The premise of many infrastructure public-private partnerships (P3s) is to deliver better life-cycle value than conventional procurement approaches. The structure of the project can either enhance or shrink project life cycle value, the so-called value “pie”. Both the size of the pie (project value created) and the size of its slices (value captured by partners) depend on a number of technical and contractual considerations. This research demonstrates an early stage life cycle evaluation of an infrastructure public-private partnership (PPP). It explicitly studies the value implications for the project partners. The discussion speaks to managers, policy-makers, and all those concerned with the development of infrastructure projects. The paper starts with an overview of the concepts central to the early stage life cycle evaluation of both general and PPP projects. It then presents the essential elements of the analysis of economic value. It further illustrates the analysis using a realistic case study of a hypothetical public-private partnership for developing and operating a major international airport.
This paper presents the most significant findings from the forensic evaluation of the long-term cracking performance of asphalt mix designs including Marshall and Superpave mixes with various performance grades of binders and RAP content of 25% total weight of aggregates. The experiment targeted comparison of permeability, stiffness, low-temperature behavior, and oxidation susceptibility of the mixes and correlation of those properties with deflection and cracking data from the six LTPP SPS-9A sections on Route 2 in Connecticut. The mechanical testing in the laboratory included measuring hydraulic conductivity by a Flexible Wall Permeameter, dynamic complex modulus by the Asphalt Mixture Performance Tester (AMPT), creep compliance and tensile strength by Indirect Tension Test, and fracture properties by Semi-Circular Beam (SCB) test. The evaluation of field performance included analysis of deflection basins and back-calculated elastic moduli from Falling Weight Deflectometer data as well as visual evaluation of surface distresses, such as cracking and weathering. The forensic laboratory testing revealed reasonable correlations between some laboratory test results and field performance. For instance, the dynamic modulus values measured by AMPT at 20 C at the highest and lowest frequency were found to be similar to the backcalculated asphalt layer moduli. The extent of transverse cracking appeared to be highly associated with the Young moduli estimated from SCB fracture energy and toughness. The amount of longitudinal wheelpath cracking correlated better with SCB fracture energy. On the other hand, neither fracture properties nor tensile strength was found to be correlated with the extent of longitudinal joint cracking observed. The laboratory testing revealed overall higher stiffness and oxidation in RAP-containing mixes. The use of those stiffer mixes, however, did not affect much load-related performance of the experimental pavement sections. On the other hand, a very fast deterioration of longitudinal joints occurred in all pavement sections, which was found most likely related to creating cold joints during paving. This phenomenon has been reduced in current practice with the introduction of wedge joints by the CTDOT.
In 2014 the Wisconsin Department of Transportation (WisDOT) and industry developed a pilot program for hot mix asphalt (HMA) with higher recycled asphalt content that required use of performance tests during mix design and production. Following the balanced mix design concept mixture tests were selected to address rutting resistance after short-term aging and durability after long-term aging. The test selected were the Hamburg Wheel Tracking (HWT) test, the semi-circular bend (SCB) test at intermediate temperature and the disc-shaped compact tension (DC(T)) test at low pavement temperatures. Asphalt binder extraction and grading from aged mix was also required. The focus of this paper is to summarize the mixture performance test and recovered binder data gathered during the pilot project on STH 77 in Ashland County, Wisconsin; suggest modifications to the SCB test procedure; and present accelerated aging protocols for continued use of performance testing in practice. Semi-circular bend test results collected during the project at 25°C did not relate well to values published in the literature or show adequate sensitivity to changes in mix properties. The effects of test temperature and an alternative analysis method are presented. Based on the results recommendations include use of a climate based approach for test temperature selection and inclusion of post peak analysis to better discriminate between mix composition and aging conditions. Accelerated long-term aging protocols involving loose mix aging at 135°C for 12 and 24 hours are compared to AASHTO R 30 compacted mix aging using recovered binder and mixture fracture properties. Results found that 12 hour loose mix aging produced similar recovered binder grading to AASHTO R 30, whereas the effect of aging on mixture fracture tests was inconclusive. The relationship between laboratory and field aging is investigated through comparison of field cores to laboratory aged plant produced mix from a project constructed in southeast Minnesota in 2006. Lastly, the laboratory performance of the high recycled and conventional mix designs are compared on the basis of mixture cracking resistance and recovered asphalt binder properties after extended aging. The high recycle mix exhibited equal or better performance relative to the conventional mix across all selected performance tests. This comparative analysis also provides an example of how the inclusion of performance testing can influence the materials selection process and produce test results indicative of improved overall performance of the mix.
