time period between actuation of the AWS and the signal change to yellow at the intersection could also be decreased. The AWS in St. George could be improved by redesigning the signal actuation mechanism.

Automated Data Collection, Analysis & Archival

Team: Dr Peter T Martin (Principal Investigator), Wu Peng (R.A), Xiaodong Wang (R.A)

Abstract: ITSs produce massive amounts of operational data originally used in real time traffic control strategies. ITS data could potentially be beneficial in many additional applications in transportation operations, planning, reporting, and research. TMSs are a basic component of Utah's ITS. These stations report traffic volume, speed, and occupancy every twenty seconds. Presently, TMS data lacks an archival and analysis strategy, which keeps it from being widely utilized; however, agencies require this information for multiple purposes. Drawing from nationwide ITS data uses, this research established guidelines for implementing an automated TMS data collection, archival, and analysis system. Five components of this system including: data aggregation, data quality control, data storage, database management system, and user interface were identified and examined. This research particularly investigated TMS data quality due to increasing concerns in regard to their accuracy. This study also developed models to derive various performance measures from TMS data, such as travel time, delay, vehicles miles traveled, vehicle hours traveled, and traffic variability. In addition, a prototype was designed to show how the system processes raw TMS data and applies them to diverse applications. In a word, this study demonstrated how to archive and better utilize TMS data and for what uses.

Detector Technology Evaluation

Team: Dr Peter T Martin (Principal Investigator), Yuqi Feng (R.A)

Abstract: Inductive loop detectors are widely used in traffic applications, but some disadvantages, such as disruption to traffic flow during installation and maintenance, higher failure rate under particular conditions, and inflexibility, urge professionals to seek alternatives to inductive loops. Market demands and technology advancement have inspired manufacturers to develop new detector devices with improved performance and capabilities. A large quantity of detector devices with different operation theories is now available on the market. This paper reports on the present status of detector technologies and on the development trends in these technologies. No single detector device is best for all applications. Each has its limitations, specializations, and individual capabilities. Because such a wide array of detector technologies and devices exists, it may be problematic to select the optimal detector technology and particular device to meet specific project requirements. To a large extent, successful application of detector technologies depends on proper device selection. This report designs a systematic selection method suitable for permanent applications. The selection method considers many factors, including data type, data accuracy (within different environmental and traffic conditions), ease of installation and calibration, costs, reliability, and maintenance. These factors are compared among a variety of detector technologies and devices in order to lead the customer to a suitable decision. This report provides comparison matrixes based on detector technology and specific devices in this field of technology. The technology matrixes offer general information about each detector technology, and the device matrixes give specific information regarding each particular detector device. Selecting an appropriate device product is more important than choosing a specific technology. The matrixes need to be continuously updated to reflect new changes on the detector market.

Expected Value Analysis

Team: Dr Peter T Martin (Principal Investigator), Yali Chen(R.A)

Abstract: In order to improve the safety of the Utah highway system, the Utah Department of Transportation (UDOT) must have accurate accident statistics for different highway classes. Although UDOT keeps a complete accident database, it does not reveal commonalities and trends behind accidents. In addition, Utah traffic facilities and services change over the years, emphasizing the need to evaluate the Utah Highway Networks' Functional Classification System. This study measures the safety of different Utah highway functional classes (FCs) according to traffic patterns, traffic volume, geometrics, and travel speed. It presents a process to analyze accident data for different FCs and also produces the expected accident range in order to identify locations with high accident rates.

Fast Track

Team: Dr Peter T Martin (Principal Investigator), Joe Perrin (Project Manager), Aleks Stevanovic (R.A)

Abstract: Fast Track The I-15 is under construction through a "Fast Track" form of contract with the contractor both designing and constructing the works. The traveling public has been disrupted through increased congestion during this large Design-Build Reconstruction Project. There would have been congestion due to the reduction in capacity had the contract been let on a traditional design/tender/build form of contract. Contractors bear greater risk with Design-Build and are paid more. The question is whether the early completion benefit of design-build is cost effective. So, how do the social costs of the fast-track, traditional and "do nothing" scenarios compare?

