Travel Time Prediction using Machine Learning and Weather Impact on Traffic Conditions (original) (raw)
Related papers
Travel time prediction using machine learning
2011
This paper investigates the application of a Machine Learning technique to predict the time that will be spent by a vehicle between any two points in an approximated area. The prediction is based on a learning process based on historical data about the movements performed by the vehicles taking into account a set of semantic variables to get estimated time
Travel Time Prediction Modelling in Mixed Traffic Conditions
INTERNATIONAL JOURNAL FOR TRAFFIC AND TRANSPORT ENGINEERING, 2018
Over the past few decades, exponential increase in vehicle ownership resulted in issues of traffic control and management. Intelligent Transportation System (ITS) is one of the solutions for the intelligent management of traffic. ITS applications like Advanced Traveler Information Systems and Advanced Traffic Management Systems need travel time as a major input. The estimation of travel time in urban network became more complicated because of the rapid change in the system and traffic. This study is done in order to assess the impact of different travel modes on travel time. Data has been collected on a stretch of 14 km length in Warangal city, India using GPS probe vehicle along with video camera. Different private modes of transportation such as 2 wheeler, passenger car and 3 wheeler have been used as test vehicles for the collection of data in different traffic flow scenarios. Artificial Neural Network and a multi linear regression model have been developed to compare the estimated travel times with the field data. Two combinations of ANN model using single hidden layer, different numbers of neurons and epochs have been compared. The travel time of different modes has been compared and the effect of vehicle composition on travel time has been analyzed. The ANN model perform better than the regression model.
Short-term Travel-time Prediction on Highway: A Review of the Data-driven Approach
Near future travel-time information is one of the most critical factors that travellers consider before making trip decisions. In efforts to provide more reliable future travel-time estimations, transportation engineers have examined various techniques developed in the last three decades. However, there have not been sufficiently systematic and through reviews so far. In order to effectively support various transportation strategies and applications including Intelligent Transportation Systems (ITS), it is necessary to apply appropriate forecasting methods for matching circumstances in a timely manner. This paper conducts a comprehensive review study focusing on literatures, including modern techniques proposed recently, related to travel time and traffic condition predictions that are based on ‘data-driven’ approaches. Based on the underlying mechanisms and theoretical principles, different approaches are categorized as parametric (linear regression and time series) and non-parametric approaches (artificial intelligence and pattern searching). Then, the approaches are analysed for their strengths, potential weaknesses, and performances from five main perspectives that are prediction range, accuracy, efficiency, applicability, and robustness.
Univariate short-term prediction of road travel times
Proceedings. 2005 IEEE Intelligent Transportation Systems, 2005., 2005
This paper presents an experimental comparison of several statistical machine learning methods for short-term prediction of travel times on road segments. The comparison includes linear regression, neural networks, regression trees, k-nearest neighbors, and locally-weighted regression, tested on the same historical data. In spite of the expected superiority of non-linear methods over linear regression, the only non-linear method that could consistently
Sustainability
Precise travel time prediction allows travelers and system controllers to be aware of the future conditions on roadways and helps in pre-trip planning and traffic control strategy formulation to lessen the travel time and mitigate traffic congestion problems. This research investigates the possibility of using the GPS trajectory dataset for travel time prediction in Indian traffic conditions having heterogeneous disordered traffic and improvement in prediction accuracy by shifting from the traditional historical average method to modern machine learning algorithms such as linear regressions, decision tree, random forest, and gradient boosting regression. The present study uses massive location data consisting of historical trajectories that were collected by installing GPS devices on the probe vehicles. A 3.6 km long stretch of the Delhi–Noida Direct (DND) flyway is selected as a case study to predict the travel time and compare the performance as well as the efficiency of various t...
Novel Approach to Predict the Travel time based on historical data using ML Techniques
Vol. 19 No. 2 FEBRUARY 2021 International Journal of Computer Science and Information Security (IJCSIS), 2021
Travel time plays a crucial role in the intelligent transport system in metropolitan cities. Predicting accurate Taxi trip travel time helps commuters to plan their trip better and reach the destination on time. Most of the existing techniques use supervised learning models to estimate the travel time. Performance obtained from the supervised learning models is insufficient. In this paper, we propose a novel approach that aims at predicting travel time by using both supervised and unsupervised techniques with a large historic dataset, and this novel method is compared with supervised techniques. The clustering approach of un-supervised learning along with supervised helps to enhance the performance of a predictive model. Clustering helps in segmenting the nearby location data into a similar group which helps in finding the underlying pattern within the large dataset. Then, a supervised algorithm is applied to this clustered data. Machine Learning (ML) techniques such as Random Forest Regressor (RFR), XGBoost Regressor (XGBR), which are supervised and RFR with k-means, XGBR with k-means which combines both supervised and unsuper-vised techniques are used to predict the trip time of the taxi trips. The results show that a combination of supervised and unsupervised models perform better than only supervised models. Also, the comparison shows that the RFR and RFR with k-means perform better than XGBR and XGBR with k-means respectively. RFR with k-means outper-forms other models with an accuracy of 84.6%. With better performance, RFR with k-means also reduces the error rate of the model significantly.
