International Conference on Transportation and Development 2019: Smarter and Safer Mobility and Cities

The proposed research aims at developing an integrated and a most efficient travel management model to generate a smart transportation system by detecting traffic data at signalized intersections. This study will utilize an emerging data acquiring technique called “floating car data (FCD)”. Even if there were several past studies on using FCD, but those have been limited into obtaining traffic related information parameters (such as speed, etc.). But this study will focus on using FCD technology to obtain information on vehicular left turn movement at different levels of congestion at protected only left turn signal. Since left turning movements are considered as the points of having most of the conflicts involving both vehicles and pedestrians this research emphasizes on collecting traffic data for protected only left turning vehicular movements. Therefore, using the vehicular performance information during traffic jam, a most effective protected left turn signal timing can be regulated providing faster and safer left turning movement. This research will develop an integrated mathematical model based on the probabilistic approach at a signal focusing on left turning traffic movement. Driver behavior can also be specifically studied during real-time information relay on congestion. This approach can be applied for congestion relief during peak hour travels as well as in the events of accidents in local roads.

The purpose of this paper is to investigate the safety impacts of adaptive signal control systems (ASCS). The authors developed a series of negative binomial regression models of different crash types to explore the relationships between the number of crashes and the factors influencing those crashes (traffic characteristic and roadway geometric features). Four hundred forty-nine observations of yearly intersection crashes were used to develop different regression models. An indicator variable tracking the installation of ASCS at the intersections was introduced into all regression models to determine the efficacy of the ASCS in terms of improving intersection safety. Analyses revealed that the deployment of ASCS in three corridors have resulted in crash reduction at a 0.05 significance level. The ASCS system resulted in a 24% reduction in the total number of crashes, a 29% reduction in the number of fatal and non-fatal injuries, a 31% reduction in the number of rear-end collisions, and an 18% reduction in the number of angle crashes. Further, a series of variables related to the roadway geometric features were deemed statistically significant at a 0.05 level in all models analyzed in this study.

The pedestrian hybrid beacon (PHB) is a high-intensity pedestrian signal that can be placed at a midblock or intersection crosswalk. This system was introduced into the MUTCD in 2009, and since has increased in popularity across the country, including in New Jersey. Recent research has raised concerns about motorist and pedestrian comprehension of the PHB due to its recent introduction in New Jersey. In this study, video data was collected at a PHB site in the city of New Brunswick, New Jersey, in order to evaluate road users’ behaviors at the PHB. An online survey was also conducted among Rowan University community including students and faculty members in order to gauge public awareness and comprehension of PHB signals in New Jersey. Together, the video data and survey results are reflective of the current level of public understanding and the effectiveness of PHBs in New Jersey.

Because traffic incidents create unexpected driving conditions, traffic safety issues can arise. In particular, unexpected traffic incidents could require drivers to decelerate rapidly, causing an unsafe situation. Because the emerging connected vehicle (CV) concept can enable sharing of real-time information and help drivers to change their driving path, lane, or speed; CVs could avoid rapid decelerations caused by unexpected traffic incidents. This study evaluated the impacts of CVs on traffic safety near traffic incidents. The researchers used VISSIM 8 to simulate a 17-mile segment of Interstate 55 in Illinois, modelling 10 different incident scenarios, and measuring the deceleration rate of vehicles approaching an incident scene. The results from this study indicate that CVs have the potential to reduce the proportion of vehicles that rapidly decelerate near incident scenes along a freeway. Reducing the number of vehicles that rapidly decelerate is expected to reduce secondary crashes and improve safety.

The study is intended to quantify the crash avoidance effectiveness of different advanced driver assistance systems (ADAS) in the real-world environment. First, twenty ADAS technologies tested in recent years are summarized. The results show that an ADAS tends to have a higher effectiveness on heavy trucks than on light vehicles, and an ADAS may achieve a maximum 70% effectiveness on heavy trucks. However, the proposed effectiveness in previous research are limited to relative ideal conditions. Then the study improves the effectiveness estimation of an ADAS by considering the effect of system limitation, system adoption rate, and system usage features in the real-world environment. Six ADASs are selected as a case study. The results show that the practical effectiveness of an ADAS could be as much as 48.25% lower than the effectiveness proposed by previous studies, and it varies between different roadway facilities.

Managed lanes (MLs) have been implemented as an important traffic management strategy for improving traffic mobility, efficiency, and safety, in addition to generating revenue for transportation agencies. Scenarios were built and tested in microsimulation to specify the safest accessibility level and to decide on the safe weaving length near access zones. A log-linear model was developed with estimation of odds multipliers for the conflict frequency analysis. The results suggested that one accessibility level was the safest option in the 9-mile corridor. The length of 1,000 ft per lane change was shown to be the safest weaving length near access zones. Additionally, further analysis will be conducted for evaluating the MLs facility in a connected vehicle (CV) environment. Three types of lanes [i.e., MLs, general purpose lanes (GPLs), CV lanes] will be built and tested in microsimulation. The findings will reveal the impact of adding CV lanes on the safety of the MLs segments.

