2013 Session: 341

• Impact of Ramp and Mainline Flow Mix on Freeway Breakdown and Capacity Characteristics
  Abstract: This study examined how the mix between ramp and mainline flows influences breakdown and capacity characteristics such as: pre-breakdown flow (PBDF), queue discharge flow (QDF), queue discharge speed (QDS), breakdown duration (BDD), and capacity change after breakdown. The mix between ramp and mainline flow was represented using the Ramp Volume Ratio (RVR) which is defined as the percentage that the ramp flow is of the total merge area traffic flow. The research also examined how results may alter if the analysis adopts the plain Ramp Volume (RV) itself, as a flow mix index, instead of the RVR. Four freeway bottlenecks in Milwaukee freeway system were used in the analysis. Because merge areas can have different utilization levels or congestion intensities, bottlenecks were ranked based on the utilization of the merge influence area and treated distinctly. For bottlenecks with heavily-utilized merge-areas, both capacities (PBDF and QDF) were reduced when RVR increased. Interestingly, BDD appeared to be very sensitive to RVR. BDD was found to double, triple or almost quadruple when ramp flow comprised a higher percentage of mainline flow. For bottlenecks with moderately-utilized merge-areas, results were usually different and counter-intuitive. For some cases, increasing the RVR resulted in a capacity flow increase and BDD decrease. The adoption of the RV index instead of the RVR index altered or reversed many results and trends for all bottlenecks. The study explains the difference between RVR and RV, demonstrates the impact of merge-area utilization, and provides further recommendations.
  Authors: Dehman, Amjad; Drakopoulos, Alexander
  Authors: Dehman, Amjad; Drakopoulos, Alexander
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-0388
• Incremental Delay Methodology for Assessing Effects of Nonrecurring Congestion on Freeway Facilities
  Abstract: This paper introduces a method to estimate the incremental effects of different sources of congestion on freeway facility travel time and reliability in a Highway Capacity Manual context. The method relies on generating a suffciently complete set of operational scenarios for a freeway facility to fully describe its operational variability over a one-year reliability reporting period, and under consideration of recurring and non-recurring sources of delay. The method calculates the associated probabilities for each scenario and produces a variety of performance measures at the facility level for each 15-minute analysis period. Using these data and associated scenario probabilities, the method calculates the contribution of each source of congestion to the performance measure of interest, distinguishing between the base recurring delay, and the incremental effects of impacts from weather and incidents. The method was applied to a 12.5 mile freeway facility on I-40 near Raleigh, North Carolina to illustrate the approach. Based on the findings, the demand level was the main contributor to delay on the facility. In fact, 85% of the delay was attributable to demand level. Inclement weather and incidents contributed at most 15% of delay for the study site of which around 12% was attributable to incidents and the remaining 3% to inclement weather. The small impact of inclement weather in our study site was expected as Raleigh is not often exposed to inclement weather events with large capacity reductions. Although inclement weather and incidents increased the delay by at most 15%, they significantly impacted travel time reliability on the study site. Non-recurring congestion yielded a threefold increase in the value of the planning time index, and significant rises in other reliability performance measures. The findings from this research are important, since they highlight that high delays oftentimes attributed to freeway incidents and weather events, may in fact be largely attributable to the underlying recurring congestion. These results are important for agencies exploring the use of active traffic and demand management strategies to improve freeway operations, as they suggest that demand management treatments may result in a larger net-benefit than strategies aimed at capacity management under adverse conditions.
  Authors: Hajbabaie, Ali; Schroeder, Bastian J.; Rouphail, Nagui M.
  Authors: Hajbabaie, Ali; Schroeder, Bastian J.; Rouphail, Nagui M.
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-0262
• Models for Estimation of Drivers’ Impatience on Two-Lane Rural Highways
  Abstract: The main objective of the study was to model the decision making process of drivers during passing maneuvers performed on two-lane rural highways, using only traffic characteristics. By limiting the input data in this way, the findings and their implications become applicable to any two-lane road segment. On two-lane highways, when traffic volume increases, drivers may be unwilling to travel in platoons, resulting in an increased willingness to execute risky passing maneuvers. This phenomenon has been called “drivers’ impatience”. The study proposes an empirical method for evaluating drivers’ impatience using the concept of a critical passing gap. It was found that the critical passing gap has a bounded decreasing function determined by the total traffic volume or the volume in the oncoming lane. The critical passing gaps were found to be Log-Normally and Normally-distributed with an average value of 16.5 seconds, and a standard deviation of 5.5 seconds.
