conference paper

Using gradient boosting machines to predict bikesharing station states

Proceedings of the 93rd annual meeting of the transportation research board

Publication Date

January 1, 2014

Author(s)

Robert Regue, Will Recker

Abstract

Bikesharing is a sustainable and environmentally friendly transportation mode that offers bikes â??on-demandâ?? to help improve daily urban mobility. However, their operation suffers from the effects of the fluctuating demand in space and time that leads to severe system inefficienciesâ??having either empty or full stations for long periods of time. To resolve the inefficiencies, bikesharing operators are forced to reposition bikes dynamically to avoid the system from collapsing. The knowledge of future demand patterns can aid in repositioning tasks, reducing relocation costs and increasing system performance. In this paper the authors use data from the Hubway Bikesharing systemâ??to which they add weather characteristicsâ??and implement Gradient Boosting Machines (GBM) to make station level forecasts at 20, 40 and 60 minutes. The authors demonstrate the advantages of GBM compared to Neural Networks (NN) and Linear Regression (LR), namely: reduced data cleaning and preparation times, insensitivity towards irrelevant explanatory variables and better prediction accuracies. A total of 18 models for the 61 stations are run and errors and optimal calibration parameters are obtained. For calibration purposes a differential evolution algorithm is implemented. The system root mean squared error (RMSE) normalized by the station capacity obtained without calibrating the GBM model is lower than all other models for all time windows. When compared to the equivalent NN, it is 1.33, 8.7 and 13.27 % better for the 20, 40 and 60 minutes predictions, respectively.

Suggested Citation

Robert Regue and Will Recker (2014) “Using gradient boosting machines to predict bikesharing station states”, in Proceedings of the 93rd annual meeting of the transportation research board, p. 16p.

conference paper

The effectiveness of state and local incentives on household ownership of hybrid electric vehicles: Results from the 2009 national household travel survey

Proceedings of the TRB workshop using national household travel survey data for transportation decision making, the keck center of the national academies, washington DC, june 6, 2011

Publication Date

January 1, 2011

Author(s)

Roberto Ayala, Jean-Daniel Saphores

Suggested Citation

Roberto Ayala and Jean-Daniel Saphores (2011) “The effectiveness of state and local incentives on household ownership of hybrid electric vehicles: Results from the 2009 national household travel survey”, in Proceedings of the TRB workshop using national household travel survey data for transportation decision making, the keck center of the national academies, washington DC, june 6, 2011.

published journal article

How to Franchise Highways

Journal of Transport Economics and Policy

Publication Date

May 1, 1993

Author(s)

Gordon (Pete) Fielding

View published journal article

Abstract

The current methods of franchising highways in America have been devised in haste and can be improved upon. The project should first be defined and then guided through environmental approval. The authors suggest that this should be done by an independent state body endowed with the mission of franchising highways. The project could then be auctioned, and the winning bidder would complete the final design and then finance, construct and operate the highway.

Suggested Citation

Gordon J Fielding (1993) “How to Franchise Highways”, Journal of Transport Economics and Policy, 27(2), pp. 113–130. Available at: https://www.jstor.org/stable/20052997.

conference paper

Dynamic Tests of a Time-Space Model of Complex Travel Behavior

Travel Behavior Research, International Association of Travel Behavior

Publication Date

September 1, 1987

Author(s)

Will Recker, Thomas Golob, Michael McNally, John Leonard

Abstract

In the research presented here, an attempt is made to analyze the entire daily activity pattern simultaneously in terms of its space, time, and activity category dimensions. The focus of the research is on developing procedures for the dynamic analysis of possible changes in activity patterns that may result from or lead to changes in the characteristics of the household. The intent is to trace the daily activity patterns of groups of travelers that have undergone significant change during multiple points in time, in a manner that will shed some light on both the nature and extent of associated changes in travel behavior. To check on the reliability of the modeling process and to understand its peculiarities, complementary analyses of pattern attributes, and time-series analyses of patterns were performed on the same data set. Results from these complementary analyses are briefly outlined here, and it is hoped that these results aid in placing the present approach in the perspective of the rich legacy of previous work.

Suggested Citation

Will Recker, Thomas F. Golob, Michael G. McNally and John D. Leonard (1987) “Dynamic Tests of a Time-Space Model of Complex Travel Behavior”. Travel Behavior Research, International Association of Travel Behavior, pp. 27–40.

conference paper

Factors influencing alternative fuel adoption decisions in heavy-duty vehicle fleets in California

The 98th annual meeting of the transportation research board

Publication Date

January 1, 2019

Author(s)

Youngeun Bae, Suman Mitra, Craig Rindt, Stephen Ritchie

Abstract

Alternative fuel adoption by heavy-duty vehicle (HDV) fleets can bring substantial benefits to both current local communities and future generations by reducing air pollutants and greenhouse gas emissions. A better understanding of HDV fleet operator perspectives towards alternative fuels can serve as the basis for developing effective policy suggestions that aim to accelerate the diffusion of these innovative technologies throughout the entire vehicle market. This study attempts to identify the factors that have influenced the decision for alternative fuel adoption or non-adoption made by HDV fleet operators in California. Through 18 in-depth qualitative interviews, a total of 25 adoption and 41 non-adoption cases were investigated across various alternative fuel options. Content analysis of the interview data yielded that the primary and most common factors comprised functional suitability, environmental consciousness, availability of vehicles, and regulations. Other important but less common factors were overall costs, vehicle purchase price, fuel price, fuel infrastructure, vehicle reliability and safety, contracts with municipalities, and financial incentives. Based on the key factors identified, testable propositions were developed and seven major hypotheses are presented. The study results help gain a better grasp of the demand-side aspects of alternative fuel adoption in the HDV sector in California.