Freeway-related safety performance functions (SPFs) in the highway safety manual (HSM) are statistical formulas that were developed based on data from California, Maine, and Washington States. Therefore, HSM recommends modification of SPFs for a certain jurisdiction following a calibration methodology or development of local SPFs to enhance accuracy of predicted crash frequencies. The study estimated calibration factors using the HSM methodology and developed calibration functions using negative binomial method for freeway speed-change lanes in Kansas utilizing crash data from 2013–2015. The quality of estimated calibration factors and calibration functions were assessed using cumulative residual plots. For the calibration, 351 entrance speed-change lanes and 366 exit speed-change lanes were used. The results indicated that HSM methodology consistently underpredicted both fatal and injury and property damage only crashes for both entrance and exit speed-change lanes in Kansas. Since, calibration functions outperformed calibration factors, the study recommends the use of calibration functions for future freeway speed-change lane related decision making in Kansas.

This study explores the link between crash frequency and driver behavior on interrupted flow segments. Among the evolving body of research around this issue, the proposed approach is unique because, among other factors, it also takes into account adjacent land use for the crash analysis segments. Driver behavior data collected from GPS sensors are used to derive a variable termed ProJaT (proportion of jerk above threshold) for each of the 68 analysis segments in three arterial corridors in Baton Rouge, LA. A data-driven approach based on the estimation of correlation coefficients between ProJaT and segment crash rate is used to select the threshold used to define ProJaT for analysis segments. To increase the explanatory power of the model, additional variables such as ADT (average daily traffic), traffic signal presence, adjacent population density, and presence of curvature were also used in the crash frequency estimation model. The results of random-effect negative binomial (RENB) model with commercial land use as a random effect suggested that ProJaT is only significantly associated with crash frequency in segments adjacent to non-commercial land use. On segments adjacent to commercial land use, ADT and influence of traffic signals are significantly related to crash frequency. Implications of the results for proactive identification of crash-prone locations through driver behavior data collected using low-cost GPS devices are discussed in the study.

A hierarchical Bayesian statistical before-after model was used to analyze locations where raised medians have been installed. Twenty locations where raised medians were installed in Utah from 2002 to 2014 were used in this analysis. A model was developed that analyzed raised medians by access category (AC). Only three AC were represented in the data. Regression plots depicting a decrease in crashes before and after installation and crash modification factor (CMF) plots presenting the CMF values estimated for different vehicle miles traveled values were created as output from the before-after model. Overall, installing a raised median gives an approximate average reduction of 53 percent for all crashes. Individual AC analysis yielded results ranging from 32 to 44 percent reduction for all severity groups except severity 4 and 5. When the model was only run for crash severity 4 and 5, a larger reduction of 57 to 58 percent was found.

Traffic safety has been a serious public health issue, and there have been considerable efforts to minimize the injury severity of crashes if it is inevitable to avoid the occurrence. If this is the case, it is important to put efforts to minimize the severity of crashes to save lives or lessen injury severities. Many previous studies have explored the severity of traffic crashes. Among those studies, many studies have analyzed the highest severity level of crashes has been analyzed with characteristics of drivers, vehicles, roadway, etc. On the other hand, some other studies have investigate the severity level of individual road users including vehicle occupants, pedestrians, bicyclists, etc. Nevertheless, no studies have explored the severity of multiple passengers simultaneously. As it is expected that vehicle occupants in a same vehicle are likely to receive a similar level of crash impact, it would be desirable to consider the shared effects the severity levels of a driver and passenger. In this study, the authors aim at analyzing the severity level of driver and school-age passenger by adopting a bivariate ordered probit model to explore four severity levels (i.e., no injury, possible injury, non-incapacitating injury, and the sum of incapacitating injury and fatal injury). The results indicate that the contributing factors for the severity level of drivers and school-age passengers are quite different. They include individual, vehicle, and residence socio-economic characteristics. It is expected that the findings from this study will contribute to an efficient strategic plan to reduce traffic crashes.

When a collision occurs, information on its time and characteristics is needed for traffic control purposes; however, the post-collision impact on traffic and the corresponding expected recovery and clearance time is still an important and under-studied subject of research. Though the collision-induced delays may be estimated given the prevalence of mobile apps like Google Maps, the durations of post-collision effects remain largely unexplored. Existing studies mainly focus on modeling the collision induced queue formation and dissipation. However, limited research focuses specifically on the modeling of collision duration per se. Therefore, the objective of this study is to utilize hazard-based modeling to estimate collision clearance time by factoring in various endogenous and exogenous factors such as collision severity, lane width, number of lanes, and surrounding traffic conditions. 2017 and 2018 collision and traffic detector data from the California freeway performance measurement system (PeMS) were mined for developing a semi-parametric hazard function of collision durations. The resulting time dependent hazard—the likelihood of clearing a collision site—can help traffic controllers and drivers make better decisions while coping with the uncertainty associated with incident events.