  Authors: Rozenshein, Sofya; Polus, Abishai; Cohen, Moshe
  Authors: Rozenshein, Sofya; Polus, Abishai; Cohen, Moshe
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-0451
• Enhancements to Freeway Facilities Method in the Highway Capacity Manual to Enable Reliability Analysis
  Abstract: The freeway facilities methodology in the Highway Capacity Manual (HCM) covers both under saturated and over saturated traffic regimes. The methodology provides a fast and reliable approach for analyzing freeway traffic operations and research is ongoing to add a reliability analysis component to the method. For that purpose, several key methodological enhancements are needed to make the method “reliability ready”, which are described and illustrated in this paper. These enhancements include: incorporation of capacity drop in queue discharge flow compared to pre-breakdown flow, adding Speed Adjustment Factor (SAF) as a new lever for modeling non-recurrent congestion sources like weather and incidents, modeling improvement of SAF and Capacity Adjustment Factor (CAF) in merge, diverge, and weaving segments, using new defaults of SAF and CAF for freeways, and adding new congestion performance measures. Further enhancements are made to explicitly incorporate CAF and SAF into methodologies for weaving and merge/diverge segments, which were typically assumed to operate as basic freeway segments for CAF less than 1.0. In order to generate a cumulative travel time distribution for reliability analysis, thousands of data points are generated with the automated version of the freeway facilities computational engine. In addition to the existing performance measures generated by the computational engine, some new output variables are proposed in the form of the Travel Time Index (TTI) and a Denied Entry Queue Length (DEQL) measure. The various enhancements are illustrated using computational examples and a discussion is offered as to their effects on conventional freeway facilities analysis, without the evaluation ofreliability.
  Authors: Sajjadi, Soheil Seyyed; Schroeder, Bastian J.; Rouphail, Nagui M.
  Authors: Sajjadi, Soheil Seyyed; Schroeder, Bastian J.; Rouphail, Nagui M.
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-0600
• Methodology for Developing HCM-Based Oversaturated Speed Flow Model
  Abstract: Oversaturated speed, flow and density relationships are of key importance for freeway operations studies. The Highway Capacity Manual (HCM) oversaturated model defined by a linear transition from flow and density at capacity to zero flow at jam density in the flow-density space provides a reasonable general representation of this relationship but does not provide an unbiased representation for all freeway facilities with different road conditions or driver behavior. This study proposes a method for fitting the HCM model to oversaturated flow and density. Fifteen-minute aggregated flow rate and speed data from Traffic.com fixed-location sensors at three sites on North Carolina (NC) urban freeways in 2010 were collected. Density was calculated as flow rate divided by speed. The fitted models for these sites were compared to the default HCM model. A set of thresholds was defined to identify eligible sensor observations that represent steady state, congested traffic conditions. Results reveal that data observations under inclement weather, lane closures, or incidents will bias the model fitting results and thus need to be filtered out. The steady state congestion data identified in the manner proposed in this study fit well with the HCM-based linear flow-density oversaturated model. This method avoids possible bias caused by capacity and jam density differences between the default HCM model and specific site models, and thus the fitted models represent the actual traffic characteristics relationships better than the default HCM models do. Therefore fitting a site-specific HCM-based model is recommended for sites with sufficient speed and flow data.
  Authors: Xu, Yilun; Williams, Billy M.; Rouphail, Nagui M.; Chase, R. Thomas
  Authors: Xu, Yilun; Williams, Billy M.; Rouphail, Nagui M.; Chase, R. Thomas
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-0823
• Assessment of Different Design Hours for Freeways
  Abstract: Conventional methods for the level-of-service assessment on freeways are based on the analysis of one specific peak hour. In Germany, like in many other countries, the 30th hour is defined as the design hour, which means that the 30th highest hourly traffic demand arising during one year is used as input value for freeway design. The paper analyzes the impact of different design hours on the level-of-service assessment for basic freeway segments. Based on traffic data from 50 freeway sections in Germany, the extent of congestion over a whole year is estimated by applying a macroscopic simulation model. With the model, all travel time losses due to congestion can be related to the hours in which the congestion was caused. It is found that on freeways with mainly recreational traffic, most time losses are caused within the first 30 hours, whereas on freeways with mainly commuter traffic, considerable time losses occur even beyond the 100th hour. Thus, the choice of a specific design hour in the range between the 30th and the 100th hour does not significantly affect the assessment results.
