- Home
- Research
- Core Topics
- Transportation Demand, Economics, and Forecasting
Transportation Demand, Economics & Forecasting
Passenger Time Use Comparison Between Traditional and Autonomous Vehicles: A Latent Class and Transition Analysis
Year: 2024
As the era of autonomous vehicles (AVs) approaches, understanding how passengers’ time use during a trip may change from a traditional vehicle (non-AV) to an AV is important to the adoption and use of AVs. In this study, a latent class analysis (LCA) as well as a latent transition analysis (LTA) are adopted to investigate the choice of travel activities of individuals as passengers in a traditional vehicle, such as a car or transit, and the anticipated shift in these activities in an AV. Since individuals may perform different activities during different trip purposes, activity choices and non-AV to AV transition dynamics are explored from two different perspectives: commute trips (e.g., to work or school) and non-commute trips (e.g., leisure, errands, or medical). Findings from the LCA models show three distinct groups of individuals with varying activity preferences in a traditional vehicle and four distinct groups that could emerge in an AV. AV users exhibited a higher preference for activities such as texting/browsing social media, relaxing, and working, suggesting that AVs may offer passengers a more productive use of their travel time. Furthermore, the LTA model shows that there is a good portion of individuals who were performing one or two activities in a traditional vehicle now becoming variety seekers that could perform at least four different activities in an AV, further corroborating the findings that AVs could provide a more productive and efficient use of travel time.
Read the full publication
Exploring the role of social networks in modeling drivers’ route choice behavior
Year: 2024
Route choice is a decision that drivers face daily under varying traffic conditions. Although many studies have focused on route choice, the actual behavior considering real-world networks and the mechanisms that govern it have eluded complete characterization. Part of the difficulty has been in accounting for the growing availability of information sources that may influence drivers’ route choices, precluding collection of adequate observational data. In addition, with the continuing emergence of new forms and sources of information, such as social media with varying degrees of interactivity, engagement, and immediacy, existing frameworks and models for representing these choices and the influence of new information sources have been lacking. This study uses an agent-based modeling approach to investigate the effects of social influence on drivers’ route choices. In the model, all drivers communicate with each other in the same social network, and this information is combined with their previous experiences. The first aim of this study is to investigate the impact of social network connectivity and information exchange on route choice. The second is to characterize situations in which all drivers in the social network are satisfied with their route choices. The simulation results show that an equilibrium is achieved when all drivers in the given social network are satisfied with their current route choices.
Read the full publicationMulti-modal Machine Learning Investigation of Telework and Transit Connections
Researcher(s): Amanda Stathopoulos, Jason Soria
Year: 2024
Public transit in the U.S. has an unsettled future. The onset of the COVID-19 pandemic saw a dramatic decline in transit ridership, with agency operations, and user perceptions of safety changing significantly. However, one new factor beyond the control of agencies is playing an outsized role in transit ridership: the shifting employment patterns in the hybrid work era. Indeed, a lasting and widespread adoption of telework has emerged as a key determinant of individual transit behaviors. This study investigates the impact of teleworking on public transit ridership changes across the different transit services in the Chicago area during the pandemic, employing a random forest machine learning approach applied to large-scale survey data (n = 5637). The use of ensemble machine learning enables a data-driven investigation that is tailored for each of the three main transit service operators in Chicago (Chicago Transit Authority, Metra, and Pace). The analysis reveals that the number of teleworking days per week is a highly significant predictor of lapsed ridership. As a result, commuter-centric transit modes—such as Metra—saw the greatest declines in ridership during the pandemic. The study's findings highlight the need for transit agencies to adapt to the enduring trend of teleworking, considering its implications for future ridership and transportation equity. Policy recommendations include promoting non-commute transit use and addressing the needs of demographic groups less likely to telework. The study contributes to the understanding of how telework trends influence public transit usage and offers insights for transit agencies navigating the post-pandemic world.
Enhancing Demand Prediction: A Multi-Task Learning Approach for Taxis and TNCs
Year: 2024
Taxis and Transportation Network Companies (TNCs) are important components of the urban transportation system. An accurate short-term forecast of passenger demand can help operators better allocate taxi or TNC services to achieve supply–demand balance in real time. As a result, drivers can improve the efficiency of passenger pick-ups, thereby reducing traffic congestion and contributing to the overall sustainability of the program. Previous research has proposed sophisticated machine learning and neural-network-based models to predict the short-term demand for taxi or TNC services. However, few of them jointly consider both modes, even though the short-term demand for taxis and TNCs is closely related. By enabling information sharing between the two modes, it is possible to reduce the prediction errors for both. To improve the prediction accuracy for both modes, this study proposes a multi-task learning (MTL) model that jointly predicts the short-term demand for taxis and TNCs. The model adopts a gating mechanism that selectively shares information between the two modes to avoid negative transfer. Additionally, the model captures the second-order spatial dependency of demand by applying a graph convolutional network. To test the effectiveness of the technique, this study uses taxi and TNC demand data from Manhattan, New York, as a case study. The prediction accuracy of single-task learning and multi-task learning models are compared, and the results show that the multi-task learning approach outperforms single-task learning and benchmark models.
Read the full publication
Faster Convergence of Integrated Activity-Based Models in Dynamic Multimodal Transit Assignment Using Macroscopic Road Congestion Estimation
Year: 2024
This paper proposes an integrated framework of an activity-based behavior model and a multimodal transit assignment-simulation tool that captures road network congestion dynamics. The framework has two levels: the upper level is the demand-side activity-based model that decides individual travelers’ behavioral choices based on up-to-date information from the lower level; the lower level consists of both transit and road network estimation models on the supply side, whose inputs are trips from the upper level. The objective of this framework is to assess impacts of transit service policies, so the transit network is simulated with an agent-based multimodal hyperpath assignment model in each iteration, while the road network is mainly estimated by a macroscopic model of congestion (metamodel) instead of a simulation-based assignment model to accelerate execution time toward an equilibrated solution. Convergence under this framework is also defined from two aspects: individual choice behaviors and transit hyperpath assignment. One contribution of this paper is to incorporate the exogenous effects of road network dynamics into the integrated demand and transit assignment model, and to reduce the time to reach convergence with macroscopic modeling. This paper uses mode choice behavior as an example to demonstrate mathematical formulations and implementation procedures to reach two-level convergence. The framework is tested with the large-scale regional network of the Greater Chicago metropolitan area. The results suggest that the major advantage of the macroscopic road model is to accelerate convergence toward equilibrium when it is used to capture the traffic network congestion effects in this integrated mode choice-transit assignment framework.
Read the full publication
Where did bike-share boom? Analyzing impact of infrastructure lockdowns on bike-sharing in Chicago
Year: 2024
The COVID-19 pandemic and the subsequent measures taken to control its spread had deep impacts on transportation behavior. This paper presents a methodology for evaluating the changes in bike-share use and focuses on the identification of extraordinary trip-making activity seen in Chicago's Divvy bike-share program prior to and over the course of the COVID-19 pandemic. An analysis period from January 2018 to April 2021 was subdivided into three time periods: pre-pandemic, citywide lockdown, and post-lockdown for all census tracts within the city of Chicago. Over the analysis window defined, an anomaly detection algorithm was used to characterize the nature of bike-share usage where anomalous trips were those that fell outside of the forecasted ranges. The changes in census tract-level bike-share usage were used to interpret how static census tract-level attributes—land use, transit connectivity, demographics, and bike infrastructure—may have contributed to varying responses across the city's bike-share system. Principal component analysis was used to interpret the relationship between the dynamic trip-making characterizations and the static census tract-level attributes. Census tracts that transitioned from exhibiting below average to above average bike-share trip-making tended to have highly correlated bike infrastructure metrics. On the other hand, census tracts with consistently above average trip-making had highly correlated demographic characteristics, the values of which were consistent with their standing as key communities for service and essential workers. Overall, results indicate that post-lockdown, bike-sharing activity had grown significantly across the entire city, supporting the claim that the pandemic worked to accelerate bicycling uptake in urban areas.