The state of the practice for municipal traffic agencies identifying high-risk road segments has been to use data on prior crashes. While historical traffic crash data is valuable in improving roadway safety, it relies on prior observations rather than future crash likelihoods. Recently, however, researchers have developed predictive crash methods based on “abnormal driving events.” These include abrupt and atypical vehicle movements indicative of crash avoidance maneuvers and/or near-crashes, especially on highways. Due to limited data, the previous research only tested the crash-jerk ratio function on highways but not on other types of roads. This paper describes research that used naturalistic driving data collected from global positioning system (GPS) sensors to locate high concentrations of abrupt and atypical vehicle movements based on vehicle acceleration and vehicle rate of change of acceleration (jerk) on two interrupted highways. Statistical analyses revealed that clusters of high magnitude jerk events while decelerating were significantly correlated to long-term crash rates at these locations. These significant and consistent relationships between jerks and crashes suggest that such observational data can be used as surrogate measures of safety and as a way of predicting safety problems, further improving crash prediction models.

Pedestrians, bicyclists, and motorcyclists are called vulnerable road users (VRU) because they lack the protection from hard shell and safety belt. In recent years, they make up over 30% traffic fatalities in the U.S. Since VRUs are exposed to ambient environment, their activities follow outdoor weather conditions and exhibit strong seasonal patterns. Understanding the seasonal patterns of VRU crashes and being able to accurately predict their future occurrences can help decision-makers to allocate safety resource by season and achieve higher return in investment. This study used four data-driven models to analyze the 2007 to 2016 VRU crashes in Pennsylvania, starting from the easy to follow proportion and projection model, then delving into the seasonal autoregressive integrated moving average model, the local level model, and the exponential smoothing state space model. They all capture the annual trends and seasonal patterns of VRU crashes well. This study emphasizes methods of predicting full year VRU crashes from partial year crash data based on knowledge of seasonal patterns. The results show good prediction can be achieved by just knowing the first five months’ crash data.

Road accident-forecasting models are indispensable modules within any efficient strategy that gauge and enhance road safety. These models are used to assess the effectiveness of road safety policies that have been introduced. The greatest share of the annual 1.25 million global road fatalities occur in developing countries, yet very little research is done in this context. This paper presents a simple approach for forecasting traffic accidents in developing countries where minimal data is collected. The model applies cointegration analysis of the road accident and economic data. Cointegration relationship is tested through Engle-Granger two-step method. The error correction model is formed using the fatality rate as dependent variable and economic growth as predictor variable. Using this approach, spurious regression is avoided. The forecasts provide the numerical context through which a mechanism is created to set new road safety targets in Turkey with the intention of assisting the monitoring of potential safety countermeasures and thus promising reduction in road traffic fatalities.

This paper summarizes the development of the Utah intersection crash prediction model (UICPM) and the intersection safety analysis methodology (ISAM). The UICPM is a Bayesian generalized linear model that creates distributions of the expected number of crashes at each intersection, compares the actual number of crashes to the distribution, and ranks the intersections based on safety concern. The ISAM is the process that was developed to execute the UICPM. The first step is to prepare the model input by combining the roadway and crash data files. Crashes are assigned to intersections if they are located with the functional area of an intersection. The second step is to execute the UICPM using the R graphical user interface (GUI) tool and R software. The third step is to create a two-page intersection safety analysis report (ISAR) for intersections of interest and maps of the state, regions, and counties to visualize the model results.

The crash prediction models provided in the Highway Safety Manual predict crashes using traffic and geometric data for various roadway facilities, which are incorporated through safety performance functions and crash modification factors. In this study, calibration factors were estimated for four types of intersections in urban and suburban locations in Kansas: 3-legged signalized intersections (3SG), 4-legged signalized intersections (4SG), 3- legged unsignalized intersections with stop control on the minor approach (3ST), and 4-legged unsignalized intersections with stop control on the minor approach (4ST). The calibration factors were estimated as 0.64 for 3SG, 0.51 for 3ST, 1.17 for 4SG, and 0.61 for 4ST. The calibration factors show that the HSM methodology underpredicted crashes for 4SG, and overpredicted crashes for the other three intersection types. The reliability of the calibration factors was checked with the help of cumulative residual plots and coefficient of variation. Furthermore, calibration functions were developed, and their reliability was examined. The results showed that the calibration functions have better reliability as compared to calibration factors for all intersection types.

In many large cities worldwide, traffic congestion is a critical issue. In general, secondary tasks are considered dangerous while driving. However, in heavy traffic, where the major driving task is significantly reduced, secondary tasks may reduce burden on drivers, maintain their arousal level, and cause them to perceive time as moving more quickly. In this study, we conducted an experiment using virtual congestion to clarify drivers’ stress, reaction delay, and perceived time under in-car activities of listening to music (passive task), and completing a quiz and talking with a passenger (active tasks). As a result, under “do nothing,” drivers perceived time as moving slowly and felt higher stress and drowsiness. Meanwhile, under the passive and active tasks, they perceived time as moving quickly and both the passive and active tasks showed stress reduction effects. Thus, it is necessary to develop driving support technology focusing on psychological, behavioral, and motivational aspects of drivers.

