The integration of electric bicycles (e-bikes) with bikesharing could increase the utility of bikesharing through a reduction of some barriers to bicycling and an increase in the number of prospective users. North America's first e-bikesharing system (cycleUshare) at the University of Tennessee, Knoxville, offers a new, sustainable transportation option for students, faculty, and staff. The cycleUshare system with two stations was launched as a small pilot project to study the technology and its users' experiences. This paper presents an overview of the cycleUshare system and reports experiences from the first year of operation. With 93 enrolled users, cycleUshare provided a unique opportunity to study not only the system's use but how individual users made trips with both regular bicycles and e-bikes and the factors that influenced those trips. The study reported here found that only 22% of users accounted for 81% of the trips. Factors of speed and convenience played major roles in participants' decisions to use the system, and speed and comfort were the most influential factors in the selection of an e-bike rather than a regular bicycle. Most of the reported trips were class related, although e-bikes were used for a wide variety of trip purposes. Walking was the mode most displaced by the system; this result indicated that e-bikesharing expanded user mobility. User perceptions about bicycle types were explored also. This model of e-bikeshare was effective in its capability to attract users to both regular bicycles and e-bikes and to expand user mobility.

This study examines the factors that influence the use of carsharing systems in Beijing, as well as the potential for carsharing systems that integrate electric vehicles. Investigated variables include weather, air quality, price, vehicle attributes and "status" indicators. Additionally, we explore how the impacts of these factors differ when carsharing is utilized for one-way trips as compared with round-trips. The study relies on a pen-and-paper survey (1,010 completed survey forms with 2,023 reported trips) which uses a stated-preference pivoting design to build hypothetical choice sets around actual trips. The data are used to estimate binomial logit regression models for one-way and round-trip carsharing. Results of the analysis indicate that age, car ownership, shelter mode, the original cost for taxi users, perceived parking availability, cold weather, and relative cost differences are significant for one-way carsharing. For round-trip carsharing, significant factors include car ownership, income, gender, environmental concern, and relative cost differences. The most statistically significant factors to attract carsharing customers are the cost gap (defined as cost of original mode-cost of carshare) for both one-way and round-trip carsharing services, and car ownership, which has a positive significant effect for one-way trips and a negative significant effect for round-trips. This paper contributes to the literature by further examining the determinants that affect the use of carsharing, distinguishing between one-way trips and round-trips, and developing models that can be applied in urban environments like Beijing.

No abstract

There have been numerous studies on traffic accidents and their severity, particularly in relation to weather conditions and road geometry. In these studies, traditional statistical methods have been employed, such as linear regression, logistic regression, and negative binomial regression modeling, which are the most common linear and non-linear regression analysis methods. In this research, machine learning architecture was applied to this problem using the random forest, artificial neural network, and decision tree techniques to ascertain the strengths and weaknesses of these methods. Three data sets were used: road geometry data, precipitation data, and traffic accident data over nine years corresponding to the Naebu Expressway, which is located in Seoul, Korea. For the model evaluation, three measures were employed: the out-of-bag estimate of error rate (OOB), mean square error (MSE), and root mean square error (RMSE). The low mean OOB, MSE, and RMSE observed in the results obtained using the proposed random forest model demonstrate its accuracy.

No abstract