Read the full publication
Microtransit adoption in the wake of the COVID-19 pandemic: Evidence from a choice experiment with transit and car commuters
Researcher(s): Amanda Stathopoulos, Jason Soria
Year: 2023
On-demand mobility platforms play an increasingly important role in urban mobility systems. Impacts are still debated, as these platforms supply personalized and optimized services while also potentially exacerbating sustainability challenges. To alleviate these concerns, microtransit projects have emerged, promising to combine the advantages of pooled on-demand rides with more sustainable fixed-route public transit services. Specifically, microtransit provides, dynamic rider-driver matching to serve demand with fewer vehicles and design optimal routes if riders accept to wait to board vehicles at curbside boarding locations. The shift to microtransit calls for new research on user behavior, motivations, and acceptability to understand demand and its role in existing mobility systems. The COVID-19 pandemic context adds an additional layer of complexity. This study investigates the potential demand for microtransit options against the background of the pandemic. We use a pivoted efficient choice experiment to study the decision-making of Israeli public transit and car commuters when offered to use novel microtransit options (sedan vs. passenger van). By estimating commuter group-specific Integrated Choice and Latent Variable models with error component terms for the microtransit alternatives, we investigate the tradeoffs related to traditional fare and travel time attributes, along with microtransit features: walking time to the pickup location, vehicle sharing, waiting time, minimum advanced reservation time, and shelter at designated boarding locations. We analyzed two latent constructs: the attitudes toward sharing and the experiences and risk perceptions related to the COVID-19 pandemic. The results reveal three key takeaways. (1) New modal attributes significantly affect the utility of the microtransit alternatives, with a notable aversion to walking and waiting among drivers; (2) car and transit commuters have structural differences in attribute elasticities; (3) significant differences are noted for the magnitude of the latent variable effects. Sharing experience and COVID Comfort play a key role for drivers evaluating the choice of microtransit.
In-person, pick up or delivery? Evolving patterns of household spending behavior through the early reopening phase of the COVID-19 pandemic
Year: 2023
Consumer reactions to COVID-19 pandemic disruptions have been varied, including modifications in spending frequency, amount, product categories and delivery channels. This study analyzes spending data from a sample of 720 U.S. households during the start of deconfinement and early vaccine rollout to understand changes in spending and behavior one year into the pandemic. This paper finds that overall spending is similar to pre-pandemic levels, except for a 28% decline in prepared food spending. More educated and higher income households with children have shifted away from in-person spending, whereas politically conservative respondents are more likely to shop in-person and via pickup.
Read the full publication
Does ridesourcing respond to unplanned rail disruptions? A natural experiment analysis of mobility resilience and disparity
Year: 2023
Urban rail transit networks provide critical access to opportunities and livelihood in many urban systems. Ensuring that these services are resilient (that is, exhibiting efficient responses to and recovery from disruptions) is a key economic and social priority. Increasingly, the ability of urban rail systems to cope with disruptions is a function of a complex patchwork of mobility options, wherein alternative modes can complement and fill occurring service gaps. This study analyzes the role of ridesourcing in providing adaptive mobility capacity that could be leveraged to fill no-notice gaps in rail transit services, addressing the question of distributional impacts of resilience. Using a natural experiment, we systematically identify 28 major transit disruptions over the period of one year in Chicago and match them, both temporally and spatially, with ridesourcing trip data. Using multilevel mixed modeling, we quantify variation in the adaptive use of on-demand mobility across the racially and economically diverse city of Chicago. Our findings show that the gap-filling potential of adaptive ridesourcing during rail transit disruptions is significantly influenced by the station-, community-, and district-level factors. Specifically, greater shifts to ridesourcing occur during weekdays, nonholidays, and more severe disruptions, in community areas that have higher percentages of white residents and transit commuters, and in the more affluent North district of the city. These findings suggest that while ridesourcing appears to provide adaptive capacity during rail disruptions, its benefits do not appear to be equitable for lower-income communities of color that already experience limited mobility options. Research implications for mobility operator collaborations to support mobility as a service are discussed. This study builds a more comprehensive understanding of transit service resilience, variation in vulnerability, and the complementarity of ridesourcing to existing transport networks during disruptions.
Read the full publication
Requiem for transit ridership? An examination of who abandoned, who will return, and who will ride more with mobility as a service
Year: 2023
To adhere to health regulations and reduce the risks associated with the COVID-19 pandemic, employers, mobility operators, and travelers alike adopted new strategies such as teleworking, rigorous sanitation, and social distancing. In this research, we examine the individual-level factors contributing to transit ridership abandonment and return decisions. We utilize comprehensive survey-based data of transit users in the Chicago metropolitan area (N = 5648) collected prior to reopening. We investigate three ridership behaviors, namely (1) discontinued public transit ridership, (2) the intent to return to pre-pandemic transit ridership levels once health concerns are alleviated, and (3) the likelihood of using public transit more often if its fare systems are integrated with other mobility services such as ridehailing and micromobility. Examining the role of sociodemographics, employment characteristics, transit investment priorities, and travel behavior before and during the pandemic, this research reveals fine-grained details about transit usage decline, as well as future intentions. The results indicate that teleworking, unemployment, and vehicle access are the major factors behind discontinued transit ridership. Analysis of race, ethnicity, and gender effects reveals that vulnerable users often have a higher risk of abandonment coupled with a lower likelihood of returning. These results point to the need for transit agencies to consider the specific concerns of ethnic/racial minorities and women. Encouragingly, there is an opportunity for agencies to attract more ridership with fare integration. Several respondent segments would use transit more if fare systems are integrated with ridehailing and micromobility, highlighting the importance of lowering the barriers to accessing these mobility services. This research informs several policies that can be adopted by transit agencies and other mobility providers. We discuss the importance of an equitable return to transit, possibilities for Mobility-as-a-Service with fare integration as a starting point and stress the significance of teleworking in future transit policies.
Shifting Mobility Behaviors in Unprecedented Times: A Multigroup MIMIC Model Investigating Intentions to Use On-Demand Ride Services During the COVID-19 Pandemic
Year: 2023
The spread of COVID-19 has been a major disruptive force in people’s everyday lives and mobility behavior. The demand for on-demand ride services, such as taxis and ridehailing, has been specifically affected given both restrictions in service operations and users’ concerns about virus transmission in shared vehicles. In the early months of the pandemic, demand for these modes decreased by as much as 80%. This study examines intentions to use on-demand ride services in the early lockdown stage of the pandemic in the United States, a period of unprecedented mobility reductions, changing household routines and transforming travel behaviors. Using data from a survey disseminated in June 2020 to 700 U.S. respondents, we use multigroup MIMIC (Multiple Indicator Multiple Cause) models to investigate the stated shift in intentions to use on-demand modes of travel. By using group-based segmentation we control for variation in ridership intentions according to personal, household, attitudinal factors, and pandemic experiences. The results point to a reduction across the board in the likelihood of using on-demand mobility associated with a significant COVID-19 effect. Beyond this general decrease, several groups are found to have more positive intentions, including younger adults, urban residents, graduate-degree holders, and people of Hispanic, Latino, Asian, and Pacific Islander ethnicities/races. The attitudinal effect of ‘‘tech-savviness’’ drives higher user intentions, revealing indirect effects of gender, education, and age. Multigroup analysis provides further evidence of potential COVID-triggered shifts in on-demand ridership intentions. The most significant drops in likelihood are observed for younger respondents (below 45), Black compared with all other racial/ethnic status, and for past users of on-demand mobility. This latter result is somewhat surprising, as riders who are younger and more experienced with on-demand travel are more likely to have been users in the past, but also more likely to reduce use during the pandemic. To conclude, we discuss the need to investigate pandemic experiences, risk attitudes, and circumstances to understand evolving mobility behavior and specific service model impacts.