Dynamic lane assignment (DLA) is one-way of improving the efficiency of traffic operation at signalized intersections by dynamically changing the number of lanes assigned for a given turning movement depending on the instantaneous demand. For the successful implementation of DLA, the drivers should be aware of the lane assignment well before approaching the intersection through variable message signs (VMS). VMS is one of the widely used intelligent transportation systems (ITS) in urban areas, which can significantly support the implementation of DLA by providing drivers with real-time information on the existing lane group configuration while approaching a signalized intersection. Saudi Arabia has very diverse drivers’ population with a large percentage of chauffeurs working for households, industries, as well as taxi drivers with a considerable variation in their educational background and driving habits. This paper investigates the factors affecting the comprehension of VMS when used in conjunction with DLA to identify drivers who need extra attention in the licensing stage if DLA is applied. Statistical analysis and artificial neural network (ANN) techniques were used to assess age, education, occupation, and driver’s experience as possible predictors of VMS comprehension.

The country of Lebanon is considering the adoption of bus rapid transit system (BRT) in its public transport sector to improve mobility in the increasingly congested urban areas. Most passengers in this country are car-dependent. This paper studies the potential for modal shift by passenger cars towards BRT system and targets in particular the studying and working young adults. It utilizes the theory of planned behavior (TPB) factors—attitude towards behavior, subjective norms, perceived behavioral control, and intention—combined with the habitual car use to predict the user’s intention to switch to BRT. Therefore, a stated preference survey is designed to examine the extent to which changes in commuting modes is influenced by habitual car use and other psychological factors and to identify the suitable incentives that will likely encourage travel mode shift by passenger cars. Overall, the results of the regression analysis reveal that mode choice behavior is a reasoned action that is largely influenced by the anticipation of the favorable outcome associated with the behavior. The analysis showed that habit and intention are not related. However, the integration of BRT in Lebanon should take into account the rigid habitual behavior of car users given that strong habits might obstruct behavior even if intended.

Self-reporting is a valuable tool in transportation research practices especially in drivers’ non-compliance behavior on- and off-road. Self-reporting often serves as a proxy and provides access to data that may be difficult to obtain by other means. Whereas assessment and validation of such subjective observations is another challenge to decipher, objective measures of drivers’ behavior (i.e. speed behavior) could be useful to assess the correlation with self-reported data. To appraise the reliability of drivers’ self-reported data, this study was conducted to analyze drivers’ perceptions of their driving speed and compare it with the actual operating speed. Data was collected for the Seward Highway in Alaska; a two-lane highway that connects Anchorage with Seward. The actual operating speed data and the self-reported data were collected at passing lanes sites where a strategy of differential speed limits was recently implemented. Similarly, to model drivers’ self-reported attitude data, a seemingly unrelated regression (SUR) was used to account for contemporaneous correlation of error terms and the endogenous behavior of variables in the system of equations. Mean free-flow speeds were compared to the expected means of the drivers’ self-reported speeds using one-way ANOVA. In terms of operating speeds, the results suggested that the actual speeds of vehicles in the passing lanes were higher than the current posted speed limits (PSLs). The results of the statistical analysis indicated a significant discrepancy between the actual operating speeds and the drivers’ self-reported speeds. The findings of this study could have significant implications on the implementation of PSLs, and regarding the perception of drivers about their own speeding behavior on two-lane highways and the passing lanes, which may be addressed to improve safety. The drivers may be educated about PSLs, to improve traffic safety and operations and strategies be devised to better reduce the above-stated discrepancy.

Drivers aged 65 and older represent the fastest increasing demographic in the United States. However, aging can influence driving-related visual, physical, and cognitive functions which can negatively influence their driving performance. Difficulties encountered during complex traffic situations include navigating intersections, making left turns against oncoming traffic, merging into traffic, and making lane changes on limited-access highways. Several studies have used driving simulators to assess the performance of older drivers. Studies have also used neuropsychological tests to assess cognitive abilities for evaluating the skills and capabilities of older drivers. Visual screening, speed of processing information, visuospatial processing, and memory assessment were found to be key predictors of older drivers’ competence. Studies have also used functional magnetic resonance imaging (fMRI) during simulated driving to examine aspects of brain activity related to specific driving tasks. There remains a gap in the research when it comes to older drivers, neuropsychological tests, and studies using fMRI.

A survey was conducted with 463 participants to investigate the effect of pavement marking configurations at roundabout approaches on yielding behavior. Participants were asked to indicate where they would yield in an overhead view, given four different pavement marking configurations in a dynamic portion of the survey. Statistical comparisons found the distributions of yielding locations differ significantly. “Shark teeth” pavement marking resulted in participants yielding five feet further upstream than scenarios without. “YIELD” alone had the greatest variance, while “shark teeth” pavement markings result in the smallest variance in yield locations. Demographics, roundabout understanding, and whether drivers thought roundabouts provided a safety benefit did not result in different driver yielding locations. The results suggest drivers yield in different locations depending on different pavement marking configurations at roundabout approaches, which may influence the number of rear-end collisions. These findings can help impact future pavement marking application decisions.