  Authors: Geistefeldt, Justin; Hohmann, Sandra
  Authors: Geistefeldt, Justin; Hohmann, Sandra
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-1151
• Statistical characterization of vehicles in 2-lane roads in Spain.
  Abstract: Two-lane roads are not easy to model. Macroscopic models omit the queue structure, essential for characterizing the level of service. Microscopic models are more adequate, but in practical terms they can be considered inadequate, since they require the introduction of data unavailable for most areas. This paper seeks to provide information to improve the use of microscopic models, in particular for the Madrid Region.Traditional vehicle types are passenger cars, trucks, buses and motorcycles. However, it is clear that congestion is caused by heavy vehicles, but also by slower cars, both having lower desired speeds. These slower vehicles can travel isolated or as platoon leaders.This paper focuses on the speed distributions of different vehicle types and queue structure, to establish measurement procedures and what errors arise from considering isolated vehicles or queue leaders. It also provides statistical information on desired speed distributions and day and nighttime driving in two-lane roads in the Madrid region. A new isolated vehicle criterion is proposed, allowing less restrictive conditions and larger sample sizes. Other conclusions are established regarding the inclusion of all platoon leaders in design speed considerations, and whether different populations can be established regarding queue length behind a vehicle.
  Authors: Castilla, Guillermo; Romana, Manuel G.
  Authors: Castilla, Guillermo; Romana, Manuel G.
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-1025
• Operational Effectiveness of Passing Zones Depending on Their Length and Traffic Volume
  Abstract: Two-lane highways operation has been deeply studied. The HCM proposes the length of no-passing zones as a parameter on the model; however, the distribution and characteristics of the passing zones along the highway are not addressed. This research presents an analysis of the effectiveness of passing zones considering their length and traffic volume. Data were collected on four passing zones of one rural highway in Spain, with two-way traffic volumes ranging from 100 veh/h to 900 veh/h and lengths between 265 and 1,270 m. More than 1,600 passing maneuvers were recorded during 53 hours. Operational effectiveness of passing zones was obtained from passing frequency, as the number of passes per time, and from passing rate, as the number of passes per time and following vehicle. The longer the passing zone is, the higher passing frequency is. However, the effectiveness of passing zones was stabilized from 1,100 m. On the other hand, balanced flows with two-way traffic volumes between 600 and 700 veh/h optimized the number of passes. Nevertheless, passing rate decreased as traffic volume increased. The results were validated with additional 46-hour data from other twelve passing zones.Highway Capacity Manual adjustments on average travel speed and percentage time spent following based on percentage of no-passing zones do not reliably represent effectiveness of passing zones. Effectiveness of every passing zone should be considered instead. Furthermore, the effect of traffic volume on the adjustment factors should be modified to maximize the passing opportunities on traffic volumes between 600 and 700 veh/h.
  Authors: Moreno, Ana Tsui; Llorca, Carlos; Garcia, Alfredo; Perez-Zuriaga, Ana Maria
  Authors: Moreno, Ana Tsui; Llorca, Carlos; Garcia, Alfredo; Perez-Zuriaga, Ana Maria
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-1483
• Observations and Analysis of Multistep Approaching Lane Changing Behavior at Two Expressway Merge Bottlenecks in Shanghai
  Abstract: From numerous field observations at two expressway merge bottlenecks, the paper identifies and studies a peculiar lane changing behavior, multistep approaching lane changing (MALC), at these bottlenecks. The characteristics of and detailed maneuvers in the MALC process are first described and compared with other three traditional lane changing behaviors (normal, cooperative and forced lane changing). Next, descriptive parameters such as lane-changing duration time, velocity and affected vehicle numbers from 132 sets of vehicle trajectory data collected at two merge bottlenecks in Shanghai, China are investigated and analyzed during the so-called trans-line ride (TLR) period. Significant differences are found between MALC and the traditional lane changing behaviors: MALC takes longer to complete (10s on average), involves lower lane changing velocity (15km/h on average during TLR period) and affects more vehicles (6 vehicles on average). As such, MALC poses more disruptive influences on the traffic flow, and could explain the occurrences of rapid capacity drop at expressway merge bottlenecks.