Read the full publication
A network-based discrete choice model for decision-based design
Year: 2023
Customer preference modelling has been widely used to aid engineering design decisions on the selection and configuration of design attributes. Recently, network analysis approaches, such as the exponential random graph model (ERGM), have been increasingly used in this field. While the ERGM-based approach has the new capability of modelling the effects of interactions and interdependencies (e.g., social relationships among customers) on customers' decisions via network structures (e.g., using triangles to model peer influence), existing research can only model customers' consideration decisions, and it cannot predict individual customer's choices, as what the traditional utility-based discrete choice models (DCMs) do. However, the ability to make choice predictions is essential to predicting market demand, which forms the basis of decision-based design (DBD). This paper fills this gap by developing a novel ERGM-based approach for choice prediction. This is the first time that a network-based model can explicitly compute the probability of an alternative being chosen from a choice set. Using a large-scale customer-revealed choice database, this research studies the customer preferences estimated from the ERGM-based choice models with and without network structures and evaluates their predictive performance of market demand, benchmarking the multinomial logit (MNL) model, a traditional DCM. The results show that the proposed ERGM-based choice modelling achieves higher accuracy in predicting both individual choice behaviours and market share ranking than the MNL model, which is mathematically equivalent to ERGM when no network structures are included. The insights obtained from this study further extend the DBD framework by allowing explicit modelling of interactions among entities (i.e., customers and products) using network representations.
Virtual leisure activity engagement: The role of childhood technology experience
Year: 2023
Activity engagement and travel behavior are dynamic concepts that are inherently complex. Recently, the generational transition from baby boomers as the largest generational cohort in the US to millennials has added a layer of complexity to our understanding of activity engagement and travel behavior. The challenge stems in part from the fact that millennials are “digital natives” or the first generation to generally be exposed to technology during childhood, with implications for adulthood preferences, activity and travel behavior. This does come with a potential benefit for the environment via reduced travel more generally and car travel more specifically. In fact, “Peak car” is one hypothesized manifestation of changes brought about by the generational transition that is gaining importance in the literature. Another, less discussed manifestation pertains to the changes in the balance between virtual and physical activities. In this study, we examine the association between millennial-specific factors, such as the childhood experience with technology, and adulthood engagement in virtual rather than physical activities. We do so by applying a comprehensive life-stage approach borrowed from a larger study examining the impacts of the generational shift on travel behavior. Results suggest that individuals who were more technologically-engaged during childhood are associated with higher levels of technology engagement and higher virtual than physical leisure activities as adults. Such results provide important insights at the policy level, helping inform efforts aimed at promoting sustainable travel and provide researchers with a comprehensive framework to tackle research questions related to young people's travel behavior. This will only gain importance as more “digital natives” become transport users.
Read the full publication
https://www.scholars.northwestern.edu/en/publications/ride-hail-vehicle-routing-river-as-a-congestion-game
Year: 2023
The RIde-hail VEhicle Routing (RIVER) problem describes how drivers in a ride-hail market form a dynamic routing strategy according to the expected reward in each zone of the market. We model this decision-making problem as a Markov decision process (MDP), and view the drivers as playing an MDP routing game, with “congestion” induced by competitive matching in a zone. The meeting probability (i.e., the probability of successfully picking up a passenger after cruising in a zone for one period) is derived from a physical model, which is specified for street-hail and e-hail service modes separately, and calibrated with simulation data. We define a Wardrop equilibrium for the MDP routing game, and then prove it exists and can be obtained by solving a fixed-point problem. We further show a system optimum – corresponding to the maximum total expected reward accumulated over time by all drivers – can be achieved if drivers make individual routing decisions according to a cooperative reward rather than the personal reward. In other words, the proposed cooperative strategy “decentralizes” the system optimal solution in the MDP routing game. The results from numerical experiments, including a case study of Chicago, indicate the service mode plays a critical role in shaping system performance. While e-hail enjoys a higher fleet utilization rate than street-hail thanks to its more advanced matching technology, it may lead to a more uneven spatial distribution of vacant vehicle supply. As expected, cooperative routing improves system performance in terms of both total reward and equal distribution of supply. Yet, its benefit is much stronger in e-hail than in street-hail, especially when the supply is overly abundant. We also find, when a local demand surge occurs, the cooperative rewards rise in sync, similar to surge pricing on the supply side. Interestingly, the impact of the price surge spreads broadly in space, propagating far beyond the epicenter of the demand surge.
Read the full publication
Managing bottleneck congestion with tradable credits under asymmetric transaction cost
Quantifying the competitiveness of transit relative to taxi with multifaceted data
Year: 2022
This paper proposes an assessment framework to quantify the competitiveness of transit relative to a taxi-like service. The framework centers on a transit route builder, which searches, using a hyperpath-based algorithm, for the best available transit route that matches the origin and the destination of a given taxi trip. Based on the optimal transit route, we then measure the relative competitiveness of the transit service according to the preference of a rational traveler, which is determined by the generalized cost defined by fare, in-vehicle travel time and other service attributes. The framework is evaluated using a case study constructed with multifaceted data sources collected in Shenzhen, China. The results show that, while 90% of all taxi trips are faster than its best alternative transit option, only about 36% is shorter. Also, the relative competitiveness of transit decreases with the passenger's value of time, and increases with the average trip distance. We also find that the preference of the middle-income passengers for transit is the most sensitive to the changes in trip distance, mode (bus or rail) and fare.
Read the full publication
A multi-hop control scheme for traffic management
Year: 2021
We propose a multi-hop control scheme (MHCS) that aims to route traffic through a set of designated intermediate checkpoints (ICs). Because travelers are allowed to freely choose routes for each “hop” that connects real (origin and destination) and ICs, MHCS promises to keep intervention at a more tolerable level, compared to conventional route-based control schemes. The MHCS problem has a natural bi-level structure: the upper level attempts to minimize congestion by adjusting the hopping ratios, which are then used in the lower level problem to route travelers according to user equilibrium conditions. Accordingly, we formulate the problem as a mathematical program with equilibrium constraints (MPEC), establish its solution existence, and propose to solve it using a sensitivity analysis based algorithm. We examine sixteen heuristic rules for choosing ICs. Results based on five hundred experiments suggest that selecting the most used and most congested nodes at system optimum as the ICs delivered the largest travel time savings. Based on this finding, a set of efficient ICs are identified and adopted to test the potential of a full-scale scheme. The results from numerical experiments indicate that these checkpoints are highly effective in reducing traffic congestion at a reasonable cost of control and unfairness. In particular, they outperform, by a large margin, other choices such as most congested nodes at user equilibrium.