A work zone crash occurred once every 5.4 minutes in the United States in 2015, according to the Federal Highway Administration (FHWA). This study investigated the impact of work zone barriers (cone pylons, concrete jersey barriers, and metal barriers) on driver behavior on a freeway using a medium-fidelity full-scale driving simulator. Traffic volumes were based on level of service (LOS) C in which 65 individuals participated in the study. A single factor analysis of variance (ANOVA) indicated that there was a statistically significant difference between mean vehicle speeds across all barriers as well as mean vehicle speeds across metal barriers for age groups 35 and above versus other age groups. An interesting observation was that drivers tend to deviate from the center of the lane, away from concrete jersey barriers on a freeway which is in complete contrast to driver behavior on an arterial road, based on a prior study.

Simulation is a proven engineering tool used to evaluate the needs of existing and future smart transportation systems. Cities are livable communities with transportation modes where drivers and pedestrians coexist. School zones are vital components in a city. A balance of mobility, speed compliance, and markings and signage in school zones is a must for all users’ safety and quality of life. Research has shown that drivers do not comply with speed limits around school zones. A survey conducted as part of this study indicated that 48% of drivers are confused regarding school zones’ signage and speed limits. Possible solution options were presented in the survey, with combinations of signage and pavement markings to better inform the road users, and to improve drivers’ awareness and compliance. Based on the survey’s results, a combination of signage and pavement markings was selected to improve the current conditions. This combination was compared to an existing school area signage and marking condition in the city of Aguadilla, Puerto Rico. Twenty-four scenarios were evaluated using the University of Puerto Rico driving simulator. The speed and acceleration noise were evaluated between the base and proposed scenarios. The principal finding is a mean speed reduction on 70.8% of the roadway segments evaluated, using the proposed combination of an overhead speed sign with alternating flashing beacons and pavement markings.

In the traditional entropy maximizing derivation of the gravity model, any micro-distribution of the individual flows within a trip linkage is seen as equally likely a priori, and the aim is to find the macro-distribution flows which is most probable, subject to relevant constraints. A method has been proposed by this author, using the principle of maximum entropy, to derive the trip probability distribution catering for the biases within the linkages. The result is a well-specified model with two deterrent factors, one, a trip frequency function and the other, a trip cost function. Calibration is performed using two criteria. The cumbersome balancing mechanism of the traditional spatial interaction model is no longer present. These two criteria are obtained directly from the principle of maximum entropy. The first criterion is the natural constraint equation. The other criterion is the selection of the probability distribution with the highest entropy value, which represents the most unbiased, most random, and most objective distribution consistent with the information contained in the form of constraints. The principle of maximum entropy thus provides its own calibration criteria and avoids the use of any inherently biased statistic. The strength of this model is its forecasting ability. This paper describes an approach for forecasting, and some examples are given using the forecasting method. A comparison is made with the highly accepted gravity model.

This paper uses data from the household travel survey of Denizli City including more than thirteen thousand passengers' information to investigate the factors affecting mode choice. In this research, by using a multinomial logit model for different trip purposes, it is identified that how much the passengers' socioeconomic characteristics and attributes of alternative modes affect the mode choice. The analysis indicates that travel time is one of the important policy-relevant factors that affect the passengers' mode choice. The finding of this research designate that the first break-even point between walking mode and other modes is 28 min for HBS, 22 min for HBW, 25 min for HBO, and 10 min for NHB. The first break-even point occurred between walking mode and the private vehicle on all trip purpose except HBS. These results help policymakers to increase walking rates by recommending some essential suggestions.

This study is focused on deriving a reliable trip-interchange matrix considering the limited availability of dependable trip-data in developing countries. In the present study on Amritsar City, India, only revealed preference data on trips made on the previous day was available. Additional data on trips produced, and the number of households and the zonal-population was obtained from secondary sources. The first phase of this study includes identification of important activity centres, preparation of base-maps, geo-referencing, determination of the zone-centroids, and determination of inter-zonal aerial distances. The second phase includes determination of corrected trip-generations, step-wise filtering of the existing erroneous revealed-preference data, development of a log-linear regression model in addition to an ANN model for predicting the final subset of filtered data points, and the comparison of the two approaches based on the trip-frequencies. The final origin-destination matrix was found to be best predicted using the ANN approach.

Tunnels in urban areas typically have restricted geometry (e.g., narrow lanes, sage curve, and low clearance) that hamper the ability of drivers to observe downstream traffic conditions beyond the immediate lead vehicle. This short-sighted view of traffic or reduced “look-ahead distance” forces drivers to reduce speeds and increase their headways, leading to less efficient traffic flow operations. With the use of connected vehicles (CVs) and advanced sensing technologies, it would be possible to monitor the speeds and trajectories of the downstream vehicles and communicate such information to the CVs in the upstream. In this paper, the impacts of increased look-ahead distance are investigated through microsimulation. A VISSIM model is developed and calibrated for the current operating conditions at the Hampton Road Bridge Tunnel in Virginia. The look-ahead distance as well as the percentage of CVs in the traffic stream is varied to understand the impacts on the tunnel throughput and travel times. With a higher look-ahead distance for the CVs, the travel times are shown to be reduced by 14% and the tunnel throughput increased by 20%. The results show the significance of look-ahead distance as a car-following parameter and the sensitivity of the system performance to varying levels of this parameter.

