1 September 2015 in International Journal of Transportation Science and Technology
In order to better predict the traffic states on freeways and make management decisions, a hybrid model of support vector machine (SVM) and K-nearest neighbor (KNN) is proposed for short-term freeway exiting volume prediction. First, a historical data set is built by using the freeway toll data. The abnormal toll records, such as records that have same entry and exit station, illogical time record and abnormal travel speed, are excluded by data quality control. Based on the historical dataset, it is found that the exiting volume has periodical variation over time which provides the basis of the short-term prediction. Then, the historical data set is cross-classified into twelves groups based on the day of week and time of day. The prediction has been done for each group. Finally, the prediction is accomplished by the hybrid-model of SVM and KNN. The exiting volumes of previous time periods are used as the feature vector for KNN and SVM. Besides, a dynamic weight is adopted for the prediction of current ti...
1 March 2018 in Case Studies on Transport Policy
1 May 2018 in Journal of Computing in Civil Engineering
Neural network (NN) models have been widely used in travel demand modeling in recently years. However, there are few studies about the spatial transferability of NN models. In this paper, the spatial transferability of NN models in travel demand modeling, especially in mode choice models, is analyzed. This paper first discusses the performance of naïve transfer when no data are available in an application context. Then, a NN model adaptation method is proposed using the classification adjustment weight vector when limited local data are available. Using the 2007/2008 Transportation Planning Board—Baltimore Metropolitan Council Household Travel Survey data, five NN models are built using trips within five areas in the Washington, DC, and Baltimore regions. Each of the five NN models is applied to the other four areas to study spatial transferability using both individual-level and aggregate-level performance measures. The result shows that the naïve transfer of NN models can perform very well between areas that share many similarities. It also indicates the transferability of NN models is not symmetric. The performance of the proposed adaptation method is evaluated for different sample sizes of local training data. For transfer between areas that have significant differences, the proposed NN model adaptation method can improve performance significantly, even with a small sample size, compared to naïve transfer.