Inter-platform competition in a regulated ride-hail market with pooling
Year: 2021
This paper studies an aggregate ride-hail market in which two platforms compete with each other, as well as with transit, under different supply and regulatory conditions. The duopoly is built on a general market equilibrium model that explicitly characterizes the physical matching process, including pairing two passengers for a pooling ride. Depending on whether drivers’ work affiliation with a platform is exclusive or not, the duopoly is said to have a single- or multi-homing supply mode. We describe the outcome of the duopoly pricing game as a Nash Equilibrium (NE) and solve it by transforming it into a variational inequality problem (VIP). When a regulatory constraint is imposed, the duopoly equilibrium becomes a generalized NE, which corresponds to a quasi VIP. We show that multi-homing may lead to disastrous outcomes in an unregulated duopoly and demonstrate it through numerical experiments constructed using data from Chicago. Specifically, passenger and driver surplus, as well as platform profits, are all significantly lower in a multi-homing duopoly than in a single-homing counterpart. This disaster arises because (i) the multi-homing duopoly is locked in a self-destructive pricing war analogous to the tragedy of the commons; and (ii) the competition among passengers limits economy of scale in trip production. We show that the negative consequences of this tragedy can be mitigated by (i) discouraging multi-homing behavior; (ii) imposing a minimum wage on both platforms; and (iii) encouraging the platforms to specialize in different services. The results also show the efficiency in matching passengers and drivers is a crucial asset for a platform's competitiveness, more so in a multi-homing duopoly. In general, the platform with a higher matching efficiency ends up making more money and providing a better level of service.
Read the full publication
Investigating socio-spatial differences between solo ridehailing and pooled rides in diverse communities
Year: 2021
Transformative mobility services present both considerable opportunities and challenges for urban mobility systems. Increasing attention is being paid to ridehailing platforms and connections between demand and continuous innovation in service features; one of these features is dynamic ride-pooling. To disentangle how ridehailing impacts existing transportation networks and its ability to support economic vitality and community livability it is essential to consider the distribution of demand across diverse communities. In this paper we expand the literature on ridehailing demand by exploring community variation and spatial dependence in ridehailing use. Specifically, we investigate the diffusion and role of solo requests versus ride-pooling to shed light on how different mobility services, with different environmental and accessibility implications, are used by diverse communities. This paper employs a Social Disadvantage Index, Transit Access Analysis, and a Spatial Durbin Model to investigate the influence of both local and spatial spillover effects on the demand for shared and solo ridehailing. The analysis of 127 million ridehailing rides, of which 15% are pooled, confirms the presence of spatial effects. Results indicate that density and vibrancy variables have analogue effects, both direct and indirect, on demand for solo vs pooled rides. Instead, our analysis reveals significant contrasting effects for socio-economic disadvantage, which is positively correlated with ride-pooling and negatively with solo rides. Additionally, we find that higher rail transit access is associated with higher demand for both solo and pooled ridehailing along with substantial spatial spillovers. We discuss implications for policy, operations and research related to the novel insight on how pooled ridesourcing relate to geography, living conditions, and transit interactions.
Read the full publication
Dueling emergencies: Flood evacuation ridesharing during the COVID-19 pandemic
Year: 2021
Volunteered sharing of resources is often observed in response to disaster events. During evacuations the sharing of resources and vehicles is a crucial mechanism for expanding critical capacity and enabling inclusive disaster response. This paper examines the complexity of rideshare decision-making in the wake of simultaneous emergencies. Specifically, the need for physical distancing measures during the coronavirus (COVID-19) pandemic complicates face-to-face resource sharing between strangers. The ability of on-demand ridesharing to provide emergency transportation to individuals without access to alternatives calls for an understanding of how evacuees weigh risks of contagion against benefits of spontaneous resource sharing. In this research, we examine both sociodemographic and situational factors that contribute to a willingness to share flood evacuation rides with strangers during the COVID-19 pandemic. We hypothesize that the willingness to share is significantly correlated with traditional emergency resource sharing motivations and current COVID-19 risk factors. To test these hypotheses, we distributed an online survey during the pandemic surge in July 2020 to 600 individuals in three midwestern and three southern states in the United States with high risk of flooding. We estimate a random parameter multinomial logit model to determine the willingness to share a ride as a driver or passenger. Our findings show that willingness to share evacuation rides is associated with individual sociodemographics (such as being female, under 36 years old, Black, or republican-identifying) and the social environment (such as households with children, social network proximity, and neighborly sharing attitudes). Moreover, our findings suggest higher levels of income, COVID-19 threat perception, evacuation fear, and household preparedness all correspond with a lower willingness to share rides. We discuss the broader implications of emergency on-demand mobility during concurrent disasters to formulate strategies for transportation agencies and on-demand ridehailing providers.
Read the full publication
Partial Demand Information and Commitment in Dynamic Transportation Procurement
Year: 2020
This paper analyzes a decision process of planning transportation procurement for a distribution lane given limited information regarding future demand for transportation services and a specified commitment horizon for procurement contracts. In contrast to variants in which either full demand information or no demand information is known over the planning horizon, our work considers the value of demand visibility for a short horizon in the future (i.e., the value of partial information). Our work also considers a commitment horizon that is much shorter than the planning horizon. We show that the availability of partial information fundamentally changes the contracting policies in the presence of such commitment horizons, and if used optimally, this information can be highly valuable. Partial visibility of demand can increase the willingness of the decision maker to commit to contracts and expand the range of capacity levels selected in settings where the capacity level of a contract is a decision variable. We also identify settings in which the value of partial information is negligible, reducing the incentive of managers to acquire additional demand information for future periods. Finally, we show that with seasonal demand, information is leveraged by properly coordinating with expected demand shocks (e.g., Black Friday) using tailored strategies.
Read the full publication
Dynamic trucking equilibrium through a freight exchange
Date: 2020
This paper proposes a new hyperpath-based dynamic trucking equilibrium (DTE) assignment model. Unlike existing freight assignment models, we focus on the interactions between individual truck operators that solely compete for loads advertised on an online freight exchange. The competitors are assumed to follow optimal bidding and routing strategies – represented using a hyperpath – to maximize their expected profit. The proposed DTE model (1) predicts system-wide truck flows (including empty truck flows), (2) identifies efficiency improvements gained by network-wide visibility, and (3) lays the foundation for building a system optimal model. We rewrite the DTE conditions as a variational inequality problem (VIP) and discuss the analytical properties of the formulation, including solution existence. A heuristic solution algorithm is developed to solve the VIP, which consists of three modules: a dynamic network loading procedure for mapping hyperpath flows onto the freight network, a column generation scheme for creating hyperpaths as needed, and a method of successive average for equilibrating profits on existing hyperpaths. The model and the solution algorithm are validated by numerical experiments constructed from empirical data collected in China. The results show that the DTE solutions outperform with wide margin the benchmark solutions that either ignore inter-truck interactions or operate trucks according to suboptimal routing and bidding decisions.
Read the full publication
Social media effects on sustainable mobility opinion diffusion: Model framework and implications for behavior change
Year: 2020
Opinions regarding emergent sustainable transportation alternatives, such as bikeshare and e-scooters, and more traditional green alternatives like public transit, spread through social networks via opinion diffusion mechanisms, like word-of-mouth and mass media. The impact of social media on diffusion of sustainable mobility opinions is not well-understood given the present lack of data. To address this gap, this paper introduces a modeling framework for the impact of social media on opinion diffusion. Inspired by Roger's diffusion theory, the framework applies different learning mechanisms (e.g., word-of-mouth and mass media) in network architectures to explore the effects of network topology on acceptance of green travel alternatives using conceptual idealizations of the complex processes involved in diffusion interactions. We present a dynamic agent-based simulation methodology capturing the impact of information and communications technology (ICT) like social media on diffusion of environmentally friendly travel mode consideration through social networks. The agent-based models provide visual comparisons of the effects of network structure and social media influence on opinion diffusion, the way opinions spread, and which agents exhibit the strongest influence. We identify types of social media influencers that most effectively encourage adoption of sustainable transportation alternatives and present an illustrative framework of the mechanisms that drive opinion diffusion. Exploratory findings suggest that: (1) scale-free networks provide the slowest initial diffusion rate but the greatest overall diffusion over time, (2) the most effective behavior incentivization strategies depend on network structure, (3) in scale-free networks, increasing the number of initial opinion leaders improves diffusion, while increasing the number of communication encounters within the network over the first year following product deployment does not noticeably improve diffusion, and (4) providing smaller financial incentives to a greater number of opinion leaders is the best strategy.
