Amy Kim

Validation of Runway Capacity Models

Many runway capacity estimation models are available today, and developers usually claim that their models have been validated. However, information about the validation process is often limited, and different models are validated at differing levels of complexity. This paper proposes two validation methodologies for use in testing model predictions against reality. It demonstrates the two methodologies on two models [the airfield capacity model (ACM) and the runway simulator (rS)] and two airports (San Francisco International and Los Angeles International in California). The results indicate that when arrivals and departures are considered separately, both ACM and rS tend to overpredict capacities under good visibility conditions and to predict larger ranges of capacity values than are seen empirically. However, when considering total operations (arrivals and departures together), the results show that both models failed to provide good estimates of total throughput at both airports. Overall, arrival and departure capacity estimates from rS typically reflect empirical capacities better than those from ACM.

Towards setting credible speed limits: identifying factors that affect driver compliance on urban roads

Road geometry, vehicle characteristics, and weather conditions are all factors that impact a drivers perception of a safe or credible speed and, consequently, the drivers decision on whether or not to comply with the posted speed limit. In fact, the role a roads environment plays in a drivers perception of a credible speed limit is a topic that has attracted the interest of many researchers in recent years. Despite that, not many studies have considered using empirical data to investigate what features of the road environment influence a drivers compliance choice. This paper aims to address this matter by exploring the relationships between features of the road surroundings (geometric, temporal factors, and weather conditions) and driver compliance with speed limits. The paper uses data from almost 600 different urban roads in the city of Edmonton, at which over 35 million vehicle spot speeds were collected. Compliance was represented using a categorical ordered response variable, and mixed-effects-logistic-regression models were fitted. Two different models were built, one for arterials and another for collector roads. In general, the findings show that the more restricted drivers become, particularly on arterials, the more likely drivers are to comply with speed limits; potential restrictions include on-street parking and the absence of lateral shoulders. Furthermore, higher traffic activity during peak hours, and presumably on shoulder weekdays, both increase the likelihood of compliance on arterials. Similarly, posted speed limits and traffic volume are both positively correlated with compliance on both arterial and collector roads. The findings of this research provide evidence of the existence of an empirical relationship between road features and compliance, highlighting the importance of setting credible speed limits on roads and the possibility of achieving higher compliance rates through modifications to the road environment.

Development and Evaluation of Adaptive Transit Signal Priority Control with Updated Transit Delay Model

Transit signal priority (TSP) strategies are widely used to reduce bus travel delay and to increase bus service reliability. State-of-the-art strategies enable dynamic (and optimal), rather than predetermined, TSP plans to reflect real-time traffic conditions. These dynamic plans are called adaptive TSP. Existing adaptive TSP strategies normally use a performance index (PI), which is a weighted summation of all types of delays, to evaluate each candidate TSP plan and the weights to reflect the corresponding priority. The performance of an adaptive TSP depends on three factors: delay estimation, weights determination, and optimization formulation. In this context, there are three key academic contributions: (a) an enhanced bus delay estimation model based on advance detection, (b) a mechanism to adjust the PI weights dynamically to reflect the changing necessity of TSP under different conditions, and (c) TSP optimization formulated into a quadratic programming problem with an enhanced delay-based PI to obtain global optimization with the use of MATLAB solvers. In addition, an adaptive TSP simulation platform using a full-scale signal simulator, ASC/3, in VISSIM was developed. The optimal TSP plans were granted or were rejected on the basis of TSP events, such as check-in, check-out, and multiple TSP requests. Through a case study in VISSIM, this research found that, compared with conventional active TSP strategies, the new adaptive TSP strategy could further reduce bus travel time while maintaining a better balance of service on non-TSP approaches along a 7.4-km bus corridor in Edmonton, Alberta, Canada.

Reconstructing an Emergency Evacuation by Ground and Air

The May 2016 wildfire in Fort McMurray in northern Alberta, Canada—the costliest wildfire disaster in Canadian history—led to an areawide evacuation by road and air. Traffic count and flight data were used to assess the characteristics of the evacuation, including estimates of people movements by vehicle and aircraft. The vehicle counts were compared first with historic values to examine traffic patterns and were then used to create an evacuation response curve, which revealed an expected S-shaped curve and highlighted how quickly the evacuation occurred. Finally, data for people evacuated by aircraft were combined with data for people evacuated by ground vehicle to construct a curve of the cumulative number of evacuees leaving the region. This study identified several key implications for evacuation planning and operations. The decision to evacuate residents to temporary shelters in the north was instrumental in the quick removal of everyone from immediate danger via all possible exits. Although an unplanned contraflow operation added roadway capacity out of Fort McMurray, the underuse of the secondary route suggested that the management of traffic routing might have reduced congestion. The evacuation response curve emphasized the volatility of the wildfires, with the resulting evacuations occurring under conditions of a greater immediacy than hurricane evacuations. Finally, the significant role of air transportation in this evacuation indicated that multimodal emergency evacuation plans may be critical for remote communities and sparse networks. These findings may be applied to evacuation planning and policy to improve the efficiency and efficacy of evacuations.