Traffic congestion is known as one of the serious global problems the transportation systems face due to the rapid urbanization in large cities. Traffic congestion imposes a major cost on the users and economy and causes significant reductions in traffic flow. The congestion problem is even more serious on urban arterials especially near driveways and unsignalized intersections’ locations where vehicle movements are subject to drivers’ judgment. Application of active traffic management (ATM) strategies is the common way to overcome the operational and safety challenges associated with congested urban arterials. Yet, the rapid increase in people movements to urban areas is becoming beyond the management capabilities of traditional ATM strategies. The emerging autonomous vehicle (AV) technology offers ample opportunities to overcome the ever-growing congestion problem, hence supplement and/or replace traditional ATM strategies. This study investigates the benefits of AVs for congestion management at driveway locations on urban arterials. The impact of AVs on the traffic density of each lane of an urban arterial in close proximity of a driveway is assessed in this study. Level of service (LOS) measure is used as a boundary to evaluate the impact of AVs under different traffic congestion levels. Eight separate traffic simulation models are developed for two scenarios (conventional vehicle and AV environments). The results indicated that AVs have a superior capability in relieving traffic congestion significantly in the vicinity of a driveway in moderate LOS. This performance is achieved by the AVs capability to distribute vehicles evenly across lanes which helps redistribute gaps along the arterial section, hence allows vehicles from the driveway to merge more smoothly to the arterial. These findings indicate that transforming into an AV network can provide optimal access point management and allow better accessibility to adjacent land-uses and facilities.

This study develops a longitudinal model framework to analyze monthly demand for for-hire services including traditional taxis, Uber, and Lyft. Structural equation modeling techniques are employed to examine the relationship between the monthly number of picks-ups for these services, built environment attributes, and measures of access to transit and bike-sharing modes within New York City neighborhoods. The results show that pedestrian-friendly street network designs are linked with fewer monthly pick-ups, whereas higher levels of neighborhood compactness are linked with increased monthly pick-ups. Further, increased access to bike-sharing and transit stations is associated with more monthly for-hire mode pick-ups. The results of the study provide a better understanding of the interrelationships between demand for for-hire modes, built environment characteristics, and measures of access to other modes of travel. These findings can be used to improve demand forecasting for taxi, Uber, and Lyft services within large metropolitan areas.

The truck congestion problem at port gates not only restricts ports’ development, but also causes exhaust pollution. One of the main causes is uneven short-term traffic volume. As a cost-effective auxiliary method to cushion the intense traffic flow, setting a buffer area has been applied in many port, such as Shanghai Port and Tianjing Port. However, the lack of standards and methods on how to decide the scale of a buffer area, leads to either land wastage or inefficiency. In this context, this paper built a simulation model to simulate the whole complex container flow and truck flow in a port, aimed to find out the appropriate scale of a buffer area with the increase of throughput under different number of gate lanes. The results show that the traffic volume per hour shows an exponent relationship with the scale of buffer area, and the expansion of the buffer area can be an effective way only when the truck volume is less than the gates’ limit capacity. For example, in case study, when traffic volume per hour is less than 225, 7 gates are enough and the buffer area should only be able to accommodate 50 trucks at most, once the traffic volume per hour climb over 240, 9 gates are needed at least. Our research method can be used for reference in engineering design on how big a buffer area should be, or when a new gate should be built with the increase of the traffic volume.

The constant increase of global trade has established maritime transportation as an integral part of the world economy. Ports play a crucial role in maritime activities, and their ability to efficiently handle demand comprises one of the main aspects affecting their operations. This paper estimates the efficiency of container ports by developing data envelopment analysis and free disposal hull models, which utilize the number of containership berths, gantry cranes, and total container yard area as inputs; and the amount of containers handled annually as outputs. The models are applied in the ten largest container ports of the Caribbean region, an area of growing importance in global trade routes. The study describes the major sources of inefficiency and some factors causing the problematic performance. The analysis can assist port authorities in identifying aspects of their operators that require improvement and contribute in better planning, managing, and decision making of seaports.

Technological advancements in the transportation industry are changing the way planners envision, plan, and manage transportation systems. In the domain of urban freight planning, recently available real-time vehicle data and new analytical tools have opened up new opportunities for planners to enhance understanding and embrace a smarter and more rigorous policy-making process. In late 2016, the New York City Department of Transportation (NYCDOT) started a SMART Truck Management Plan as a key initiative to address the challenges of freight management and achieve the freight goals identified in NYCDOT’s 2016 Strategic Plan. A notable feature of the plan is its keen commitment to the use of new data and tools throughout the development process. This paper provides a detailed description of this data-driven planning process. Specifically, it presents how planners leveraged truck activity data to (1) measure truck movement efficiency, (2) explore the environmental factors behind off-route driving behavior, and (3) identify and prioritize projects and strategies. Through these practices, the paper seeks to initiate a conversation on how today’s cities can capitalize on new data and models to expand the capacity of their freight management systems and adopt a more robust, informed, and integrated planning process.