Read the full publication
Paired-line hybrid transit design considering spatial heterogeneity
Year: 2020
This study attempts to incorporate spatial heterogeneity into the optimal design of paired-line hybrid transit systems, which aims to strike a better balance between accessibility and efficiency by leveraging the flexibility of a demand adaptive service. A simple trip production and distribution model is introduced to differentiate the central business district (CBD) of a city from its periphery. To cope with the heterogeneous demand pattern, the transit system is also configured differently inside and outside the CBD, for both its fixed route and demand adaptive services. Allowing the supply heterogeneity complicates transit users’ route choice modeling considerably. As a result, user costs must be estimated separately for six subregions that constitute the feasible set of the fixed route headway. Each subregion corresponds to a unique route choice behavior, hence leading to a distinctive design model that is formulated as a mixed integer program and solved by a commercial solver. Results of numerical experiments show that concentrating demand in the CBD significantly reduces the average system cost, and this benefit increases as the average demand density becomes larger. Also, recognizing demand heterogeneity and responding to it with a tailored design can be highly beneficial. However, this benefit diminishes as the average demand density increases.
Read the full publication
Pricing carpool rides based on schedule displacement
Year: 2020
This paper considers a carpool matching (CaMa) problem in which participants price shared rides based on both operating cost and schedule displacement (i.e., the absolute difference between the desired and actual arrival times). By reporting their valuation of this displacement, each participant in effect bids for every possible shared ride that generates a unique value to her. The CaMa problem can be formulated as a mixed integer program (MIP) that maximizes the social welfare by choosing matching pairs and a departure time for each pair. We show the optimal departure time can be determined for each pair a priori, independent of the matching problem. This result reduces the CaMa problem to a standard bipartite matching problem. We prove that the classical Vickrey-Clarke-Groves (VCG) pricing policy ensures no participant is worse off or has the incentive to misreport their valuation of schedule displacement. To control the large deficit created by the VCG policy, we develop a single-side reward (SSR) pricing policy, which only compensates participants who are forced by the system to endure a schedule displacement. Under the assumption of overpricing tendency (i.e., no participant would want to underreport their value), we show the SSR policy not only generates substantial profits, but also retains the other desired properties of the VCG policy, notably truthful reporting. Even though it cannot rule out underreporting, our simulation experiments confirm that the SSR policy is a robust and deficit-free alternative to the VCG policy. Specifically, we find that (1) underreporting is not a practical concern for a carpool platform as it never reduces the number of matched pairs and its impact on profits is largely negligible; and (2) participants have very little to gain by underreporting their value.
Read the full publication
On-demand ridesourcing for urban emergency evacuation events: An exploration of message content, emotionality, and intersectionality
Researcher(s): Elisa Borowski, Amanda Stathopoulos
Year: 2020
Evacuation mode choice has been researched over the past decade for disaster management and planning, focusing primarily on established modes such as personal automobiles, carpooling, and transit. Recently, however, on-demand ridesourcing has become a viable mode alternative, most notably through the growth of major transportation network companies, such as Uber and Lyft. The availability of this new transportation option is expected to have important implications for adaptive disaster response. The goal of this work is to investigate the influence of internal and external contextual factors on preferred ridesourcing applications during small-scale urban evacuations. A case study was conducted in the three most populous metropolitan areas in the United States. Data were collected using an internet-based stated preference survey, and a discrete choice model was estimated to analyze the 185 responses. Determinants of on-demand ridesourcing for evacuation include internal factors, such as interactions between race, gender, and income, and external contextual factors, such as the evacuation notification source, consequence severity, immediacy, evacuation distance, unfamiliarity of surroundings, and traveling with others. Findings are illustrated through three ridesourcing applications based on specific evacuation needs. Policy recommendations are provided for the design of equitable evacuation services, soft policy communication strategies, and public-private partnerships.
The covering path problem on a grid
Year: 2019
This paper introduces the covering path problem on a grid (CPPG) that finds the cost-minimizing path connecting a subset of points in a grid such that each point that needs to be covered is within a predetermined distance of a point from the chosen subset. We leverage the geometric properties of the grid graph, which captures the road network structure in many transportation problems, including our motivating setting of school bus routing. As defined in this paper, the CPPG is a biobjective optimization problem comprising one cost term related to path length and one cost term related to stop count. We develop a trade-off constraint, which quantifies the trade-off between path length and stop count and provides a lower bound for the biobjective optimization problem. We introduce simple construction techniques to provide feasible paths that match the lower bound within a constant factor. Importantly, this solution approach uses transformations of the general CPPG to either a discrete CPPG or continuous CPPG based on the value of the coverage radius. For both the discrete and continuous versions, we provide fast constantfactor approximations, thus solving the general CPPG.
Read the full publication
Evaluating the impact of spatio-temporal demand forecast aggregation on the operational performance of shared autonomous mobility fleets
Year: 2019
Fleet operators rely on forecasts of future user requests to reposition empty vehicles and efficiently operate their vehicle fleets. In the context of an on-demand shared-use autonomous vehicle (AV) mobility service (SAMS), this study analyzes the trade-off that arises when selecting a spatio-temporal demand forecast aggregation level to support the operation of a SAMS fleet. In general, when short-term forecasts of user requests are intended for a finer space–time discretization, they tend to become less reliable. However, holding reliability constant, more disaggregate forecasts provide more valuable information to fleet operators. To explore this trade-off, this study presents a flexible methodological framework to evaluate and quantify the impact of spatio-temporal demand forecast aggregation on the operational efficiency of a SAMS fleet. At the core of the methodological framework is an agent-based simulation that requires a demand forecasting method and a SAMS fleet operational strategy. This study employs an offline demand forecasting method, and an online joint AV-user assignment and empty AV repositioning strategy. Using this forecasting method and fleet operational strategy, as well as Manhattan, NY taxi data, this study simulates the operations of a SAMS fleet across various spatio-temporal aggregation levels. Results indicate that as demand forecasts (and subregions) become more spatially disaggregate, fleet performance improves, in terms of user wait time and empty fleet miles. This finding comes despite demand forecast quality decreasing as subregions become more spatially disaggregate. Additionally, results indicate the SAMS fleet significantly benefits from higher quality demand forecasts, especially at more disaggregate levels.
Read the full publication
A Comment on "Subsidization of Urban Public Transport and the Mohring Effect"
Researcher(s): Ian Savage
Year: 2010
Van Reeven (2008) argues that the Mohring effect is not relevant to the determination of transit subsidies because a profit-maximizing monopolist would supply frequencies that are the same as, or greater than, those that are socially optimal. We find that his results depend on the reduction or elimination of the effect of fares on demand, causing optimal prices to be indeterminate within broad ranges. Consequently, his model is an unsatisfactory tool for discussing subsidies in general, and the optimal combination of fare and frequency in particular.