Refinements to a Procedure for Estimating Airfield Capacity

This paper presents a method for obtaining airfield capacity estimates using historical data from FAAs Aviation System Performance Metrics (ASPM) database. The process first involves merging individual flights and quarter-hour airport runway operations data sets from ASPM to create a new data set. Data for Newark Liberty International Airport (EWR) in New Jersey and San Diego International Airport in California from 2006 to 2011 were used. Then, filters for meteorological condition, runway configuration, called rates, and fleet mix were applied to the two airport data sets. The filtered data sets were then used in a censored regression model of capacity that included queue length (number of aircraft waiting to arrive or depart) and arrival–departure throughput count splits as independent variables. These attributes were found to affect airfield capacity at statistically significant levels, and parameters had expected signs and magnitudes. Additionally, capacities under ideal conditions were found to be reasonably close to other sources. The model also confirmed that average capacities at EWR during hours when a ground delay program (GDP) was running were lower than when there was no GDP in effect. The method described in this paper could be used to more precisely quantify airfield capacities in specific conditions of particular interest to air traffic controllers and airport operators to better facilitate decisions that rely heavily on a good understanding of capacity in these conditions. The data exploration and preparation undertaken as part of the study reveal some of the finer points of the ASPM data and how they can be used in a more meaningful way for airfield capacity estimation.

Before-and-After Empirical Bayes Evaluation of Automated Mobile Speed Enforcement on Urban Arterial Roads

Speeding is a leading factor in road collisions and is found to contribute to approximately one-third of all fatal collisions. Speed enforcement is one of the most common countermeasures used to reduce speed. However, a gap exists in the literature regarding the effectiveness of automated mobile photo enforcement on urban arterial roads. This study addresses this gap using the before-and-after empirical Bayes method to account for regression to the mean effects and other confounding factors. Locally developed safety performance functions and yearly calibration factors for different collision severities were obtained by using a reference group of urban arterial roads. The evaluation period covers 8 years. Collision records, deployment information, traffic counts, and road geometric data were collected. The results showed consistent reductions in different collision severities; the reductions ranged from 14% to 20%, with the highest reductions observed for severe collisions. The enforced segments were further categorized according to site selection criteria and deployment hours to examine the effect of enforcement on collision reduction. More reductions were found at segments that had more collisions during the before period and longer deployment hours. The study also compared the safety effects of continuous and discontinuous enforcement strategies on different arterials, and the analysis revealed that continuous enforcement achieved more reductions across all severities and types of collisions. The study also investigated the spillover effects on adjacent unenforced approaches. Significant reductions were found, and these results are discussed with regard to the general and specific deterrence of the enforcement.

Exploration of the Evolution of Airport Ground Delay Programs

This study introduces a novel method of merging disparate but complementary datasets and applying machine learning techniques to ground delay program (GDP) data. More specifically, it aims to characterize GDPs with respect to changing weather forecasts, GDP plan parameters, and operational performance. The analysis aims to gain insights into GDP usage patterns (implementation and revisions), with respect to these key dimensions. It also aims to gain insights into how GDP cancelations and revisions correlate with operational efficiency and predictability. The results could be used to help traffic managers and air carriers understand complex patterns in the evolution of GDPs, so that they might, for example, better anticipate or even plan a response to a change in weather conditions. The focus is on GDPs at Newark Liberty International Airport (EWR), from 2010 through 2014. A master dataset was generated by merging several datasets on GDPs, weather forecasts, and individual flight information. Several scenarios of GDP evolution were then identified by reducing the dimensionality of the master GDP dataset, then applying cluster analysis on the lower dimensional data. It was found that GDPs at EWR can be categorized into 10 types based on weather forecasts, realized weather, GDP scope, arrival rates, and duration. The characteristics of these 10 GDP clusters were further explored by examining the relationships between GDP scenarios and their performance. It was found that GDPs under stable, low-severity weather and with large scope may score higher on the efficiency metric than expected. When GDPs called in the same weather conditions have high program rates, medium durations, and narrow scopes, capacity utilization was higher than expected—less affected flights lead to fewer cancelations and more arrivals (albeit delayed), and therefore, higher capacity utilization. Results also suggest that program rates are set more conservatively than needed for some poor weather conditions that end earlier than expected. GDPs with fewer revisions were associated with a higher predictability score but lower efficiency score. These findings can provide greater insights and knowledge about GDPs for future planning purposes. More specifically, the findings could, for example, be used to support discussion around, or even future guidance regarding, how to set and adjust GDP program rates. In future work additional data could be utilized to provide a more comprehensive operational picture of GDPs, and a wider range of performance metrics could be considered. It is also recommended that the patterns of how GDPs evolve over their lifetimes be further explored using other machine learning techniques that may provide new and useful insights.

