Sevgi Erdogan - Academia.edu (original) (raw)

Papers by Sevgi Erdogan

Transportation Research Record: Journal of the Transportation Research Board, 2020

The proliferation of micromobility, evolving from station-based to dockless bikeshare programs, h... more The proliferation of micromobility, evolving from station-based to dockless bikeshare programs, has dramatically accelerated since 2017 with an influx of investment from the private sector to a new product, dockless e-scooter share. As an alternative to pedal bikes, e-scooters have become widespread across the U.S.A. owing to the unprecedented convenience they bring to commuters and travelers with electric-power propulsion and freedom from docking stations. In cities like Washington, D.C., e-scooter share can play an important role to support transportation sustainability and boost accessibility in less-connected communities. This study takes advantage of publicly available but not readily accessible e-scooter share data in Washington, D.C. for an initial view of the travel patterns and behaviors related to this new mode. The study adopted an innovative approach to scrape and process general bikeshare feed specification data in real time for e-scooters. Not only locational time seri...

Transportation Research Record: Journal of the Transportation Research Board, 2022

This paper introduces an innovative way to incentivize increased person throughput on priced high... more This paper introduces an innovative way to incentivize increased person throughput on priced highway facilities such as toll roads, express toll lanes, and high-occupancy toll (HOT) lanes, using cash rewards for carpoolers and transit riders. An exploratory evaluation of the concept is demonstrated for the I-66 HOT facility inside the Capital Beltway in Northern Virginia. The results of the analysis suggest that the concept could provide significant benefits at a relatively low cost to public agencies for cash incentives.

Socio-Economic Planning Sciences, 2021

Abstract In this paper, we apply an optimization-based approach to locate electric vehicle (EV) f... more Abstract In this paper, we apply an optimization-based approach to locate electric vehicle (EV) fast charging stations that prioritize the deployment of stations considering the designated EV corridors. A designated corridor is a segment of a highway selected for EV charging infrastructure deployment to increase the adoption of EVs. We formulate the problem such that candidate charging stations on corridors are prioritized over the other candidate station locations outside the corridor. We also introduce two new objective functions based on corridor-utilizing and corridor-weighted traffic flow concept and a new constraint that primarily focuses on the corridor building. These new objectives fit better for enabling EV corridors when the underlying objective is to locate chargers along these corridors with high density. We conduct numerical experiments using the Maryland highway network, major population centers, and select EV corridors as input. Experiment results suggest that the traditionally used objectives in the literature, maximizing traffic flow and maximizing vehicle-miles traveled (VMT), may not be well-suited for corridor building. However, a corridor-focused objective function causes the removal of many O-D pairs from consideration; hence, its impact on the overall EV flow refueled is significant as well. Experiment results also indicate that corridor prioritization is crucial for selecting a solution with more stations on corridors.

Sustainability, 2019

This study aimed to develop a model to estimate the impacts of zero-emission vehicle (ZEV) adopti... more This study aimed to develop a model to estimate the impacts of zero-emission vehicle (ZEV) adoption on CO2 emissions and to evaluate efficacy of ZEV deployment strategies in achieving greenhouse gas (GHG) emission reduction goals. We proposed a modeling scheme to represent ZEVs in four-step trip-based travel demand models. We then tested six ZEV scenarios that were a cross-combination of three ZEV ownership levels and two ZEV operating cost levels. The proposed modeling scheme and scenarios were implemented on the Maryland Statewide Transportation Model (MSTM) to analyze the impacts of different ZEV ownership and cost combinations on travel patterns and on CO2 emissions. The main findings were the following: (1) A high-ZEV ownership scenario (43.14% of households with ZEVs) could achieve about a 16% reduction in statewide carbon dioxide equivalent (CO2Eq) emissions from 2015 base year levels; and (2) CO2Eq emissions at a future year baseline (2030) (the Constrained Long-Range Plan) ...

