Yongjun Shen

Research on Restrained Study Areas for Effective Activity-based Travel Demand Forecasting

Recently, considerable attention has been devoted to studying the activity-based approach for transportation planning and forecasting. However, one of the practical limitations of applying most of the currently available activity-based models is their computation time. This research investigated the possibility of restraining the size of the study area to reduce the computation time when applying an activity-based model. By introducing an accuracy level of the model, the authors proposed an iterative approach to determine the minimum size of the study area required for a target territory. In the application, the authors investigated the required minimum size of the study area surrounding each of the 327 municipalities in Flanders with regard to two different transport modes: car as driver and public transport. Additionally, a validation analysis was conducted. All of the experiments were carried out by using the Forecasting Evolutionary Activity-Travel of Households and their Environmental Repercussion (FEATHERS) framework, an activity-based micro-simulation modeling framework currently implemented for the Flanders region of Belgium.

Measuring driver's relative performance over time using DEA and window analysis

Reference (Published version): Babaee, Seddigheh; Shen, Yongjun; Hermans, Elke; Brijs, Tom & Wets, Geert(2016) Measuring drivers relative performance over time using DEA and window analysis. In: Zeng, Xianyi; Lu, Jie; Kerre, Etienne E.; Martinez, Luis; Koehl, Ludovic (Ed.). Uncertainty Modelling in Knowledge Engineering and Decision Making, Proceeding of the 12th International FLINS Conferense (FLINS 2016), World Scientific Publishing,p. 885-890

BOOTSTRAPPING DEA SCORES FOR TRAFFIC FATALITY RISK ASSESSMENT IN BRAZIL

In this paper, three risk indicators on road safety (i.e., traffic fatalities per vehicle, per traveled kilometer and per inhabitants) are combined into a composite indicator in order to assess the overall fatality risk for the 27 Brazilian states. The so-called multiple layer data envelopment analysis (DEA) model is used in this respect. Given their remarkable diversity in terms of road safety context, the states are first clustered (into four clusters) and next, a range of bootstrapped scores is generated to manifest the estimated variability in the road safety performance of each state. Bootstrapping the original DEA scores showed to be a useful strategy to assess the robustness of the states’ ranking, which is subjected to uncertainties in input information and DEA model determinism.

Assessing the Overall Driving Performance of Older Drivers

This study investigates the relative performance of older drivers at the individual level, based on specific measures of functional and driving abilities. To do so, 55 participants (ages 70 years and above) completed tests of an assessment battery of psychological and physical aspects as well as knowledge of road signs. Moreover, they took a driving simulator test in which specific driving situations (known to cause difficulties for older drivers) were included. To evaluate the overall performance of each driver, all the above information was combined by using the concept of composite indicators. The technique of data envelopment analysis (which is an optimization model for measuring the relative performance of a set of units or drivers in this study) was employed. The model output distinguishes the best performers from under-performing drivers and helps in guiding future development of training interventions tailored to each individual by specifically targeting those functions that are (mostly) impaired.

Assessing Road Safety Performance by Data Envelopment Analysis - The Case of Brazil

The intense economic growth experienced by Brazil in recent decades and its consequent explosive motorization process have generated an undesirable impact: the increasing and unbroken trend in road traffic fatalities. In order to contribute to road safety diagnosis, this study presents a research on two main available indicators in Brazil: mortality rate (represented by fatalities per inhabitants); and fatality rate (represented by two sub-indicators, i.e., fatalities per vehicle and fatalities per vehicle kilometers traveled). These indicators were aggregated into a composite indicator (CI) through a multiple layer data envelopment analysis CI model (ML DEA-CI), in which the optimum combination of indicators’ weights is searched for each decision-making unit (DMU), in this case, the 27 Brazilian states (BR-27). The composite indicator represents the road safety performance on which a ranking of states can be made. As a result, valuable differences associated to DMU’s socioeconomic attributes were outlined after calibrating the model.