Rural roads are the backbone of a transportation system by carrying a significant volume of traffic, and connecting local roads and highway, and serving villages and cities. Local transportation agencies such as counties do maintenance of rural roads and often face financial constraints due to a limited budget. Rural road conditions are evaluated frequently since large agricultural vehicles, and environmental condition causes severe damage to these roads. Illinois Department of Transportation (IDOT) is currently exploring the use of smartphone applications to evaluate the roughness of the rural roads in terms of International Roughness Index (IRI). In this research Roadroid and RoadBump smartphone applications are used to assess seal coated road roughness mounted in sedan car using four different types of cell phone mounts and driving at various speed limits. The data is collected from multiple rural roads using different phone mounts like a windshield, air vent, and dashboard mounts. It has been observed that the RoadBump app gives higher IRI value compared to Roadroid measured value. The IRI measured using cell phone apps are validated using Rolling Profiler and Laser Profiler, and it was observed that Roadroid provides consistent results compared with RoadBump.

Aggregate gradation plays a vital role in the performance of an asphalt mixture as the stress distribution in the mix depends upon the aggregate gradation. The grain to grain contact depends upon the gradation and could be analyzed by examining the structure using image analysis technique. Bailey method evaluates a gradation and optimizes the gradation to improve its grain-to-grain contact and facilitate anti-rutting properties. The current study focuses on the relation between the Bailey method’s gradation parameters and the permanent deformation characteristics through internal aggregate structure characterization using image analysis technique. Image processing techniques using MATLAB software was used to determine the contact points (CP) and contact length (CL) of the gradations. Rut resistance increased with increase in coarse aggregate (CA) ratio and decrease in coarser portion of fine aggregate (FA_c) and finer portion of fine aggregate (FA_f) ratios. The mixes which have higher rut depths were found to have lower contact points and contact length. Through this study, it was concluded that Bailey method of gradation would be useful in selecting certain gradations which are resistant to permanent deformation. Correlation matrix revealed CA ratio significantly affects contact points and contact length.

Resilient modulus (M_r) is a parameter that is being used to characterize the bituminous mixture by many agencies under the linear viscoelastic region. Indian flexible pavement design procedure utilizes M_r as an input parameter for analyzing the response of a pavement structure. M_r is measured at a temperature of 35°C with a loading period of 0.1 s and a rest period of 0.9 s. However, different speeds of vehicles and traffic conditions lead to various loading periods and rest periods. Hence, it is essential to understand the material behavior at different loading and rest periods. The objective of this study is to investigate the effect of loading frequency on the resilient modulus and time lag of bituminous mixes. Results showed that time lag increases whereas M_r and percent recovery decreases with a decrease in ratio of rest period to loading period (R/L). It was observed that mixes with longer time lag have poor rut resistance.

The cement concrete (CC) and asphalt layer interface condition (IC) greatly influences the structural response of thin white-topping (TWT) pavements. Limited literature is available on the quantification of the influence of CC-asphalt IC on stresses in CC layer. An attempt was made in the present study to evaluate the influence of the IC on the structural response of TWT pavement with the help of finite element models developed for 1 m×1 m panels of different thicknesses placed over an asphalt layer. These models were analyzed for different loading conditions and different coefficient of friction (CoF) values assigned to the CC-asphalt interface. The analysis was done for isolated single panel with zero temperature gradient. The variation in CoF from 0.1 to 10 resulted in a change in the maximum tensile stress by (-)9% to 15% for different combinations of TWT pavement considered.

The construction manager/general contractor (CM/GC) is one of the modern contracting methods being advanced in the transportation industry. Although the passage of the Moving Ahead for Progress in the 21st Century Act (MAP-21) in 2012 eliminates the Special Experimental Project Number 14 (SEP-14) approval, only 19 states have legislation enabling them to use CM/GC. As other states adopt CM/GC, it is imperative to understand the current best practices and how to effectively harness the benefits of CM/GC. The aim of this paper is to present a literature review, case studies, and critical analysis of current practices of CM/GC in transportation projects to further facilitate research and applications in this domain. A total of 112 publications including journal articles, conference proceedings, case studies, and published reports were reviewed. In addition, content analysis was performed on 30 CM/GC requests for proposals/requests for qualifications located in 16 states. Based on the results of the analysis, evaluation methods, types of projects, contracts, partnering, alternative technical concepts, BIM use, benefits, challenges and limitations, and research gaps are discussed. Furthermore, the results show that although BIM is making in-roads in the transportation industry, it has not been used to garner optimum benefits on CM/GC projects and hence the paper provides a framework of how to incorporate BIM on CM/GC projects. The paper contributes to CM/GC knowledge that can serve as a foundation to facilitate further research and applications for both academia and industry stakeholders to promote the use of CM/GC.

Public–private partnership (PPP) schemes have been widely used in delivering infrastructure projects such as highways, bridges, and tunnels. The concession period as one of the most important parameters in the contractual arrangements of PPP projects is usually predetermined by governments without plausible reasoning. This practice induces frequent renegotiation of concession contracts and even project failure, since it neglects market uncertainties. To identify a reasonable length for the concession period, various decision support models have been developed by researchers and practitioners. However, existing models are usually established based on the assumption that perfect information is available for decision-making, which is unreasonable in relation to real-life projects. Therefore, focusing on highway PPPs, this paper proposes a methodology to help governments identify a specific length for the concession period in the context of imperfect information. Bargaining game theory is adopted to explore the bargaining behavior of both governments and private investors. Referring to a real highway PPP project in China, Project CB is used as a numerical example to validate the proposed method. The outcome shows that the proposed decision method is capable of identifying the optimal length of the concession period for highway PPPs. Even though the negotiation may not be conducted in most cases, the identified concession period provides a good reference for governments to define the length of the concession life prior to launching the request for bids. Limitations and future research are also discussed.