  Authors: Zhao, Li; Sun, Jian; Zhang, H. Michael
  Authors: Zhao, Li; Sun, Jian; Zhang, H. Michael
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-2400
• Lane-Based Breakdown Identification Method at Diverge Sections for Modeling Breakdown Probability
  Abstract: As for modeling breakdown probability on expressways, identification of breakdown occurrence is a decisive issue. To date, most of the existing identification methods have treated all lanes of mainline as one unit for a facility, and are conducted mainly through a check of the aggregated data of cross-section. However, this cross-section based method is oversimplified for bottlenecks at expressway facilities where lane usage preferences of each lane significantly differ like nearby diverge sections.Therefore, a lane based method is proposed to identify breakdown on each lane, and timing of breakdown occurrence is determined by a critical speed which is optimized through obtaining most significant speed drops accompanying with breakdown occurrences. The proposed lane based method is applied to four bottlenecks at diverge sections on intercity expressways in Japan. Superiorities of lane based method are highlighted as follows. Firstly, it can identify and exclude semi-congested cases where some lanes are congested and others are not. At a diverge section of Toyota JCT, 15 semi-congested cases are checked out through lane based method among 198 breakdown events which are identified by using cross-section based method. Secondly, timing of breakdown occurrence can be appropriately determined by lane based method which applies to 34 cases for this diverge section at Toyota JCT. These superiorities significantly improve accuracy of extracting breakdown flow rates which are underestimated by the existing cross-section based method. Breakdown probability is modeled by using the extracted breakdown flow rates through lane based method, and impacts of traffic condition characteristics are discussed. Diverge rate and lane utilization rate are found to have significant impacts on breakdown probability at diverge sections.
  Authors: Danpeng, Ma; Nakamura, Hideki; Asano, Miho
  Authors: Danpeng, Ma; Nakamura, Hideki; Asano, Miho
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-2514
• Use of HCM Freeway Facility Analysis and Microscopic Simulation Models as Part of Real-Time Management Decision Support Systems
  Abstract: Traffic management centers (TMCs) have successfully applied advanced management strategies to reduce incident impacts. However, the effectiveness of the management strategies can be enhanced by predicting, in real-time, the impacts of incidents based on the identified attributes of these incidents. This study investigates the implementation of macroscopic and microscopic traffic simulation models to estimate incident delays, as part of a real-time traffic management system. The two tools used in the investigation are FREEVAL, the computational engine of the Highway Capacity Manual 2010 (HCM 2010) and the CORSIM microscopic simulation tool. It can be concluded based on the results of this study that both CORSIM and FREEVAL are able to predict traffic delays due to incidents, however, the accuracy of prediction is affected by the accuracy of incident duration prediction and capacity drop estimates. The analysis also indicates that both models produce similar estimates of incident delays when the same analysis parameters are input to these two models, at least for the case studies investigated in this paper.
  Authors: Hadi, Mohammed; Xiao, Yan; Wang, Tao
  Authors: Hadi, Mohammed; Xiao, Yan; Wang, Tao
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-2783
• Rethinking the Driver Population Factor:Examination of Interstate 80 in California
  Abstract: The calculations for freeway capacity and demand include the input of a driver population factor, known as Fp, which allows the analyst to adjust the demand depending on the familiarity of the drivers on the roadway. This is based upon the assumption that unfamiliar drivers will drive at slower speeds and lower densities. However, there has been little research supporting the use of the Fp factor, and the Highway Capacity Manual cautions against its use unless the analyst is fairly certain the traffic stream is actually unfamiliar. As an experiment, two bottlenecks in California were selected and were analyzed both during the weekday peak and Sunday afternoons in periods where the traffic stream was likely to be non-local. Results showed that changes in queue discharge flow were minor at these two locations. Further research with additional sites and an increased awareness of the definition of familiarity will be required confirm results from this research.