Scheduling resources in a mobile photo enforcement program

This paper presents a model for scheduling resources in a mobile photo enforcement (MPE) program. An MPE program deploys operators driving vehicles equipped with radar and photo equipment to roadway locations where speeding and collision problems are significant. We developed a binary integer linear programming model to allocate MPE shifts over the course of a month to selected enforcement locations. These locations are pre-determined from a set of city neighborhoods chosen for enforcement through a prior multi-objective programming step. A major feature of the scheduling model is to observe the time halo effects of enforcement; our optimization model minimizes visits to an enforcement location over consecutive shifts. The model was applied to data obtained from the currently operating MPE program in Edmonton, Canada. Based on the resource allocation in neighborhoods determined in the previous stage, this scheduling model produces a scheduling plan that allocates resources to individual enforcement sites within each city neighborhood during a month. The purpose of this scheduling model, in combination with the first stage neighborhood allocation model, is to provide enforcement agencies with a tool to systematically and transparently assign limited resources in an efficient manner, providing greater efficacy in achieving program-level objectives such as reducing speeding, reducing collisions, and providing enforcement presence in areas with many vulnerable pedestrians (i.e. school zones).

Factors Affecting Classification of Road Segments into High- and Low-Speed Collision Regimes

The safety of locations operating under high-speed conditions could significantly differ from that of locations operating under low-speed conditions. Therefore, different approaches must be adopted when speed and safety are analyzed and managed at locations operating under different regimes. However, it is necessary first to understand the factors affecting the speed–collision classification of a site. Locations operating under high speeds are typically expected to have more collisions compared with locations in which speeds are low. Some locations, however, might experience a high collision rate even when speeds are low, or vice versa. This study aimed to identify the factors that affected the site classification into any of those categories by using data collected on roads in Edmonton, Alberta, Canada. Locations were divided into four speed–collision bins (high collision, high speed; high collision, low speed; low collision, high speed; low collision, low speed), and geographic information system maps of locations were produced to explore the spatial distribution of those locations. Moreover, logistic regression was used to understand the role of different factors in identifying the speed–collision bin to which a certain location belonged. The results reveal that locations with high collision rates but low speeds have a relatively high population of heavy vehicles and trucks as well as high speed variability. As for locations with low collision rates and high speeds, these sites were found to have a high level of protection through the presence of medians and shoulders with relatively low access density.

Relationship between road safety and mobile photo enforcement performance indicators: A case study of the city of Edmonton

ABSTRACT Mobile photo enforcement (MPE) programs are commonly implemented to regulate speed and improve road safety. However, most previous research focuses mainly on validating the safety effects of MPE, with very minimal discussion on the enforcement performance indicators (EPIs). Therefore, the goal of this study is to provide a better understanding of the relationship between the three selected EPIs (number of enforced sites, average check length, and number of issued tickets) and the programs safety outcomes. In total, 8 years (2005–2012) of monthly citywide data were collected and used in a generalized linear Poisson model. The results show that as the number of enforced sites and issued tickets increased, the number of speed-related collisions decreased. Also, as the average check length decreased, a greater reduction of speed-related collisions was observed. These results indicate that collision reductions were associated with a MPE program that promoted: higher spatial coverage (i.e., more enforceable locations), more frequent checks (i.e., shorter average check length), and more issued tickets. The marginal effects of enforcing 100 sites and issuing 10,000 tickets per month were calculated to be 47 and 140 fewer speed-related collisions, respectively.