Pricing Congestion for Arriving Flights at Chicago O'Hare Airport
Researcher(s): Ian Savage
Year: 2010
This paper estimates congestion feeds for arriving flights at Chicago O’Hare Airport. The analysis finds that the level of congestion is only about a fifth of the magnitude of the congestion associated with departing flights. Congestion is much worse in poor weather conditions, and mitigating these weather delays is a primary objective of the current program to reconfigure the airfield. The analysis finds that the nonlinearities inherent in models of congestion mean that even a very modest change in flight patterns reduces delays and congestion fees quite considerably.
Existence of Self-financing and Pareto-improving Congestion Pricing: Impact of Value of Time Distribution
Researcher(s): Yu Nie, Yang Liu
Year: 2010
This paper considers a static congestion pricing model in which travelers select a mode from either, driving on highway or taking public transit, to minimize a combination of travel time, operating cost and toll. The focus is to examine how travelers’ value of time (VOT), which is continuously distributed in a population, affects the existence of a pricing-refunding scheme that is both self-financing (i.e. requiring no external subsidy) and Pareto-improving (i.e. reducing system travel time while making nobody worse off). A condition that insures the existence of a self-financing and Pareto-improving (SFPI) toll scheme is derived. Our derivation reveals that the roll authority can select a proper SFPI scheme to distribute the benefits from congestion pricing through a credit-based pricing scheme. Under mild assumptions, we prove that an SFPI toll always exists for concave VOT functions, of which the linear function corresponding to the uniform distribution is a special case. Existence conditions are also established for a class of rational functions. These results can be used to analyze more realistic VOT distributions such as log-normal distribution. A useful implication of our analysis is that the existence of an SFPI scheme is not guaranteed for general functional forms. Thus, external subsidies may be required to ensure Pareto-improving, even if policy-makers are willing to return all toll revenues to road users.
Improving Our Understanding of How Pricing and Congestion Affect Travel Demand
Researcher(s): Hani Mahmassani
Year: 2010
Dr. Mahmassani serves as co-PI on this study, awarded by the National Academy of Science to PB Americas, Inc. (with Northwestern University Transportation Center). The work plan can be conceptualized in three interconnected levels of behavioral rigor and practical application, with varying levels of sophistication:
Level 1 – Behavioral Foundations. The first level corresponds to behavioral models intended for a deep understanding and quantitative exploration of travel behavior. These models seek to address the full range of possible short and long‐term responses, but also may focus on select choice dimensions (for example, route and departure time choices, or usual workplace location choice).
Level 2 – Advanced Operational. The second level relates to relatively advanced, yet operational, Activity‐Based (AB) models, integrated with state of the art DTA (Dynamic Traffic Assignment) models. These models allows for the incorporation of a wide range of possible short‐ and long‐term responses that are embedded in the choice hierarchy. The integrity of operational models requires that each and every choice dimension should be allocated a proper “slot” in the hierarchy, with upward and downward linkages to related choices. Operational/computing time requirements often limit the total number of choice dimensions and alternatives, but this restriction is lessening with time.
Level 3 – Opportunities for Prevailing Practice. The third level relates to existing model systems used by most of MPOs and state DOTs, in the form of aggregate trip-based models (frequently referred to as 4-step models). Though rather restrictive in design, such models offer opportunities for meaningful and immediate contributions to the state of travel demand modeling practice. A serious restriction of 4‐step models is that these rely on static assignment procedures. Static assignments generate only crude average travel time and cost variables, and reliability can be incorporated only through certain proxies.
The SHRP 2 C04 project has completed an inventory of available datasets to support the research, and demonstrated an integrated application of user response models with a simulation-based DTA platform for the New York region Best Practice Model network.
Emissions and Energy Costs in Marginal Cost Pricing for Roadways
Researcher(s): Hani Mahmassani
Year: 2009
Funded as part of Northwestern’s Initiative on Sustainability and Energy, the objective of this study is to develop a methodology for analyzing and setting user prices (under marginal cost pricing) to better reflect vehicle greenhouse gas emissions and energy consumption.
Congestion results in increased greenhouse emissions and wasted fuel. Pricing is considered along with other operational strategies, such as signal timing, lane use controls and real-time traffic management, as a mechanism to influence user behavior and the resulting flow patterns towards less congested and more energy and environmentally sustainable states.
Outcomes of this ongoing study are to:
- extend congestion pricing principles to account explicitly for emissions and fuel consumption costs, in addition to travel time costs;
- provide a reliable basis and network-level methodology for analyzing pricing mechanisms for reducing CO2 emissions and fuel consumption.
Improved Framework and Tools for Highway Pricing
Researcher(s): Hani Mahmassani
Year: 2009
Dr. Mahmassani served as PI for Northwestern University Transportation Center’s participation through a subcontract to PB Americas, Inc., contracted by the National Academy of Science to implement a research synthesis on “Improved Framework and Tools for Highway Pricing”.
The current state of U.S. practice for highway pricing decisions is characterized by the following tendencies:
- Incorporation of road pricing in the general framework of regional travel demand models and planning processes by state DOTs and MPOs. These include 4‐step models based on static assignment procedures as well as recent practical implementations of activity‐based models and dynamic traffic assignment.
- Financial analyses by consultants hired for bond‐rating agencies, resulting in revenue and demand projections often based on simplified modeling tools, such as corridor‐level traffic assignment using aggregate demand elasticities.
Public agencies seek to improve the overall performance of regional transportation networks, so their interest in toll road profitability may be bounded by opportunities for re-investing net revenues in further system enhancements and/or pursuing other welfare-enhancing policies for their community of travelers, including both transit and highways users. In order to adequately reflect the number of potential behavioral shifts (for example, destination and time-of-day choice) and policy strategies that may apply under significant implementations of roadway pricing (for example, variable and/or occupancy-based tolls), travel models should be comprehensive. They should reflect regional networks and modes, times of day and traveler types. To permit robust evaluation of welfare impacts, they should be behaviorally founded. The ultimate purpose of these models is a description of travel behavior and demand response.
Understanding, planning, and managing road pricing as part of a regional system’s operation is a critical objective of this study. NUTC’s role focused on the development of network-based methodologies to capture the short and long term term responses of users to different pricing schemes, as part of decision-support capabilities for public and private entities.
Asymptotic Properties of Bridge Estimators in Sparse High-dimensional Regression Models
Researcher(s): Joel Horowitz, J. Huang, S. Ma
Year: 2008
We study the asymptotic properties of bridge estimators in sparse, highdimensional, linear regression models when the number of covariates may increase to infinity with the sample size. We are particularly interested in the use of bridge estimators to distinguish between covariates whose coefficients are zero and covariates whose coefficients are nonzero. We show that under appropriate conditions, bridge estimators correctly select covariates with nonzero coefficients with probability converging to one and that the estimators of nonzero coefficients have the same asymptotic distribution that they would have if the zero coefficients were known in advance. Thus, bridge estimators have an oracle property in the sense of Fan and Li [J. Amer. Statist. Assoc. 96 (2001) 1348–1360] and Fan and Peng [Ann. Statist. 32 (2004) 928–961]. In general, the oracle property holds only if the number of covariates is smaller than the sample size. However, under a partial orthogonality condition in which the covariates of the zero coefficients are uncorrelated or weakly correlated with the covariates of nonzero coefficients, we show that marginal bridge estimators can correctly distinguish between covariates with nonzero and zero coefficients with probability converging to one even when the number of covariates is greater than the sample size.
