Forecasting Multiple-Period Freeway Link Travel Times Using Modular Neural Networks

1998 ◽  
Vol 1617 (1) ◽  
pp. 163-170 ◽  
Author(s):  
Dongjoo Park ◽  
Laurence R. Rilett
Keyword(s):  
Travel Time ◽  
Real Time ◽  
Transportation Systems ◽  
Time Profile ◽  
Identification System ◽  
Route Guidance ◽  
Travel Times ◽  
Test Bed ◽  
Link Travel Time ◽  
Time Information

With the advent of route guidance systems (RGS), the prediction of short-term link travel times has become increasingly important. For RGS to be successful, the calculated routes should be based on not only historical and real-time link travel time information but also anticipatory link travel time information. An examination is conducted on how realtime information gathered as part of intelligent transportation systems can be used to predict link travel times for one through five time periods (of 5 minutes’ duration). The methodology developed consists of two steps. First, the historical link travel times are classified based on an unsupervised clustering technique. Second, an individual or modular artificial neural network (ANN) is calibrated for each class, and each modular ANN is then used to predict link travel times. Actual link travel times from Houston, Texas, collected as part of the automatic vehicle identification system of the Houston Transtar system were used as a test bed. It was found that the modular ANN outperformed a conventional singular ANN. The results of the best modular ANN were compared with existing link travel time techniques, including a Kalman filtering model, an exponential smoothing model, a historical profile, and a real-time profile, and it was found that the modular ANN gave the best overall results.

Prediction Model of Bus Arrival Time for Real-Time Applications

2005 ◽  
Vol 1927 (1) ◽  
pp. 195-204 ◽  
Author(s):  
Ranhee Jeong ◽  
Laurence R. Rilett
Keyword(s):  
Prediction Model ◽  
Travel Time ◽  
Real Time ◽  
Arrival Time ◽  
Historical Data ◽  
Transportation Systems ◽  
Test Bed ◽  
Travel Time Prediction ◽  
Ann Models ◽  
Real Time Applications

Advanced traveler information systems (ATIS) are one component of intelligent transportation systems (ITS), and a major component of ATIS is travel time information. Automatic vehicle location (AVL) systems, which are a part of ITS, have been adopted by many transit agencies to track their vehicles and to predict travel time in real time. Because of the complexity involved, there is no universally adopted approach for this latter application, and research is needed in this area. The objectives of the research in this paper are to develop a model to predict bus arrival time using AVL data and apply the model for real-time applications. The test bed was a bus route located in Houston, Texas, and the travel time prediction model considered schedule adherence, traffic congestion, and dwell times. A historical data-based model, regression models, and artificial neural network (ANN) models were used to predict bus arrival time. It was found that ANN models outperformed both the historical data-based model and the regression model in terms of prediction accuracy. It was also found that the ANN models can be used for real-time applications.

Estimating Travel Time Summary Statistics of Larger Intervals from Smaller Intervals Without Storing Individual Data

2002 ◽  
Vol 1804 (1) ◽  
pp. 39-47 ◽  
Author(s):  
Dongjoo Park ◽  
Laurence R. Rilett ◽  
Parichart Pattanamekar ◽  
Keechoo Choi
Keyword(s):  
Travel Time ◽  
Time Estimation ◽  
Transportation Systems ◽  
Identification System ◽  
Time Interval ◽  
Travel Time Estimation ◽  
Time Data ◽  
Probe Vehicles ◽  
Mean And Variance ◽  
Link Travel Time

Historically, real-time intelligent transportation systems data are aggregated into discrete periods, typically of 5 to 10 min duration, and are subsequently used for travel time estimation and forecasting. In a previous study of link and corridor travel time estimation and forecasting by using probe vehicles, it was shown that the optimal aggregation interval size is a function of the traffic condition and the application. It is expected that traffic management centers will continue to collect travel time statistics (e.g., mean and variance) from probe vehicles and archive this data at a minimum time interval. Statistical models are developed for estimating the mean and variance of the link and route or corridor travel time for a larger interval by using only the observed mean travel time and variance for each smaller or basic interval. The proposed models are demonstrated by using travel time data obtained from Houston, Texas, which were collected as part of the automatic vehicle identification system of the Houston TranStar system. It was found that the proposed models for estimating link travel time mean and variance for a larger interval were easy to implement and provided results that had minimal error. The route or corridor travel time mean and variance model had considerable error compared with the link travel time mean and variance models.