Transportation Research Record: Journal of the Transportation Research Board, 2017

Under worldwide environmental stress, zero-emission vehicles (ZEV) are rapidly coming to market. ... more Under worldwide environmental stress, zero-emission vehicles (ZEV) are rapidly coming to market. However, it is not clear how such vehicles reduce vehicular emissions at a spatially explicit level, which is crucial for developing specific policies. This study proposed a quantitative approach to estimate the effectiveness of ZEVs in reducing emissions to support investment decisions promoting the use of ZEVs. The approach uses existing statewide travel demand and mobile emission models in an integrated framework. Scenarios are designed to measure the emissions reduction effects of ZEVs at different spatial scales (statewide, county, and roadway) and characteristics (densely and sparsely populated counties) and with various levels of market penetration and driving range limits. Results show significant spatial differentiation of the impact of ZEV deployment from county to roadway levels. Offering greater spatial detail and new insights on decision-making processes, this study describe...

Transport Policy, 2016

Fast and Intertwined Regular (FAIR) highway had been proposed as a road pricing concept that dist... more Fast and Intertwined Regular (FAIR) highway had been proposed as a road pricing concept that distributes credits to the low-income commuters. The paper models the FAIR highway operations considering the heterogeneity of commuters and lane-by-lane variations. A bilevel programming framework is established to concurrently model the FAIR operator's pricing strategies (at the upper level) and commuters' mode and lane choices (at the lower level). Various operation policy scenarios are designed and tested to evaluate the FAIR highway performance on an experimental highway corridor. A series of sensitivity analyses are conducted with respect to the key conditional factors such as travel demand levels and lane-by-lane variations. Numerical results show that compared to the do-nothing case the FAIR scheme can significantly alleviate the traffic congestion and promote the use of high-occupancy vehicles. The FAIR highway, however, has two potential disadvantages: (i) the toll level is increased when subsidizing the low-income commuters and (ii) the divergence of lane utilization is aggravated on the FAIR highways compared to the do-nothing case. The study provides insights for the decision-makers and practitioners on the potential outcomes of a FAIR highway application.

Transportation Research Record, 2006

A subarea analysis capability is needed in conjunction with dynamic network analysis models to al... more A subarea analysis capability is needed in conjunction with dynamic network analysis models to allow consideration and rapid evaluation of a large number of scenarios and to support transportation network planning and operations decisions for situations that may not require analysis on a complete network representation. Focusing on how to provide an up-todate time-dependent OD demand matrix for the subarea network, this paper describes a twostage subarea demand estimation procedure. The first stage uses path-based traffic assignment results from the original network to generate an induced OD demand matrix for the subarea network. The second stage incorporates an iterative bi-level subarea OD updating procedure to find a consistent network flow pattern by utilizing the induced OD demand information and archived traffic measurements in the subarea network. An excess-demand traffic assignment formulation is adopted to model the external trips that traverse or bypass the subarea network. This formulation allows vehicular flow to respond to traffic conditions resulting from network and operational changes in the subarea, and it can be nicely interpreted in an entropy maximization framework. The proposed procedure is illustrated in a case study using the Los Angeles subarea network extracted from the Southern California Association of Governments regional transportation planning network.

Journal of the American Planning Association, 2020

This paper presents development steps, results and the lessons learned from an application of vol... more This paper presents development steps, results and the lessons learned from an application of volume-delay functions-based dynamic network model on the Maryland Statewide Transportation Model (MSTM) project. The MSTM was developed to analyze traffic issues throughout the state including traffic in rural areas, freight movements and travel between the Baltimore and Washington Metropolitan areas. An issue facing the MSTM and other statewide models is how to account for the dynamic conditions of a large-scale network and to understand the impact of these conditions on trips which span multiple time periods. As a way of improving the understanding of daily network conditions but avoiding the problems of detailed data collection and validation, a proof of concept dynamic network model of MSTM is developed. The model is able to track individual travelers but does not simulate vehicle movements. Instead an analytical approach is used where volume-delay functions are used to update link del...