The Dynamics of Fare and Frequency Choice in Urban Transit
Nonparametric Instrumental Variables Estimation of a Quantile Regression Model
Researcher(s): Joel Horowitz, S. Lee
Year: 2007
We consider nonparametric estimation of a regression function that is identified by requiring a specified quantile of the regression "error" conditional on an instrumental variable to be zero. The resulting estimating equation is a nonlinear integral equation of the first kind, which generates an ill-posed-inverse problem. The integral operator and distribution of the instrumental variable are unknown and must be estimated nonparametrically. We show that the estimator is mean-square consistent, derive its rate of convergence in probability, and give conditions under which this rate is optimal in a minimax sense. The results of Monte Carlo experiments show that the estimator behaves well in finite samples.
A Nonparametric Test of Exogeneity
Researcher(s): Joel Horowitz, Richard Blundell
Year: 2007
This paper presents a test for exogeneity of explanatory variables that minimizes the need for auxiliary assumptions that are not required by the definition of exogeneity. It concerns inference about a non-parametric function "g" that is identified by a conditional moment restriction involving instrumental variables (IV). A test of the hypothesis that "g" is the mean of a random variable "Y" conditional on a covariate "X" is developed that is not subject to the ill-posed inverse problem of non-parametric IV estimation. The test is consistent whenever "g" differs from "E"("Y"|"X") on a set of non-zero probability. The usefulness of this new exogeneity test is displayed through Monte Carlo experiments and an application to estimation of non-parametric consumer expansion paths.
The Scaling Law of Human Travel - A Message from George
Researcher(s): Dirk Brockmann, Lars Hufnagel
Year: 2006
The dispersal of individuals of a species is the key driving force of various spatiotemporal phenomena which occur on geographical scales. It can synchronize populations of interacting species, stabilize them, and diversify gene pools.1–3 The geographic spread of human infectious diseases such as influenza, measles and the recent severe acute respiratory syndrome (SARS) is essentially promoted by human travel which occurs on many length scales and is sustained by a variety of means of transportation4–8. In the light of increasing international trade, intensified human traffic, and an imminent influenza A pandemic the knowledge of dynamical and statistical properties of human dispersal is of fundamental importance and acute.7,9,10 A quantitative statistical theory for human travel and concomitant reliable forecasts would substantially improve and extend existing prevention strategies. Despite its crucial role, quantitative assessment of human dispersal remains elusive and the opinion that humans disperse diffusively still prevails in many models.11 In this chapter we will report on a recently developed technique which permits a solid and quantitative assessment of human dispersal on geographical scales.12 The key idea is to infer the statistical properties of human travel by analysing the geographic circulation of individual bank notes for which comprehensive datasets are collected at online bill-tracking websites. The analysis shows that the distribution of traveling distances decays as a power law, indicating that the movement of bank notes is reminiscent of superdiffusive, scale free random walks known as L´evy flights.13 Secondly, the probability of remaining in a small, spatially confined region for a time T is dominated by heavy tails which attenuate superdiffusive dispersal. We will show that the dispersal of bank notes can be described on many spatiotemporal scales by a two parameter continuous time random walk (CTRW) model to a surprising accuracy. We will provide a brief introduction to continuous time random walk theory14 and will show that human dispersal is an ambivalent, effectively superdiffusive process.
Closed Form Discrete Choice Models
Researcher(s): Frank S. Koppelman
Year: 2006
Abstract: Random utility maximization discrete choice models are widely used in transportation and other fields to represent the choice of one among a set of mutually exclusive alternatives. The decision maker, in each case, is assumed to choose the alternative with the highest utility to him/her. The utility to the decision maker of each alternative is not completely known by the modeler; thus, the modeler represents the utility by a deterministic portion which is a function of the attributes of the alternative and the characteristics of the decision-maker and an additive random component which represents unknown and/or unobservable components of the decision maker's utility function.
Early development of choice models was based on the assumption that the error terms were multivariate normal or independently and identically Type I extreme value (gumbel) distributed (Johnson and Kotz, 1970). The multivariate normal assumption leads to the multinomial probit (MNP) model (Daganzo, 1979); the independent and identical gumbel assumption leads to the multinomial logit (MNL) model (McFadden, 1973). The probit model allows complete flexibility in the variance-covariance structure of the error terms but it's use requires numerical integration of a multi-dimensional normal distribution. The multinomial logit probabilities can be evaluated directly but the assumption that the error terms are independently and identically distributed across alternatives and cases (individuals, households or choice repetitions) places important limitations on the competitive relationships among the alternatives. Developments in the structure of discrete choice models have been directed at either reducing the computational burden associated with the multinomial probit model (McFadden, 1989; Hajivassiliou and McFadden, 1990; Börsch-Supan and Hajivassiliou, 1992; Keane, 1994) or increasing the flexibility of extreme value models.
Two approaches have been taken to enhance the flexibility of the MNL model. One approach, the development of open form discrete choice models is discussed by Bhat in another chapter of this handbook. This chapter describes the development of closed form models which relax the assumption of independent and identically distributed random error terms in the multinomial logit model to provide a more realistic representation of choice probabilities.
Comparisons of Urban Travel Forecasts Prepared with the Sequential Procedure and a Combined Model
Researcher(s): Justin D. Siegel, Joaquın De Cea, Jose Enrique Fernandez, Renan E. Rodriguez, David Boyce
Year: 2006
Detailed analyses and comparisons of urban travel forecasts prepared by applying the state-of-practice sequential procedure and the solution of a combined network equilibrium model are presented. The sequential procedure for solving the trip distribution, mode choice and assignment problems with feedback is the current practice in most transportation planning agencies, although its important limitations are well known. The solution of a combined model, in contrast, results from a single mathematical formulation, which ensures a well converged and consistent result. Using a real network, several methods for solving the sequential procedure with feedback are compared to the solution of the combined model ESTRAUS. The results of these methods are shown to have various levels of instability. The paper concludes with a call for a new paradigm of travel forecasting practice based on an internally consistent model formulation that can be solved to a level of precision suitable for comparing alternative scenarios.
Modeling Household Activity-Travel Interactions as Parallel Constrained Choices
Researcher(s): J.P. Gliebe, F.S. Koppelmann
Year: 2005
The daily activity-travel patterns of individuals often include interactions with other household members, which we observe in the form of joint activity participation and shared rides. Explicit representation of joint activity patterns is a widespread deficiency in extant travel forecasting models and remains a relatively under-developed area of travel behavior research. In this paper, we identify several spatially defined tour patterns found in weekday household survey data that describe this form of inter-agent decision making. Using pairs of household decision makers as our subjects, we develop a structural discrete choice model that predicts the separate, parallel choices of full-day tour patterns by both persons, subject to the higher level constraint imposed by their joint selection of one of several spatial interaction patterns, one of which may be no interaction. We apply this model to the household survey data, drawing inferences from the household and person attributes that prove to be significant predictors of pattern choices, such as commitment to work schedules, auto availability, commuting distance and the presence of children in the household. Parameterization of an importance function in the models shows that in making joint activity-travel decisions significantly greater emphasis is placed on the individual utilities of workers relative to non-workers and on the utilities of women in households with very young children. The model and methods are prototypes for tour-based travel forecasting systems to represent the complex interaction between household members in an integrated model structure.