According to Beijing Subway's extensive expansion plan, the total length of subway lines will reach 1,050 km by 2020, in comparison with 638 km by the end of 2018. The new Austrian tunneling method (NATM) is widely adopted in Beijing subway construction. This paper proposes a fuzzy analytic network process to evaluate the risk for Beijing subway tunnel construction using NATM. Firstly, the risk breakdown structure (RBS)–work breakdown structure (WBS) was introduced for risk identification and 5 major risk factors of the construction projects were identified. Then the model is refined by two-round of Delphi survey and 16 sub-factors are selected. Secondly, the fuzzy analytic network process (FANP) was used to structure the FANP model hierarchically and the weights of the factors and sub-factors of the FANP were calculated. Finally, the refined model is applied to Beijing Subway Line 6 project and a series of risk control measures were taken. The risk control system was shown to be effective and ensure the success of tunnel excavation. This model is believed to serve as a valuable reference for future subway construction in Beijing using NATM in congested urban area.

This paper aims to compute the track modulus. For this purpose, it reviews the computation of the track modulus of a ballasted track, and the formula of the track modulus of a direct fixation (DF) track by Transit Co-operative Research Program. The computation has been found to be wrong, and the formulation has theoretical deficiencies because of incorrect interpretation of the definition of the track modulus although there is an equation based on the beam on an elastic foundation (BOEF) to compute the track modulus of a ballasted track. These are investigated in the paper in detail. A discreetly supported rail structure can be analyzed using numerical means. However, for a static assessment and simple configurations, the BOEF model may be applied on a DF track as an approximation method. The current industry practice applies the BOEF model on a DF track to analyze rail stress, deflection, reaction, etc. Thus, a formulation of the track modulus of a DF track is based on the BOEF model. As a sideline work, this study shows a new expression for the characteristic length, the track modulus, and the relation between track stiffness and mean track stiffness. It also presents limiting values of fastener stiffness and then a formula for fastener spacing.

This paper describes the development and successful utilization of a method to measure average railroad track crosstie-ballast interfacial pressures using pressure cells specially designed for granular materials. The test method was developed and verified using a series of laboratory tests using controlled loads applied to sections of trackbed constructed in specially designed resilient frames/boxes. For the in-track tests, a number of successive crossties were fitted with recessed pressure cells at the crosstie-ballast interface below the rail seat. Trackbed pressure measurements were collected for numerous revenue freight trains during a 20-month period of traffic including numerous loaded and empty unit trains, mixed trains, and intermodal trains. Test results are also presented for complementary static and dynamic loading using the FRA DOTX 218 comprehensive test and inspection train. Measured pressures at the crosstie-ballast interface directly below the rail seat range from 20 to 30 psi (140 to 210 kPa) for typical locomotives and heavily loaded freight cars having basically smooth wheels. Wheel loads measured by nearby wheel-impact load detectors (WILDs) were compared to trackbed pressure data for the same trains traversing the test site. Measured WILD parameters were compared to recorded tie-ballast pressures. Increases in WILD loadings, due to heavier wheel loads or increased impacts, relate favorably to increases in recorded trackbed pressures. Trackbed pressures produced by revenue freight trains compare favorably with those obtained using DOTX 218 applied wheel loads.

Regular maintenance of railroad tracks is not only essential for the proper functioning of the railway, but also significantly improves ride quality, reduces track damage, and increases track life. Deviations from the designed track geometry can result in poor ride quality and possible derailments. This research investigates the potential use of low-cost sensors aboard hi-rail vehicle to monitor automatically and continuously for inertial events caused by irregular track geometry. However, the GPS receivers of low-cost sensors do not always provide reliable position information, mostly due to signal occlusion, multipath interference, and low update rates. This study introduces a signal processing and statistical method to estimate the position of peak inertial events from multiple traversals and furthermore, the study characterizes and validates its accuracy by comparing the estimated positions of detected irregularities with the actual positions of irregularities that the railroad inspector observed.

The objective of this study is to analyze heterogeneity in users’ perception regarding service quality of passenger ferry. For this purpose, to collect the data, a structured questionnaire survey was conducted on the ferry users travelling in three different in-land water routes of Bangladesh. The data was later segmented according to five criteria, i.e., users’ income, gender, seat-type, purpose of trips, and availability of alternative modes. Finally, thirteen ordered probit models were developed to analyze heterogeneity in users’ service quality perception based on these criteria. From the analysis, we found that users’ preferences for different SQ factors are heterogeneous among different user groups. However, “Frequency”, “Travel cost”, and “Security” are observed as the significant factors irrespective of the user groups.