Ramp Metering

Team: Dr Peter T Martin (Principal Investigator), Joe Perrin (Project Manager), Robert Lambert (R.A)

Abstract: Ramp Metering has been implemented since 1997 in Utah on I-15 in Davis County and it appears to be an effective tool in helping reduce traffic congestion. Ramp meters are also to be installed as part of the Advanced Traffic Management System (ATMS) to help manage traffic on other critical freeway segments such as I-15, I-215 and I-80. Ramp metering has been proven in other states and around the world to help reduce the onset of traffic congestion by forced offline storage and reduced merging turbulence. There is a need, however, to evaluate the various ramp metering strategies available and adopt the appropriate one to optimize ramp meter control along the Wasatch Front. This study will focus on simulating several coordinated ramp-metering strategies on the Utah freeway network. The results will be useful in developing recommendations for the best candidate strategy to use. Although ramp metering has been implemented and additional ramp meters are proposed as part of the UDOT ATMS, additional benefits can be realized by coordinating the ramp meters with each other. As of today, each ramp meter is locally controlled by its' own calculated or preset ramp meter rates. Each locally controlled ramp meter works independently and does not "assist" the other ramp meters in controlling the traffic flow onto the freeway system. Certain efficiencies are not realized through local ramp meter control.

SCOOT

Team: Dr Peter T Martin (Principal Investigator), Joe Perrin (Project Manager), Chintan Jhaveri (R.A), Bhargav Rama (R.A), Wu Peng (R.A), Xiaodong Wang (R.A), YuQi Feng (R.A)

Abstract: Poor traffic signal coordination not only frustrates the motoring public, but also increases congestion and pollution. Signal coordination problems are due to outdated or inadequate signal timing plans. This project seeks to determine the effects on traffic conditions in downtown Salt Lake City's street network by simulating an adaptive traffic signal control system. An adaptive system recalculates and implements optimized signal coordination according to the traffic that is on the street at that time. This will be the first such simulation using the most widely recognized and adaptive signal control system, SCOOT (Split, Cycle, and Offset Optimization Technique). The results of the study are likely to receive both local and international attention and future funding.

Surface Street LOS

Team: Dr Peter T Martin (Principal Investigator), Joe Perrin (Project Manager), Brad Coleman (R.A)

Abstract: The Traffic Operation Center (TOC) for the Utah Department of Transportation (UDOT) gained the ability to monitor progression on I-15 by embedding loop detectors in the road that detect speed and flow data. The TOC is able to instantly see the Level of Service (real-time LOS, or RTLOS) along the I-15 corridor so they can see how the commute is progressing. In an attempt to improve public service, UDOT seeks to extend this technology to the main arterials dispersed around the Salt Lake Valley. However, there has not been any research or development into providing RTLOS measurements for arterials streets. A requirement for this project is to determine the RTLOS of the arterials using existing equipment along the arterial. The most common equipment is the stop bar detectors embedded at the signalized intersections. The goal of this research is to find a way to measure RTLOS with Salt Lake's existing infrastructure and deliver that information to the TOC for distribute it to the public. The uniqueness of this project is also conditional to the type of data with which the LOS would be calculated. The aim of this research is unprecedented and may be the foundation to develop a unique tool for traffic operation centers across the country.

Pedestrian Safety

Team: Dr Wayne Cotrell (Principal Investigator), Joe Perrin (Project Manager), Bhargav Rama