Comparative Analysis of Sequential and Simultaneous Choice Structures for Modeling Intra-Household Interactions
Researcher(s): P. Vovsha, J. Gliebe, E. Petersen, F.S. Koppelman
Year: 2005
Intra-household interactions constitute an important aspect in modeling activity and travel-related decisions. Recognition of this importance has recently produced a growing body of research on various aspects of modeling intra-household interactions and group decision making mechanisms as well as first attempts to incorporate intra-household interactions in regional travel demand models. This paper presents an attempt to build a general framework for incorporation of intra-household interactions in the regional travel demand model. The approach distinguishes between three principal levels of intra-household interactions: 1) Coordinated principal daily pattern types, 2) Episodic joint activity and travel, 3) Intra-household allocation of maintenance activities. The adopted models are discrete choice constructs of the Generalized Extreme Value class. These models together create an analytical framework for integrative modeling of the daily activity and travel of multiple household members, taking into account their interactions.
Contractibility and Asset Ownership: On-board Computers and Governance in U.S. Trucking
Researcher(s): George P. Baker, Thomas N. Hubbard
Year: 2004
We investigate how contractual incompleteness affects asset ownership in trucking by examining cross-sectional patterns in truck ownership and how truck ownership has changed with the diffusion of on-board computers (OBCs). We find that driver ownership of trucks is greater for long than short hauls, and when hauls require equipment for which demands are unidirectional rather than bidirectional. We then find that driver ownership decreases with OBC adoption, particularly for longer hauls. These results are consistent with the hypothesis that truck ownership reflects trade-offs between driving incentives and bargaining costs, and indicate that improvements in the contracting environment have led to less independent contracting and larger firms.
Information, Decisions, and Productivity: On-Board Computers and Capacity Utilization in Trucking
Researcher(s): Thomas N. Hubbard
Year: 2003
Productivity reflects not only how efficiently inputs are transformed into outputs, but also how well information is applied to resource allocation decisions. This paper examines how information technology has affected capacity utilization in the trucking industry. Estimates for 1997 indicate that advanced on-board computers (OBCs) have increased capacity utilization among adopting trucks by 13 percent. These increases are higher than for 1992, suggesting lags in the returns to adoption, and are highly skewed across hauls. The 1997 estimates imply that OBCs have enabled 3-percent higher capacity utilization in the industry, which translates to billions of dollars of annual benefits
Make versus Buy in Trucking: Asset Ownership, Job Design, and Information
Researcher(s): George P. Baker, Thomas N. Hubbard
Year: 2003
Explaining patterns of asset ownership is a central goal of both organizational economics and industrial organization. We develop a model of asset ownership in trucking, which we test by examining how the adoption of different classes of on-board computers (OBCs) between 1987 and 1997 influenced whether shippers use their own trucks for hauls or contract with for-hire carriers. We find that OBCs' incentive-improving features pushed hauls toward private carriage, but their re- source-allocation-improving features pushed them toward for-hire carriage. We con- clude that ownership patterns in trucking reflect the importance of both incomplete contracts and of job design and measurement issues.
The First Use of a Combined-value Auction for Transportation Services
Researcher(s): John O. Ledyard, Mark Olson, David Porter, Joseph A. Swanson, David P. Torma
Year: 2002
Combined-value auctions (CVAs) allow participants to make an offer of a single amount for a collection of items. These auctions provide value to both buyers and sellers of goods or services in a number of environments, but they have rarely been implemented, perhaps because of lack of knowledge and experience. Sears Logistics Services (SLS) is the first procurer of trucking services to use a CVA to reduce its costs. In 1993, it saved 13 percent over past procurement practices. Experimental economics played a crucial role in the development, sale, and use of the CVA.
Hub and Network Pricing in the Northwest Airlines Domestic System
Researcher(s): Robert J. Gordon, Darryl Jenkins
Year: 1999
This paper investigates the “hub premium” hypothesis that major carriers with the major share of traffic in and out of a hub exploit so-called "monopoly power." The hypothesis states that these carriers charge hub-city residents higher fares for travel originating or terminating at the hub than they charge other passengers traveling on the rest of their systems. Some have even gone so far as to claim that consumers living in hub cities live in “pockets of pain.”
By contrast, virtually everyone agrees that consumers who choose to take one stop flights enjoy the full benefits of competition. If a passenger is traveling, say, from Newark to Los Angeles or Seattle and is willing to include a stop in the itinerary, that person has a choice of flying perhaps seven or eight different airlines – including all of the major carriers. Those flying shorter distances, even from Washington to Chicago, have the choice of connecting through cities like Cleveland, Pittsburgh, Detroit, and Cincinnati, rather than going nonstop. This rich array of choices for connecting traffic guarantees a competitive fare to the passenger willing to make a connection.
The surprising result of this study is that the passenger originating or terminating his or her trip in the three major Northwest Airlines hub cities actually enjoys the same competitive fare as the connecting passenger, holding constant the effect of mileage on fares. And yet this study makes no adjustment whatsoever to the benefit to the hub-originating passenger of his or her freedom from the inconvenience or time penalty of connecting or stopping enroute.
Learning about Transport Costs
Researcher(s): Ronald R. Braeutigam
Year: 1999
This chapter examines the progress made during the past few decades in understanding transportation costs. Specifically, the chapter focuses on the costs incurred by carriers in providing railroad, motor carrier, airline, or other transportation services. Many other types of costs are not addressed here, such as congestion costs, pollution costs and other externalities, and other costs to users of transport services, such as the value of time in travel. Progress in understanding has come on three fronts. First, there have been significant advances in the theoretical understanding of costs. For example, early cost studies did not recognize the proper role of factor prices in a cost function. Researchers such as McFadden and Nerlove showed the importance in empirical work of specifying cost functions that are consistent with production theory, including not only a proper treatment of factor prices, but also variables that might contribute to a change in technology over time. Although these principles are now part of the material covered in standard graduate and even undergraduate courses in microeconomics, they helped define a renaissance in empirical studies of costs and production functions.
Second, improvements in empirical techniques have made it possible to learn more than ever about the underlying structure of technology in an industry. Early cost studies were often based on simple functional forms that embodied very strong assumptions about the nature of technology. They also were highly aggregated, effectively treating transportation firms as single product enterprises, and they often paid little attention to the quality of services provided. Empirical work in the past two decades has been advanced by the introduction of more flexible functional forms that contain as special cases many of the more specialized functional forms used by early investigators. Researchers have also improved techniques for studying the costs of multiproduct firms, allowing at least some degree of disaggregation of products.
Finally, as regulatory reform has been implemented, researchers have asked new kinds of questions about technology. For example, in the past regulators often studied costs to determine whether a firm’s revenues would cover its costs or to measure the extent to which total costs could be divided into fixed and variable costs. Over time researchers have learned the importance of incorporating features of the transportation network into cost studies, for example, by distinguishing economies of size from economies of density. As regulatory reform became a real possibility, researchers began to ask whether is was likely to lead to an industry structure compatible with competition.
To understand the evolution of transportation cost studies, it is useful to begin with a brief discussion of the kinds of cost studies that the Interstate Commerce Commission (ICC) commonly used before regulatory reform. Because issues of rail rate making were important even before the turn of the century, much of the early effort to measure transport costs focused on the railroad industry. Several academic researchers succeeded in pointing out the limitations of regulatory costing procedures and inspired a generation of improved studies. From that beginning point, I follow the flow of literature through a series of improvements in the use of theory and empirical techniques.
After discussing several studies that have made important methodological contributions to the literature, I summarize findings from several of them about the major characteristics of selected transport modes, including economies of scale, density, size, and scope. At the outset, however, I note that this chapter is not intended to provide a comprehensive survey of transportation cost studies, an effort well beyond the scope of this paper and also one that has attempted elsewhere, including the recent excellent survey by Oum and Waters.