This paper examines the emissions and traffic pattern impacts of converting high-occupancy vehicl... more This paper examines the emissions and traffic pattern impacts of converting high-occupancy vehicle (HOV) lanes into high-occupancy toll (HOT) lanes in Maryland. A multi-state integrated travel demand and mobile emissions model is used to measure emission changes at multiple scales, specifically at statewide, county, corridor and individual highway lane levels. A case study is conducted on two existing HOV facilities, I-270 and US 50, in the state of Maryland. The results of the analysis show that the emission impacts of HOV-to-HOT conversion are minor at the statewide and county levels, but prominent when measured at a corridor and lane scale. Additional network performance measures indicate that the utilization of managed lanes is considerably improved after the conversion. The study illustrates the importance of conducting analyses at multiple scales, leading to important implications for policy and decision-making.

Substantial development in Dynamic traffic assignment (DTA) methods has resulted in a new generat... more Substantial development in Dynamic traffic assignment (DTA) methods has resulted in a new generation of applicable DTA-based tools for operational planning and evaluation of ITS deployment. However, the computational intensiveness of simulation-based DTA methodologies places limitations on the application of such tools to large-scale networks. To allow the consideration and rapid evaluation of large number of scenarios, a subarea analysis capability is critically needed to support transportation network planning and operations decisions for situations that may not require analysis on a complete network representation. This paper describes a two-stage subarea network extraction and zonal aggregation procedure using a simulation-based DTA model. The first stage extracts a subarea network from the original network and generates an induced OD demand matrix. Based on an iterative bi-level subarea OD estimation method, the second stage finds a consistent network flow pattern by utilizing ...

This paper addresses the problem of appropriately setting time-varying link tolls to maintain tar... more This paper addresses the problem of appropriately setting time-varying link tolls to maintain target service levels on toll links by applying a linear feedback controller similar to ones proposed for ramp metering control. Specifically, the deviations between link concentrations and a given set of target concentrations on toll links are calculated, and then link tolls are adjusted by some control regulator. With this method, a set of static link tolls can be established if the historical link concentration data are given; a set of time-varying reactive link tolls can be set if link concentrations are extracted from prevailing network states; while a set of time-varying anticipatory link tolls can be determined if predicted link concentrations are available as well. To examine and compare the performance of static, reactive and anticipatory link pricing strategies, these methods are embedded in a real-time traffic estimation and prediction system (TrEPS), which constantly publishes p...

This paper describes a simplified queue-based, capacity-constrained Dynamic Traffic Assignment (D... more This paper describes a simplified queue-based, capacity-constrained Dynamic Traffic Assignment (DTA) methodology that allows running large scale regional/statewide transportation models with 24 hour planning time horizon. The aim is to address challenges introduced by the regional/statewide models such as the need to accommodate long distance trips that typically traverse multiple geographies (i.e. regions, states) and multiple time periods (i.e. peak and off-peak periods), use of less detailed demand and supply representation due to almost prohibitively large network size, and model calibration and validation for real applications. The authors propose a simplified simulation-based DTA method that can represent congestion dynamics i.e. queue buildup and bottleneck formation without the need for detailed network coding and signal timing input through use of various traffic flow models. The model’s capability is demonstrated on the Maryland Statewide Transportation Model (MSTM), the f...

Natural or man-made hazards that require evacuation put already vulnerable populations in a more ... more Natural or man-made hazards that require evacuation put already vulnerable populations in a more precarious situation. When plans and decisions about evacuation are made, access to a private car is typically assumed, and differences in income levels across a community are rarely taken into account. The result is that carless members of a community can find themselves stranded. Low-income carless residents need alternative transportation means to reach shelters in case of an emergency. Thus, evacuation plans, decisions, and models need necessary information that identifies and locates these populations. In this study, data from the American Community Survey, U.S. Census, Internal Revenue Service, and the National Household Travel Survey are used to generate a synthetic population for Anne Arundel County, Maryland, using the copula concept. Geographic locations of low-income residents are identified within each subarea of the county (census tract) and their car ownership is estimated with a binomial logit model. The developed population synthesis method allows officials to have a more accurate account of populations for emergency planning and identify locations of shelters and triage points as well as planning carless transportation services.