Abstract: The Mean Streets 2000 report, one of a series published by the Surface Transportation Policy Project (STPP), has attracted media attention. Among the report's findings are that average statewide spending on pedestrians is about 0.6% of total transportation spending, 13% of all traffic deaths involve pedestrians, 59.1% of all pedestrian deaths occur where no crosswalk is available, and that safety programs focus on keeping pedestrians out of the way rather than on safe driving or pedestrian amenities. The report provides a pedestrian danger index (PDI) for each metropolitan area in the U.S. with a population of one million or more. The Salt Lake City-Ogden area has the 12th highest index in the nation. Also, in the Salt Lake City-Ogden area, pedestrian traffic deaths represent 25% of all traffic deaths, which is twice the national average and sixth highest among large metropolitan areas. The child pedestrian death rate in Utah is the third highest in the country. These statistics indicate that pedestrian safety is a problem, and that Utah's pedestrian safety performance is among the U.S.' worst in several areas. The Utah Crash Outcome Data Evaluation System (Utah CODES) database provides information on pedestrian-motor vehicle crashes in the state. During 1998, there were 748 reported pedestrian-vehicle crashes; 679 of these involved pedestrian injuries, and 41 involved pedestrian fatalities. A total of 694 of the crashes, 629 of the injuries, and 38 of the fatalities occurred in counties featuring substantial urbanization: Cache, Davis, Salt Lake, Utah, Washington and Weber. The remaining 54 crashes, 50 injuries and 3 fatalities occurred in counties that are predominantly rural. Clearly, pedestrian safety is a concern in both urban and rural areas in Utah.

Accident Data Availability (ADA)

Team: Peter T. Martin (Principal Investigator), Joe Perrin (Project Manager), Blake Hansen (Project Coordinator), Alex Barrios (R.A)

Goals & Objectives: The goal of this research is to develop an accident database-Geographic Information Systems (GIS )interface that allows for wider dissemination of the information. Legal issues often arise from accident information requests and maybe why UDOT has been reluctant in the past to make the accident information so readily available. These issues will be explored to determine a department-wide policy on accident data access. In addition, the support of the accident data base is based on providing timely and accurate accident information input in the proper format. An accident data collection methodology will be developed that includes an electronic data input template, Global Position Systems (GPS) and recommended practices. This research effort will be completed by fulfilling the following objectives:

Integrate the data into a GIS format interface most likely using arc view
Identify issue related to accident data access that should be identified before a department-wide policy for dissemination of accident data.
Determine requirements for getting hardware into all patrol cars in all jurisdictions in the State (Laptops for data entry, (GPS).

Abstract: Accident information is valuable for identifying problem areas or locations of interest. This is true not only to the Traffic and Safety section. UDOT is comprised of many departments that would benefit from access to accident information. Not only is the data not readily available, its current format is one of summary spreadsheet format utilizing codes and mileposts to denote location, type and severity of accidents. With little graphical interface, the usefulness of the accident information by most not familiar with the codes is questionable. This proposal promotes the integration of the accident database with a GIS format to allow a graphical and spatial interface. The issues being addressed by this research focus on information circulation. Two specific issues have been identified as key elements in the research.

Making accident information more readily available.
Incorporating a GIS graphical interface with the accident information to allow for query searches on various accident attributes.

Trapeze

Team: Dr Peter T Martin (Principal Investigator), Joe Perrin (Project Manager), Aleks Stevanovic (R.A)

Abstract: This project should propose alternatives to help the paratransit operator to make paratransit operations more efficient and less expensive. On the other side we have to keep in mind that paratransit service should provide more reliable and faster service for paratransit users. Using operational data from the transit agency in the Salt Lake City a parametric analysis of the operating policy should be made. For example, passengers with travel distances above a specified threshold could be assigned taxi service instead of transit, depending on system-wide impacts on travel time and cost. Similarly, this project should consider alternative that could more restrict reservation policies and encourage prescription trips, or involve different pricing systems for different types of service. Further, project could propose division among current paratransit users in different categories according to their capabilities to use more spatially or timely restricted service. Given such operational questions, transit managers, after consulting the guidelines and associated figures developed by this project, should be able to identify the local situation, and see the potential impacts on costs and passengers of different policies.

Real-Time Modeling of SCOOT

Team: Peter T. Martin(Principal Investigator), Joe Perrin(Project Manager), Blake Hansen (Project Coordinator)

Goals & Objectives: To simulate a twenty-intersection downtown Salt Lake City street network under optimized fixed-time and adaptive traffic signal control in low, medium, and high traffic flows.