Suggested Citation

Youngeun Bae, Suman Kumar Mitra, Craig Ross Rindt and Stephen G. Ritchie (2019) “Factors influencing alternative fuel adoption decisions in heavy-duty vehicle fleets in California”, in The 98th annual meeting of the transportation research board.

published journal article

The perceived psychological benefits of third places for university students before and after COVID-19 lockdowns

Cities

Publication Date

October 1, 2024

Author(s)

Narae Lee, Doug Houston

View published journal article

Abstract

The impact of COVID-19 in the relationship between consumer-facing businesses and health and well-being remains understudied. In response, we utilized mixed methods to examine the perceived psychological benefits of urban parks and consumer-facing businesses among students from the University of California, Irvine before and after the COVID-19 lockdown. Overall, findings from photo-based close-ended questions indicated no significant disparity in perceived restorative quality, sentiment, and preference for urban third places before and after COVID-19 lockdowns. However, crowded third places exhibited a notable decline in sentiment and preference for third spaces following the COVID-19 lockdowns. Inductive content analysis of open-ended questions indicated a rise in visits to eating/drinking establishments and shops/stores post-COVID-19 lockdowns and a shift towards outdoor venues for activities. Findings support policy and planning efforts that promote healthy communities by maximizing the benefit of third places.

Suggested Citation

Narae Lee and Douglas Houston (2024) “The perceived psychological benefits of third places for university students before and after COVID-19 lockdowns”, Cities, 153, p. 105299. Available at: 10.1016/j.cities.2024.105299.

conference paper

Evaluation of relative clock stability in cellular networks

Proceedings of the 30th international technical meeting of the satellite division of the institute of navigation (ION GNSS+ 2017)

Publication Date

November 1, 2017

Author(s)

Joe Khalife, Zaher Kassas

View conference paper

Suggested Citation

Joe J. Khalife and Zaher M. Kassas (2017) “Evaluation of relative clock stability in cellular networks”, in Proceedings of the 30th international technical meeting of the satellite division of the institute of navigation (ION GNSS+ 2017). Institute of Navigation, pp. 2554–2559. Available at: 10.33012/2017.15221.

published journal article

Daily activity pattern recognition by using support vector machines with multiple classes

Transportation Research Part B: Methodological

Publication Date

December 1, 2013

Author(s)

Mahdieh Allahviranloo, Will Recker

View published journal article

Suggested Citation

Mahdieh Allahviranloo and Will Recker (2013) “Daily activity pattern recognition by using support vector machines with multiple classes”, Transportation Research Part B: Methodological, 58, pp. 16–43. Available at: 10.1016/j.trb.2013.09.008.

published journal article

Demand for clean-fuel vehicles in California: A discrete-choice stated preference pilot project

Transportation Research Part A: Policy and Practice

Publication Date

May 1, 1993

Author(s)

David Bunch, Mark Bradley, Thomas Golob, Ryuichi Kitamura, Gareth P. Occhiuzzo

View published journal article

Abstract

A study was conducted to determine how demand for clean-fuel vehicles and their fuels is likely to vary as a function of attributes that distinguish these vehicles from conventional gasoline vehicles. For the purposes of the study, clean-fuel vehicles are defined to encompass both electric vehicles and unspecified (methanol, ethanol, compressed natural gas or propane) liquid and gaseous fuel vehicles, in both dedicated or multiple-fuel versions. The attributes include vehicle purchase price, fuel operating cost, vehicle range between refueling, availability of fuel, dedicated versus multiple-fuel capability and the level of reduction in emissions (compared to current vehicles). In a mail-back stated preference survey, approximately 700 respondents in the California South Coast Air Basin gave their choices among sets of hypothetical future vehicles, as well as their choices between alternative fuel versus gasoline for hypothetical multiple-fuel vehicles. Estimates of attribute importance and segment differences are made using discrete-choice nested multinomial logit models for vehicle choice and binomial logit models for fuel choice. These estimates can be used to modify present vehicle-type choice and utilization models to accommodate clean-fuel vehicles; they can also be used to evaluate scenarios for alternative clean-fuel vehicle and fuel supply configurations. Results indicate that range between refueling is an important attribute, particularly if range for an alternative fuel is substantially less than that for gasoline. For fuel choice, the most important attributes are range and fuel cost, but the predicted probability of choosing alternative fuel is also affected by emissions levels, which can compensate for differences in fuel prices.

Suggested Citation

David S. Bunch, Mark Bradley, Thomas F. Golob, Ryuichi Kitamura and Gareth P. Occhiuzzo (1993) “Demand for clean-fuel vehicles in California: A discrete-choice stated preference pilot project”, Transportation Research Part A: Policy and Practice, 27(3), pp. 237–253. Available at: 10.1016/0965-8564(93)90062-p.

working paper

Electronic Integration in the Air Cargo Industry: An Information Processing Model of On-Time Performance

Publication Date

January 1, 2003

Author(s)

Amelia Regan

Areas of Expertise

Freight, Logistics, & Supply Chain

View working paper

Abstract

This study examines electronic integration in logistics supply chains using the non-integrated US-international air cargo industry as a case study. We ask what impact electronic integration has on interorganizational task performance, and hypothesize about factors limiting the effectiveness of electronic integration. Surprisingly, our study does not find evidence of direct impacts of electronic integration on performance though it does find evidence of indirect impacts of information systems use and performance. The findings suggest that the use of electronic integration as a strategy to improve operational performance across firms is limited by the nature of the interorganizational task, environmental dynamism, and the power relationship between firms in the supply chain.