Nos publications et articles scientifiques !
0
Articles scientifiques
0
Mémoires et thèses
0
Autres documents
Articles scientifiques
Abstract: : Long-distance (LD) or intercity travel is getting less attention by researchers than usual daily trips. There is no specific definition for this kind of trip at the national, provincial, and inter-regional levels. At the same time, it has a high contribution to transportation in terms of distance travelled. This paper presents a model comparison method for (LD) trip generation model for LD trips performed by residents of Canada. The terms of « long-distance » trip based on Travel Survey for Residents in Canada (TSRC) survey is considered as non-frequent overnight and day trip. This study compared several machine learning methods for the trip generation model. Since LD trip is relatively rare, the data set of TSRC is considered imbalanced data, three different techniques on the data preparation level as part of rare event modelling (over, under, and synthetically oversampling) employed to handle the issue of imbalanced data. TSRC data from 2012 to 2017 was used for model estimation. Among the random forest, CART, CTree, and logit models, it was found that the random forest has the best performance in prediction, and decision tree models have the best overall accuracy. Also, Income level and educational level play an essential role in the occurrence of an intercity trip. The paper highlights the importance of improvement in intercity travel survey methods and other data collection methods.
Abstract: : Evolutionary algorithms have been used extensively over the past 2 decades to provide solutions to the Transit Network Design Problem and the Transit Network and Frequencies Setting Problem. Genetic algorithms in particular have been used to solve the multi-objective problem of minimizing transit users’ and operational costs. By finding better routes geometry and frequencies, evolutionary algorithms proposed more efficient networks in a timely manner. However, to the knowledge of the authors, no experimentation included precise and complete pedestrian network data for access, egress and transfer routing. Moreover, the accuracy and representativeness of the transit demand data (Origin Destination matrices) are usually generated from fictitious data or survey data with very low coverage and/or representativity. In this paper, experiments conducted with three medium-sized cities in Quebec demonstrate that performing genetic algorithm optimizations using precise local road network data and representative public transit demand data can generate plausible scenarios that are between 10 and 20% more efficient than existing networks, using the same parameters and similar fleet sizes.
Abstract: : Accurate forecasting of passenger flow (i.e., ridership) is critical to the operation of urban metro systems. Previous studies mainly model passenger flow as time series by aggregating individual trips and then perform forecasting based on the values in the past several steps. However, this approach essentially overlooks the fact that passenger flow consists of trips from each individual traveler. For example, a traveler’s work trip in the morning can help predict his/her home trip in the evening, while this causal structure cannot be explicitly encoded in standard time series models. In this paper, we propose a new forecasting framework for boarding flow by incorporating the generative mechanism into standard time series models and leveraging the strong regularity rooted in travel behavior. In doing so, we introduce returning flow from previous alighting trips as a new covariate, which captures the causal structure and long-range dependencies in passenger flow data based on travel behavior. We develop the return probability parallelogram (RPP) to summarize the causal relationships and estimate the return flow. The proposed framework is evaluated using real-world passenger flow data, and the results confirm that the returning flow—a single covariate—can substantially and consistently improve various forecasting tasks, including one-step ahead forecasting, multi-step ahead forecasting, and forecasting under special events. And the proposed method is more effective for business-type stations with most passengers come and return within the same day. This study can be extended to other modes of transport, and it also sheds new light on general demand time series forecasting problems, in which causal structure and long-range dependencies are generated by the user behavior.
Abstract: : Inferring trip destination in smart card data with only tap-in control is an important application. Most existing methods estimate trip destinations based on the continuity of trip chains, while the destinations of isolated/unlinked trips cannot be properly handled. We address this problem with a probabilistic topic model. A three-dimensional latent dirichlet allocation model is developed to extract latent topics of departure time, origin, and destination among the population; each passenger’s travel behavior is characterized by a latent topic distribution defined on a three-dimensional simplex. Given the origin station and departure time, the most likely destination can be obtained by statistical inference. Furthermore, we propose to represent stations by their rank of visiting frequency, which transforms divergent spatial patterns into similar behavioral regularities. The proposed destination estimation framework is tested on Guangzhou Metro smart card data, in which the ground-truth is available. Compared with benchmark models, the topic model not only shows increased accuracy but also captures essential latent patterns in passengers’ travel behavior. The proposed topic model can be used to infer the destination of unlinked trips, analyze travel patterns, and passenger clustering.
Abstract: : Despite the desired transition toward sustainable and multimodal mobility, few tools have been developed either to quantify mode use diversity or to assess the effects of transportation system enhancements on multimodal travel behaviors. This paper attempts to fill this gap by proposing a methodology to appraise the causal impact of transport supply improvement on the evolution of multimodality levels between 2013 and 2018 in Montreal (Quebec, Canada). First, the participants of two household travel surveys were clustered into types of people (PeTys) to overcome the cross-sectional nature of the data. This allowed changes in travel behavior per type over a five-year period to be evaluated. A variant of the Dalton index was then applied on a series of aggregated (weighted) intensities of use of several modes to measure multimodality. Various sensitivity analyses were carried out to determine the parameters of this indicator (sensitivity to the least used modes, intensity metric, and mode independency). Finally, a difference-in-differences causal inference approach was explored to model the influence of the improvement of three alternative transport services (transit, bikesharing, and station-based carsharing) on the evolution of modal variability by type of people. The results revealed that, after controlling for different socio-demographic and spatial attributes, increasing transport supply had a significant and positive impact on multimodality. This outcome is therefore good news for the mobility of the future as alternative modes of transport emerge. Mobility as a Service (MaaS) and multimodality (or intermodality) are increasingly fashionable concepts in urban transport research. They are generally presented as essential in the move toward more sustainable mobility, even more now that alternatives to the exclusive use of a private car are emerging. Consequently, many projects focus on the development of technologies (such as mobile applications) to integrate several transportation modes. However, there is a lack of knowledge about multimodal behaviors among travelers, their associated attributes, and motivations. Few tools have been proposed to quantify and explain mode use variability. It, therefore, remains difficult for practitioners to measure the impacts of strategies implemented to promote a more diversified mobility. This paper aims at assessing the influence transport supply may have on the multimodality level of travel behaviors over a five-year period. This required first defining an indicator to summarize the intensities of use of several modes of transport. This indicator was calculated by type of people (aggregated by socio-demographic and spatial attributes) using the last two (2013 and 2018) Montreal household surveys. The evolution by type of people between the two surveys was then cross-analyzed with changes in transport supply. To this end, a difference-in-differences approach that allows causal inference was applied. As a result, this paper provides a means to guide interventions to increase multimodality. This paper is organized as follows. First, some former studies into analyzing and quantifying multimodal mobility, as well as factors affecting multimodal travel behavior, are reviewed. The household surveys used in this paper are then described, and some methodological elements are exposed. The results are reported in three parts: the typology of people is presented; a multimodality indicator is then applied by type; and the causal effect of transport supply on the evolution of multimodality between 2013 and 2018 is modeled. Finally, the paper is concluded and the discussion feeds into the programming of future work.