In 2007, the Virginia Department of Transportation piloted a specification allowing up to 30% reclaimed asphalt pavement (RAP) in certain dense-graded asphalt surface mixtures while changing virgin binder grade requirements. The change affected only mixtures requiring an end binder grade of either PG 64-22 or PG 70-22. For mixtures specifying PG 64-22 binder, the virgin binder grade at RAP contents of 30% or less was no longer required to change. For mixtures specifying PG 70-22 binder, the virgin binder grade at RAP contents of 21-30% was no longer required to change from PG 64-22 to PG 64-28. Prior to this, both types of surface mixtures were allowed to contain only up to 20% RAP before binder grade adjustments were required. An initial laboratory study of mixtures produced under the pilot specification indicated that there were no significant differences for fatigue, rutting, and susceptibility to moisture between the higher content (21-30%) RAP mixtures and control mixtures (having 20% RAP or less). The current study evaluated the inservice performance of these mixtures after approximately 7 years and encompassed field visits and a laboratory investigation of a sample of 23 in-service pavement sites used in the initial laboratory evaluation. Cores were collected from each site and used to evaluate the binder and mixture properties. Binder data were compared to data from the original construction when available to assess the changes in properties over time. Overall study results revealed no systematic effect on field and laboratory performance with increasing RAP contents up to 30%. Test results from roadway cores showed no conclusive trends in performance with RAP content. Testing of extracted binder indicated that RAP content appears to have an influence on the rate of aging of virgin binder–RAP blends; initial grades were lower for blends having lower RAP contents, although after 7-8 years of service, all blends aged to similar grades. Binder analysis also revealed that depth within the surface layer (in this case, the top half versus the bottom half) significantly affects binder properties, with stiffness decreasing with depth. However, increasing RAP contents appeared to mitigate the difference in failure temperature before and after aging, possibly attributable to the preexisting aged composition of the RAP and its influence on the virgin binder properties.
Urban sprawl along with the wide spread of motorized vehicles downplayed the role of public transit in Canada. As such, the use of public transit in Canadian cities is less common compared to other developed countries like Europe and Japan. Obviously, the low usage of transit and high auto-dependency is no sustainable in the long run especially in larger cities. Therefore, academics, planners, community organizations among other stakeholders have been working on strategies to decrease the number of motorized trips and promote more transit usage in urban areas (Cervero & Kockelman, 1997). A major thrust of the conducted research to date have been to explore the impacts that socio-economic and demographic factors have on public transit usage (see for example Wiley et al. 2011). However, less have been done to identify if the built environment has any role to play when it comes to transit ridership. The research conducted in this paper is concerned with transit ridership for the 2011 Journey-to-Work in the four largest Canadian metropolitan areas. The analysis is focused on evaluating the degree at which the built environment, depicted by the design of the road network, affects transit ridership while controlling for socio-economic variables. The analysis also investigates whether these factors are systematic across the four studied metropolitan. Meeting these objectives will contribute to the existing body of literature and will allow planners to better understand the relationship between urban form and transit ridership in large metropolitan areas. The statistical analysis will use the Simultaneous Auto-Regressive (SAR) modeling technique given the spatial nature of the problem in hand.