Comparative Assessment of Dynamic Travel Time Prediction Models in the Developing Countries Cities
INTERNATIONAL JOURNAL FOR TRAFFIC AND TRANSPORT ENGINEERING
Providing real and accurate travel time information usually, assists road users to plan their trips and choose the appropriate mode of transport. However, accurate prediction of travel time is a challenging problem, especially in developing countries where heterogeneous flow conditions exist and there are no records of information about the travel time for travelers. Most of the dynamic travel-time prediction models developed emphasize on link travel time without taking into account delay time at the intersections and waiting time at the bus stops. The objective of this study was to compare Multi-Linear Regression and Artificial Neural Network models to obtain a suitable model for developing a dynamic travel-time prediction model using waiting time at the bus stop, intersection delay time, link distance, traffic volume, link travel time, peak hours and off-peak hours as model inputs. Link travel time was modeled by a well-trained Neural Network and Kalman filtering dynamic algorithm using field survey data collected by employing public buses in Dar es Salaam city. The model was validated by using data collected in five main routes in Dar es Salaam City. The Root Mean Square Error and Mean Absolute Percent Error were used to evaluate the performance of the model by comparing it with other prediction models. Findings indicate that the integration of the Artificial Neural Network and Kalman Filter algorithm model (ANN-KF) promised to be a reasonable model for predicting dynamic travel time in Dar es Salaam city.
Travel Time Prediction on Highways
2015 IEEE International Conference on Computer and Information Technology; Ubiquitous Computing and Communications; Dependable, Autonomic and Secure Computing; Pervasive Intelligence and Computing, 2015
We describe the development of a predictive model for vehicle journey time on highways. Accurate travel time prediction is an important problem since it enables planning of cost effective vehicle routes and departure times, with the aim of saving time and fuel while reducing pollution. The main information source used is data from roadside double inductive loop sensors which measure vehicle speed, flow and density at specific locations. We model the spatiotemporal distribution of travel times by using local linear regression. The use of real-time data is very accurate for shorter journeys starting now and less reliable as journey times increase. Local linear regression can be used to optimally balance the use of historical and real time data. The main contribution of the paper is the extension of local linear models with higher order autoregressive travel time variables, namely vehicle flow data, and density data. Using two years of UK Highways Agency (HA) loop sensor data we found that the extended model significantly improves predictive performance while retaining the main benefits of earlier work: interpretability of linear models as well as computationally simple predictions.
Development of a Data-Driven Framework for Real-Time Travel Time Prediction
Computer-aided Civil and Infrastructure Engineering, 2016
Travel time prediction is one of the most important components in Intelligent Transportation Systems implementation. Various related techniques have been developed, but the efforts for improving the applicability of long-term prediction in a real-time manner have been lacking. Existing methods do not fully utilize the advantages of the state-of-the-art cloud system and large amount of data due to computation issues. We propose a new prediction framework for real-time travel time services in the cloud system. A distinctive feature is that the prediction is done with the entire data of a road section to stably and accurately produce the longterm (at least 6-hour prediction horizon) predicted value. Another distinctive feature is that the framework uses a hierarchical pattern matching called Multilevel k-nearest neighbor (Mk-NN) method which is compared with the conventional k-NN method and Nearest Historical average method. The results show that the method can more accurately and robustly predict the long-term travel time with shorter computation time.
Real-Time Travel Time Prediction Framework for Departure Time and Route Advice
Transportation Research Record: Journal of the Transportation Research Board, 2015
Heavily used urban networks remain a challenge for travel time prediction because traffic flow is rarely homogeneous and is also subject to a wide variety of disturbances. Various models, some of which use traffic flow theory and some of which are data driven, have been developed to predict traffic flow and travel times. Many of these perform well under set conditions. However, few perform well under all or even most urban traffic conditions. As part of the Amsterdam Practical Trial, a comprehensive field operation test for traffic management, a real-time travel time prediction framework, was developed to make use of an ensemble of traffic modeling techniques to predict travel times with great accuracy for arterial roads as well as urban roads. The various models in the framework include both traffic theoretical models and data-driven approaches, making use of some of the largest real-time traffic data sets currently available to limit errors to less than 20% for any time of day or ...