Travel time prediction of urban public transportation based on detection of single routes

Improving travel time prediction for public transit effectively enhances service reliability, optimizes travel structure, and alleviates traffic problems. Its greater time-variance and uncertainty make predictions for short travel times (≤35min) more subject to be influenced by random factors. It requires higher precision and is more complicated than long-term predictions. Effectively extracting and mining real-time, accurate, reliable, and low-cost multi-source data such as GPS, AFC, and IC can provide data support for travel time prediction. Kalman filter model has high accuracy in one-step prediction and can be used to calculate a large amount of data. This paper adopts the Kalman filter as a travel time prediction model for a single bus based on single-line detection: including the travel time prediction model of route (RTM) and the stop dwell time prediction model (DTM); the evaluation criteria and indexes of the models are given. The error analysis of the prediction results is carried out based on AVL data by case study. Results show that under the precondition of multi-source data, the public transportation prediction model can meet the accuracy requirement for travel time prediction and the prediction effect of the whole route is superior to that of the route segment between stops.

Travel Time Prediction and Explanation with Spatio-Temporal Features: A Comparative Study

Travel time information is used as input or auxiliary data for tasks such as dynamic navigation, infrastructure planning, congestion control, and accident detection. Various data-driven Travel Time Prediction (TTP) methods have been proposed in recent years. One of the most challenging tasks in TTP is developing and selecting the most appropriate prediction algorithm. The existing studies that empirically compare different TTP models only use a few models with specific features. Moreover, there is a lack of research on explaining TTPs made by black-box models. Such explanations can help to tune and apply TTP methods successfully. To fill these gaps in the current TTP literature, using three data sets, we compare three types of TTP methods (ensemble tree-based learning, deep neural networks, and hybrid models) and ten different prediction algorithms overall. Furthermore, we apply XAI (Explainable Artificial Intelligence) methods (SHAP and LIME) to understand and interpret models’ predictions. The prediction accuracy and reliability for all models are evaluated and compared. We observed that the ensemble learning methods, i.e., XGBoost and LightGBM, are the best performing models over the three data sets, and XAI methods can adequately explain how various spatial and temporal features influence travel time.

Using artificial intelligence methods for shear travel time prediction: A case study of Facha member, Sirte basin, Libya

Shear wave travel time logs are major acoustic logs used for direct estimation of the mechanical properties of rocks. They are also important for prediction of critical drawdown pressure of the reservoir. However, core samples are sometimes not available for direct laboratory measurements, and the time-consuming dipole shear imager tool is generally not used. Hence, there is a need for simple indirect techniques that can be used reliably. In this study, cross-plots between the available measured shear travel time and compressional travel time from three oil wells were used, and three artificial intelligence tools (fuzzy logic, multiple linear regression and neural networks) were applied to predict the shear travel time of Facha member (Gir Formation, Lower Eocene) in Sirte Basin, Libya. The predicted times were compared to those obtained by the equation of Brocher. The basic wireline data (gamma ray, neutron porosity, bulk density and compression travel time) of five oil wells were used. Based on principle component analysis, two wireline data sets were chosen to build intelligent models for the prediction of shear travel time. Limestone, dolomite, dolomitic limestone and anhydrite are the main lithofacies in the Facha member, with an average thickness of about 66 m. The simple equation gave 87% goodness of fit, which is considered comparable to the measured shear travel time logs. The Brocher equation yielded adequate results, of which the most accurate was for the Facha member in the eastern part of the Sirte basin. On the other hand, the three intelligent tools’ predictions of shear travel time conformed with the measured log, except in the eastern area of the basin.

Modeling of BRT System Travel Time Prediction Using AVL Data and ANN Approach

Improving the quality of public transportation systems and encouraging passengers to use them are effective solutions for reducing transportation problems in metropolitan. Prediction of travel time and providing information to passengers are significant factors in this process. In this research not only the travel time components in Bus Rapid Transit (BRT) system were investigated but also an Artificial Neural Network (ANN) model and a regression model for travel time prediction were presented. To enhance this aim, data was collected by AVL data and field observation and after investigating the primary independent variables, the significant ones were determined using statistical analysis, then ANN development was done. Moreover, linear regression method was used for this purpose. The results prove that although both models have high level of prediction accuracy, ANN model outperform the regression model and the accuracy for the route sections with no signalized intersections is higher than the others.

A review of travel and arrival-time prediction methods on road networks: classification, challenges and opportunities

Transportation plays a key role in today’s economy. Hence, intelligent transportation systems have attracted a great deal of attention among research communities. There are a few review papers in this area. Most of them focus only on travel time prediction. Furthermore, these papers do not include recent research. To address these shortcomings, this study aims to examine the research on the arrival and travel time prediction on road-based on recently published articles. More specifically, this paper aims to (i) offer an extensive literature review of the field, provide a complete taxonomy of the existing methods, identify key challenges and limitations associated with the techniques; (ii) present various evaluation metrics, influence factors, exploited dataset as well as describe essential concepts based on a detailed analysis of the recent literature sources; (iii) provide significant information to researchers and transportation applications developer. As a result of a rigorous selection process and a comprehensive analysis, the findings provide a holistic picture of open issues and several important observations that can be considered as feasible opportunities for future research directions.