The technologies and models linking individual vehicle control and network operations, despite th... more The technologies and models linking individual vehicle control and network operations, despite their critical role in determining whether automated vehicle (AV) technologies can eventually evolve to massive real-world deployments, seem to be an under-represented topic at the Automated Vehicles Symposium (AVS) in the past years. This chapter documents lecture notes of the first AVS breakout session on network modeling, which discusses the latest developments in network AV operations, modeling and simulation from academia, government, and industry perspectives. Specifically, the consensus reached in an attempt to answer questions on how to manage AV fleets in a networked environment, and control mixed traffic to optimally utilize network capacity and challenges ahead, as well as directions for research, practice and policy making are summarized.

This paper examines the effectiveness of two transit station-area development policies, park-and-... more This paper examines the effectiveness of two transit station-area development policies, park-and-ride (PNR) and transit-oriented development (TOD), by integrating a households’ residential relocation model with a conventional four-step travel demand model. Scenarios are designed on an experimental network with a single candidate metro station to assess the effectiveness of: i) building a new PNR lot; ii) replacing an existing PNR facility with TOD; and iii) developing new TOD communities. In addition, the analysis is extended to multiple stations where PNR versus TOD comparison is made considering multiple station sites with the goals of maximizing the transit patronage and minimizing the vehicle kilometers traveled (VKT) and vehicle hours of delay (VHD). The system performance in each scenario is evaluated with utilizing various measures including transit patronages, VKT, and VHD. Results show that: i) deployment of a new PNR facility or a new TOD can considerably increase the metr...

Journal of Transport & Health

Transportation Research Record: Journal of the Transportation Research Board, 2015

Bicycle-sharing programs have emerged around the world. Theoretically, the effect of bicycle shar... more Bicycle-sharing programs have emerged around the world. Theoretically, the effect of bicycle sharing on more conventional transit modes can take a substitute or complementary form. On one hand, bicycle sharing could substitute for conventional transit as a convenient and sustainable travel option. On the other hand, bicycle sharing may complement such transit by seamlessly connecting transit stations with origins and destinations and thus increase accessibility. However, the questions of how and to what extent bicycle-sharing programs affect public transit ridership remain to be answered, despite the attempts of a few empirical and quantitative studies. This study examined the impact of the Capital Bikeshare (CaBi) program on Metrorail's ridership in Washington, D.C. When CaBi trips were mapped, it was observed that Metrorail stations had been important origins and destinations for CaBi trips. Six of seven CaBi stations producing more than 500 trips were located close to Metrora...

Title of Document: TOOLS TO SUPPORT TRANSPORTATION EMISSIONS REDUCTION EFFORTS: A MULTIFACETED AP... more Title of Document: TOOLS TO SUPPORT TRANSPORTATION EMISSIONS REDUCTION EFFORTS: A MULTIFACETED APPROACH Sevgi Erdoğan, Doctor of Philosophy, 2011 Directed By: Associate Professor, Elise Miller-Hooks, Department of Civil and Environmental Engineering The transportation sector is a significant contributor to current global climatic problems, one of the most prominent problems that today’s society faces. In this dissertation, three complementary problems are addressed to support emissions reduction efforts by providing tools to help reduce demand for fossil fuels. The first problem addresses alternative fuel vehicle (AFV) fleet operations considering limited infrastructure availability and vehicle characteristics that contribute to emission reduction efforts by: supporting alternative fuel use and reducing carbon-intensive freight activity. A Green Vehicle Routing Problem (G-VRP) is formulated and techniques are proposed for its solution. These techniques will aid organizations with AF...