Changes in ridership at individual stations on Chicago’s mass-transit rail system following fare increases in 2004, 2006 and 2009 are analyzed to determine whether the price elasticity of demand varies with the per capita income in the neighborhood surrounding each station. For two of the three fare changes, the fare elasticity becomes more inelastic for weekday trips as the neighborhood income per capita increases. However, a contradictory result is found for one of the fare increases. The relationship is even less clear for weekend trips. These mixed finding are in line with the prior literature which also found an inconsistent relationship.
Using the example of the Yonge-University-Spadina (Line 1) and Sheppard (Line 4) heavy rail transit (HRT) lines in the City of Toronto, we utilize spatial hedonic regression to isolate the effects of transit and transit-oriented development (TOD) on single-detached home values. To overcome the issue of heterogeneity in implementations of station area TOD the present research adopts the TOD typology method proposed by Higgins and Kanaroglou (2016) to segment and control for different TOD contexts directly. Results show significant land value uplift (LVU) effects for transit and TOD, though as hypothesized, these effects vary by the type of station area TOD. This suggests that transit access and TOD create different bundles of local goods and that individual sorting is at least partly responsible for the increases in land value seen within them.
In this paper, we present our results from our dynamic mixed logit route choice model based purely on trip characteristics using the next-direction method, accounting for mixed-effects of individual heterogeneity observed from panel data and serial correlation of sequential observations. This paper will highlight the process in deciding choice set for estimation, explaining our dynamic mixed logit model and report results from our dataset. We will conclude with several hypothetical uses for our forecasting methods in travel models and also implications in transportation policy. Our research goal is to develop a framework for a predictive bike sharing load balancing application by observing future trajectories from our predictive model. We can forecast the travel behaviour of cyclists ahead of time and manage the transport of bikes from station to station in a more efficient and intuitive manner.
The gathering of public transportation statistics requires a system for classifying data by mode. The majority of naming conventions have consistently recognized transit operations as 'heavy rail," "commuter rail," and "light rail" for the past 40 years (although some others still use older terms). New systems now emerging have unique characteristics, which have led some classifying organizations such as the National Transit Database (NTD) to begin using terms such as "hybrid rail" and "streetcar" to include systems which were part of commuter rail and light rail until 2011. Similarly, NTD designation of some bus operations as "bus rapid transit" and "commuter bus" also requires an updated classification system. This presentation will take inventory of all types of bus and rail mode classifications, discuss the issues associated with changing classifications, and put forth a revised classification of transit modes.
Transit user mode choice behaviour in response to service disruptions is more complex than the everyday commuting mode choice. Recent studies on this topic have been reviewed, categorized and summarized. There are three main categories and six sub-categories with regards to the timeline of transit user behaviour in response to disruptions: immediate (pre-trip and en-route), pre-planned (short-term and long-term), and gradual (short-term and long-term). The challenges and shortcomings of the studies are identified and recommendations on future studies are presented.
In the US and across most of the European Union, the population is aging. The fraction of the population in the US that is over the age of 65 has risen to 13 percent, while in the UK it is even higher at 17 percent. The US center for disease control (CDC) estimates the nearly 10 percent of adults have activity limiting arthritis, with more than twice that number suffer from physician diagnosed arthritis. And, more than 13 percent of older Americans have significantly impaired vision. These and other mild or more severe disabilities significantly reduce the mobility of older pedestrians. This research project will develop simple data collection methods that will allow pedestrians to record information about unsafe pedestrian routes, and, cutting edge machine learning and optimization techniques will produce usable real-time routes based on the specific needs of each user, and will provide transportation systems managers with tools to efficiently characterize, identify and analyze urban pedestrian networks.