  Authors: Seeherman, Joshua Lawrence; Skabardonis, Alexander
  Authors: Seeherman, Joshua Lawrence; Skabardonis, Alexander
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-4349
• Freeway Facilities Methodology Calibration for Travel Time Reliability Analysis: I-40 Case Study in North Carolina
  Abstract: This paper aims to calibrate the Travel Time Index (TTI) distribution for freeway reliability analysis in the Highway Capacity Manual (HCM). For this purpose, three calibration parameters were identified: the demand adjustment factor, the percent drop in capacity during queue discharge mode and the jam density. The calibration parameter effects on the TTI distribution were studied for a 12.5 miles facility on EB I-40 in Raleigh, North Carolina for which segment and facility travel times for the calendar year 2010 were available from INRIX web based database. Based on the initial model runs, the team experience with the HCM computational engine, and previous studies, three candidate values for each calibration parameter were selected resulting in 27 distinct combinations. The calibration analysis was limited to conditions in which there was no weather or incident events, although it is possible that some such events have gone underreported, and thus could bias the calibration process. The application of the two-sample Kolmogorov-Smirnov (K-S) test indicated that using higher demand adjustment factor in the seed file as 1.03 and selecting a value of 9.0% for queue discharge capacity drop yielded estimated TTI distributions that were not significantly different from the empirical INRIX distribution, with the caveat that the latter may contain some unreported incidents. Jam density values, on the other hand, showed very little effect on the resulting TTI distribution. The model verification, considering all the weather and incident events, showed that the calibrated TTI cumulative distribution functions of the HCM and INIRX got closer but the K-S test result yielded that the difference between the two distributions is still significant. This can be justified as there exists very high TTI values in the HCM model with very low probabilities and one year data of INRIX may not be sufficient enough to cover all the possible combinations.
  Authors: Sajjadi, Soheil Seyyed; Hajbabaie, Ali; Rouphail, Nagui M.
  Authors: Sajjadi, Soheil Seyyed; Hajbabaie, Ali; Rouphail, Nagui M.
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-5199
• Estimating Incident Propensity for Reliability Analysis in the Highway Capacity Manual
  Abstract: This paper presents the methodology used for generating incident probabilities required in the Freeway Scenario Generator (FSG) for Travel Time Reliability analysis in the Highway Capacity Manual (HCM). FSG requires the estimation of specified study period monthly probabilities of different incident types. Incident probability in this context is the fraction of time that a specific incident type is active somewhere on the freeway facility during the study period for the month considered. The proposed methodology is designed to recognize and deal with the varying levels of incident and facility data availability at the implementing agencies. A M/G/∞ queuing model is proposed for conversion of incident frequencies into incident probabilities when agencies have access to only frequencies instead of probabilities.
  Authors: Aghdashi, Seyedbehzad; Rouphail, Nagui M.; Hajbabaie, Ali
  Authors: Aghdashi, Seyedbehzad; Rouphail, Nagui M.; Hajbabaie, Ali
  Year: 2013
  Document Type: Paper
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-5058
• Statistical characterization of vehicles in 2-lane roads in Spain.
  Authors: Castilla, Guillermo
  Authors: Castilla, Guillermo
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-1025
• Statistical characterization of vehicles in 2-lane roads in Spain.
  Authors: Romana, Manuel
  Authors: Romana, Manuel
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-1025
• Assessment of Different Design Hours for Freeways
  Authors: Geistefeldt, Justin
  Authors: Geistefeldt, Justin
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-1151
• Lane-Based Breakdown Identification Method at Diverge Sections for Modeling Breakdown Probability
  Authors: Danpeng, Ma
  Authors: Danpeng, Ma
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-2514
• Use of HCM Freeway Facility Analysis and Microscopic Simulation Models as Part of Real-Time Management Decision Support Systems
  Authors: Xiao, Yan
  Authors: Xiao, Yan
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-2783
• Rethinking the Driver Population Factor: Examination of Interstate 80 in California
  Authors: Seeherman, Joshua
  Authors: Seeherman, Joshua
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-4349
• Impact of Ramp and Mainline Flow Mix on Freeway Breakdown and Capacity Characteristics
  Authors: Dehman, Amjad
  Authors: Dehman, Amjad
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-0388
• Methodology for Developing HCM-Based Oversaturated Speed Flow Model
  Authors: Xu, Yilun
  Authors: Xu, Yilun
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-0823
• Observations and Analysis of Multistep Approaching Lane Changing Behavior at Two Expressway Merge Bottlenecks in Shanghai
  Authors: Sun, Jian
  Authors: Sun, Jian
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-2400
• Observations and Analysis of Multistep Approaching Lane Changing Behavior at Two Expressway Merge Bottlenecks in Shanghai
  Authors: Zhao, Li
  Authors: Zhao, Li
  Year: 2013
  Document Type: Presentation; Poster
  Subject: Operations and Traffic Management
  Session: 341
  Paper Number: 13-2400