Abstract: SCOOT (Split, Cycle, and Offset Optimization Technique) is a sophisticated adaptive traffic signal control system. If a city considers procuring a SCOOT system, then offline modeling with TRANSYT7F provides some indication of the potential benefit. The user, however, does not realize the full potential of SCOOT until it has been installed. This project explains how SCOOT has been configured to provide a simulated connection to CORSIM. CORSIM allows a "run-time extension" which means that network state information can be extracted and that the traffic signal timings can be modified during the simulation. CORSIM is set up such that any RT-TRACS algorithm can be added in the form of a dynamic link library. SCOOT, which is one of the first adaptive traffic control systems, runs on a completely different platform than any of the RT-TRACS control algorithms. SCOOT runs on the VMS operating system, CORSIM on Windows NT. The two are connected across Ethernet with a dynamic link library interface that extracts the signal state and detector information from CORSIM and converts it to a format that SCOOT understands. SCOOT processes the information and sends it back across the Ethernet. The waiting dynamic link library receives this new signal timing information from SCOOT, converts it into CORSIM signal timing commands and gives CORSIM permission to process the simulation for the next second. This work offers traffic engineers the opportunity to evaluate the impact of SCOOT in a simulated environment. For the first time, in North America, it will be possible to "try before you buy".

Turning Movement Estimation In Real Time - TMERT

Abstract: Traffic flow detectors are expensive. There are many already installed on our city streets and highways across the Nation. Advances in modern communication technologies means that now we can access the traffic flow data both remotely and in real time. This wealth of information from sparse collections of detectors can be fused, managed and manipulated. This project demonstrates how a new model, Turning Movement Estimation in Real Time (TMERT) infers unknown flows from those detected. The impact of this model is far reaching. Widely deployed, it will eliminate the need for costly turning movement detectors or surveys. The estimates serve as "virtual" detectors.

The model was developed under support from the UK Government, California PATH and the University of Utah. The work reported here, as supported by the ITS IDEA program, is a comprehensive evaluation of the effectiveness of the quality of the estimates. With the benefit of a rigorously developed theoretical network, the model has been tested with some 7,000 simulations.

The literature tells us that most other real-time methods of traffic flow estimation rely on some notion of flow distribution, usually associated with the familiar Origin and Destination Matrix. These are appropriate for the wider view and will provide information on trends. TMERT however, coming from a fundamentally different basis makes no behavioral assumptions. More simply it serves to "fit" unknown flows into a network constrained by a partial set of detected data. This State Estimation technique allows network-wide evaluation instead of the individual intersection in contrast to the other techniques focus. The network oriented, non-behavioral characteristics of TMERT provide multiple application possibilities including on-line functions, such as turning movement estimates and support of adaptive control traffic signal systems, and a range of off-line functions from updating fixed time signal plans to planning level land use impact estimates.

This project addresses the relationship between estimation reliability and network flow-level, vehicle detector coverage, and detector pattern layout.

Final Report (PDF, 461 KB, new window)

A Comparative Analysis of the Alternative Pavement Marking Materials for the State of Utah

Abstract: Pavement markings provide guidance for traffic, separate opposing lanes of traffic, prohibit passing maneuvers, and delineate roadway edges. Pavement markings convey traffic regulations and warnings to drivers. Effectiveness depends on markings visibility. Most types of pavement markings are retro-reflective. Tiny glass beads are embedded in a white or yellow matrix. Retro-reflectivity is the reflection of light in the direction from which it came. This describes the reflection a driver views when illuminated from the vehicle headlights. An effective pavement marking system facilitates driver guidance, improves traffic flow, contributes to driving comfort, and enhances traffic safety.

Approximately 140,000 m3 (37,000,000 gallons) of traffic marking paint, 50,000 Mg (55,000 tons) of thermoplastic marking material, 118,000 Mg (130,000 tons) of glass beads, and an unknown quantity of raised, tape, and thermosetting markings are laid annually in the United States.

Final Report (PDF, 217 KB, new window)