In response to a growing awareness of the desire toward walkable, safe urban communities, many government agencies are evaluating their pedestrian planning decisions. With a culture shift away from suburban areas toward urban communities the issues of pedestrian safety and pedestrian improvements are the topic of many local forums. In defining these issues and developing new policies, agencies are drafting pedestrian plans. There are numerous pedestrian plans published by agencies ranging from state departments of transportation to cities and townships. The content of these plans vary widely. Few common practices exist today, and little research has been performed to identify if any common trends exist. In order to address this gap in research this study will evaluate current trends on pedestrian planning efforts by analyzing plans from twenty-five states. Individual elements analyzed include the an analysis of the span of the policy included in the plan, the funding, major components, the vision, the goals & objectives and the performance measures. This paper also discussed broad observations in terms of the potential impacts to transportation planning practices and investments in pedestrian improvements. This study found significant variability in the practices of pedestrian planning efforts across the United States in state departments of transportation. There is no one standard for how individual pedestrian plans should look, the elements or components that each plan contains and the overarching issues that a plan should address. This research identified several key trends and identified the key elements that are present in nearly half of the nation’s bicycle and pedestrian plans. This paper provides a preliminary discussion in the area of determining the quality of individual published plans and how practitioners can improve pedestrian plans to ensure that the public needs are being addressed.
Building bike lanes in Toronto has always been a challenge because the city’s top priority has always been and will remain improving traffic congestion throughout the city. Given the magnitude of the problem across the Greater Toronto Area, this approach is certainly warranted. However as a byproduct the focus on cars has led to a limited cycling network in the city, as ‘sensible locations’ for bike lanes are necessarily tied to their impact on level of service for cars and transit. The Network Robustness Index is an applied transportation geography method that uses traffic simulations to measure the impact of reduced road capacity on network travel time for cars. Since bike lanes require road capacity the method can be used to estimate the impact of a bike lane prior to implementation.
In recent years, there has been growing attention on bicycle sharing systems (BSS) as an alternative and complementary mode of transportation (Shaheen et al. 2010; Faghih-Imani et al. 2014). A bicycle-sharing system provides increased flexibility to ride a bicycle without the costs and responsibilities associated with owning a bicycle (such as the need to secure their bicycles or perform regular maintenance). With the growing installation of BSS infrastructure across the world there is a substantial interest in understanding how these systems impact the urban transportation system. Research efforts examining BSS employed a wide range of sample sizes depending on the temporal or spatial aggregation. While it is beneficial to use large sample sizes for analysis, increase in sample sizes are associated with increased data preparation effort, and longer model run times. In this context, the main objective of this paper is to investigate the impact of sample size on BSS analysis using data from New York City’s BSS (CitiBike). Specifically, the research evaluates the impact of sample size on model parameter estimates, inference measures and prediction capabilities. The findings provide analysts and planners guidelines on the “minimum” and “ideal” size of data necessary for examining BSS.
Connected vehicle (CV) and automated vehicle (AV) technologies are rapidly entering the fleet and are expected to profoundly change personal, freight, and public transport. The potential benefits to society of these technologies are immense but there are also substantial risks. This report assesses policy and planning strategies at the state, regional, and local levels that could influence private-sector AV and CV choices to positively affect societal goals. The report will be useful to staff responsible for developing plans for reacting to these technologies and is accompanied by a briefing document that may be appropriate for agency decision makers.
Life cycle assessment (LCA) for transportation fuels and vehicles, known as “well-to-wheel” analysis, is a common technique to evaluate life cycle greenhouse gas (GHG) emissions for trucks. The energy consumption and GHG emissions of medium-duty trucks are highly dependent on vehicle characteristics, such as truck configuration, payload and driving conditions (Tong et al., 2015). Few of the previous LCAs of compressed natural gas (CNG) trucks comprehensively capture the impacts of both weight and drive cycle on fuel consumption and life cycle GHG emissions. For example, the fuel consumption estimates in TIAX LLC, (2008) and Kliucininkas et al., (2012), two LCA studies of CNG trucks, do not reflect the impacts of driving conditions. The objective of this study is to estimate the energy consumption, life cycle GHG emissions, and the lifetime costs for a diesel truck and a CNG truck based on real-world medium-duty vehicle drive cycles in Toronto.
