No abstract

A growing interest in the behaviour of travelers to university campuses has recently emerged whether by university administrators or transport officials. Understanding the modal choice determinants of university travellers increases the opportunity for finding appropriate policies and solutions to reduce traffic congestion and parking needs as well as to encourage active transportation hence achieving more sustainable mobility. This research study investigates the differences in mode choice habits among the various groups of travellers to Sharjah University City (SUC) in the United Arab Emirates (UAE), including students, staff, faculty, and university visitors. A revealed preference survey was distributed randomly throughout SUC. Using information collected from this survey, multinomial discrete logit choice models were developed to evaluate the SUC travellers’ mode-choice likelihood for the following modes: car, private bus, public bus, taxi, and active transport (walking and biking). It was found that travel time, travel distance, trip makers’ characteristics (gender, citizenship, car ownership, car sharing, and the number of cars per household), and other contributing factors ( Weathers conditions, Infrastructure adequacy, and bus services quality) are the main factors that affect significantly the mode choice at SUC. Further, a sensitivity analysis was conducted to study how the considered factors influence the mode choice. The developed model can be used in future studies to predict travel demand at SUC in response to new policies and solutions set by university administrators or transport officials.

No abstract

Due to the ambiguity between risk attitudes, this study aims at ranking the different risk attitudes considering the factors that affect the behaviour of the decision-makers. Both the technique for order preference by similarity to ideal solution (TOPSIS) and analytical hierarchy process (AHP) are employed to address the characteristics of risk attitudes aiming to highlight the criteria significance and finally to rank the most impactful risk attitude. It was found that regret aversion and risk aversion attitudes have higher impact in real life decision-making problems. In contrast, the maximin and maximax risk attitudes have the lowest importance. Risk seeking and regret aversion attitudes demonstrated the highest importance using TOPSIS of equal-weights while the importance of loss aversion and regret aversion have the highest for the AHP-TOPSIS approach. The results of this study can be beneficial for decision-makers who encounter a variety of risk attitudes in their decision problems.

This research proposes a hybrid approach for predicting incident duration that integrates the salient features of both factorial design of experiments (DOE) and machine learning (ML). This study compares DOE with another widely used technique, forward sequential feature selection (FSFS). Moreover, to confirm the effectiveness and robustness of the proposed approach, multiple ML techniques are employed, including linear regression, decision trees, support vector machines, ensemble trees, Gaussian process regression, and artificial neural networks. The study results are validated using data from the Houston TranStar incidents archive with over 90,000 records. The accuracy of the developed predictive models is compared based on multiple techniques (i.e., no feature selection–ML, FSFS–ML, and DOE–ML). The results revealed that the significant factors affecting incident duration identified by both DOE and FSFS include the type of vehicles involved, type of lanes affected, number of vehicles involved, number of emergency responses dispatched, incident severity level, and day of the week. The comparative results of the different feature selection and modeling approaches revealed that the hybrid DOE–ML approach outperformed the other tested analysis approaches. The best‐performing model under the DOE–ML approach was the SVM with cubic kernel model. It reduced the modeling time by 83.8% while increasing the prediction error by merely 0.02%, which is not significant. Therefore, the prediction accuracy could be slightly downgraded in return for a substantial reduction in the number of variables utilized, resulting in substantial savings in the modeling time and required dataset.