Air dispersion modelling is typically conducted across large spatio-temporal scales. However, microscopic simulations are needed to investigate the effectiveness of strategies aiming at improving air quality in near-road environments. In this work, we developed a traffic simulation linked with an emission model, and a street-canyon model to simulate nitrogen dioxide (NO2) concentrations in near-road environments. Our study is set in Montreal, Canada where a small portion of the downtown road network is simulated. The network consists of 28 links and 10 intersections. First, we explain the criteria for selection of the traffic, emission, and air quality models according to their input data and scales. Next, we address the development of a modelling chain to link these models. Then, we investigate two approaches to vehicle behavior for emission modelling: one Eulerian which considers the behavior of a link in its entirety, the second considering the trajectory of each vehicle (Lagrangian). This study examines the performance of these approaches by validating modelled air pollutant concentrations against roadside measurements.
Traffic-related air pollution is associated with a number of chronic and acute health effects (Crouse, Goldberg, Ross, Chen, & Labrèche, 2010; United States Environmental Protection Agency, 2008). In particular, a number of studies have established positive associations between various health outcomes (e.g. cancers, heart attacks, asthma) and exposure to nitrogen dioxide (NO2), an accepted marker of traffic-related air pollution (Parent et al., 2013; Wu, Wilhelm, Chung, & Ritz, 2011). In urban areas, air pollution is affected largely by the built environment, traffic composition and meteorological conditions (Weichenthal, Farrell, Goldberg, Joseph, & Hatzopoulou, 2014). In this study, we developed a traffic simulation for four consecutive road segments along a busy corridor in Montreal. The four segments exhibit different configuration and built environments, yet they share almost the same traffic characteristics. Based on the traffic simulation results, we simulated emissions of nitrogen oxides (NOx) and modeled the resulting NO2 concentrations along the road. We conducted the dispersion of traffic emissions along the four different segments under varying traffic and meteorological conditions using three different dispersion models with vastly different dispersion algorithms (CALINE 4, OSPM, and SIRANE). In-situ roadside measurements of NO2 concentrations were also conducted in order to validate and inter-compare the models.
At times, the taxicab industry has come under attack, for service quality, control of supply, tariffs and the regulatory system. Examples of this are statements such as ... The local taxi service in Toronto was labelled as the worst in North America. “Tight control on taxicab supply and government-set tariffs have resulted in consumers paying an exaggerated price for a stale and uninspired service.” More recently one writer states “...Toronto’s taxicab industry is in a state of crisis… Public perception of the entire industry is low, the government appears to have little understanding of how to manage and regulate industry efficiently, and none of these issues show any signs of improving.” The purpose of this paper is to examine the failure of closed entry taxi regulation. The paper provides a comparative analysis of the regulatory experience in Toronto and Ottawa. The next section presents the historical background of the taxicab regulation and experimentation in Toronto. Subsequently, the same is done for the City of Ottawa. Opinions on why the cities failed to deregulate are offered in the penultimate section. Finally, a few concluding remarks are made.
This paper reports on the development of SMARTPLANS, a successor of the IMULATE Integrated Urban Model (IUM) that was originally developed for the Hamilton metropolitan area. SMARTPLANS is developed as a full-fledged IUM that can be used to simulate various land use and transportation scenarios. While it shares some features with IMULATE, it also builds on ideas from the MUSSA and URBANSIM models. On the land use side, the model can simulate the decisions affecting land development, land prices, household location and firm location, whereas the transportation system of SMARTPLANS is based on the well-known four-stage model. The User Interface (UI) of the model is developed with three features in mind: 1) fine tuning the model parameters through the user interface without the need for computer programming; 2) using the UI to extend the model by including additional land use and transportation components or categories; and 3) transferability and application of the model to different urban areas. The remainder of the paper is organized as follows: a background section is provided at first to highlight the underlying principles that need to be considered when developing an IUM. This is followed by a section that describes the modeling framework of SMARTPLANS. The fourth section provides an empirical illustration of SMARTPLANS based on model parameters from Halifax, Nova Scotia. It also provides recommendations for future research.