scholarly journals Highway travel time information system based on cumulative count curves and new tracking technologies

2016 ◽  
Author(s):  
Margarita Martínez-Díaz ◽  
Francesc Soriguera Martí ◽  
Ignacio Pérez Pérez
Keyword(s):  
Travel Time ◽  
Real Time ◽  
New Technologies ◽  
Time Estimation ◽  
Travel Times ◽  
Important Indicator ◽  
Loop Detectors ◽  
Real Time Estimation ◽  
Cumulative Count ◽  
Time Information

Travel time is probably the most important indicator of the level of service of a highway, and it is also the most appreciated information for its users. Administrations and private companies make increasing efforts to improve its real time estimation. The appearance of new technologies makes the precise measurement of travel times easier than never before. However, direct measurements of travel time are, by nature, outdated in real time, and lack of the desired forecasting capabilities. This paper introduces a new methodology to improve the real time estimation of travel times by using the equipment usually present in most highways, i.e., loop detectors, in combination with Automatic Vehicle Identification or Tracking Technologies. One of the most important features of the method is the usage of cumulative counts at detectors as an input, avoiding the drawbacks of common spot-speed methodologies. Cumulative count curves have great potential for freeway travel time information systems, as they provide spatial measurements and thus allow the calculation of instantaneous travel times. In addition, they exhibit predictive capabilities. Nevertheless, they have not been used extensively mainly because of the error introduced by the accumulation of the detector drift. The proposed methodology solves this problem by correcting the deviations using direct travel time measurements. The method results highly beneficial for its accuracy as well as for its low implementation cost.DOI: http://dx.doi.org/10.4995/CIT2016.2016.3209 

Evaluating Effectiveness of Real-Time Advanced Traveler Information Systems Using a Small Test Vehicle Fleet

1997 ◽  
Vol 1588 (1) ◽  
pp. 41-48 ◽  
Author(s):  
Christopher L. Saricks ◽  
Joseph L. Schofer ◽  
Siim Sööt ◽  
Paul A. Belella
Keyword(s):  
Travel Time ◽  
Real Time ◽  
Time Of Day ◽  
Guidance System ◽  
Effective Characteristics ◽  
Local Network ◽  
Route Guidance ◽  
Time Data ◽  
Traveler Information ◽  
Link Travel Time

ADVANCE was an in-vehicle advanced traveler information system (ATIS) providing route guidance in real time that operated in the northwestern portion and northwest suburbs of Chicago, Illinois. It used probe vehicles to generate dynamically travel time information about expressways, arterials, and local streets. Tests to evaluate the subsystems of ADVANCE, executed with limited availability of test vehicles and stringent scheduling, are described; they provided useful insights into both the performance of the ADVANCE system as a whole and the desirable and effective characteristics of ATIS deployments generally. Tests found that the user features of an in-route guidance system must be able to accommodate a broad range of technological sophistication and network knowledge among the population likely to become regular users of such a system. For users who know the local network configuration, only a system giving reliable real-time data about nonrecurrent congestion is likely to find a market base beyond specialized applications. In general, the quality and usefulness of systemwide real-time route guidance provided by other means are enhanced significantly by even a small deployment of probes: probe data greatly improve static (archival average) link travel time estimates by time of day, although the guidance algorithms that use these data should also include arterial traffic signal timings. Moreover, probe- and detector-based incident detection on arterial networks shows considerable promise for improved performance and reliability.

Predicting Travel Times for the South Jersey Real-Time Motorist Information System

2003 ◽  
Vol 1855 (1) ◽  
pp. 32-40 ◽  
Author(s):  
Steven I. J. Chien ◽  
Xiaobo Liu ◽  
Kaan Ozbay
Keyword(s):  
Information System ◽  
Prediction Model ◽  
Simulation Model ◽  
Travel Time ◽  
Real Time ◽  
Travel Times ◽  
The South ◽  
Travel Time Prediction ◽  
Real Time Information ◽  
Time Information

A dynamic travel-time prediction model was developed for the South Jersey (southern New Jersey) motorist real-time information system. During development and evaluation of the model, the integration of traffic flow theory, measurement and application of collected data, and traffic simulation were considered. Reliable prediction results can be generated with limited historical real-time traffic data. In the study, acoustic sensors were installed at potential congested places to monitor traffic congestion. A developed simulation model was calibrated with the data collected from the sensors, and this was applied to emulate traffic operations and evaluate the proposed prediction model under time-varying traffic conditions. With emulated real–time information (travel times) generated by the simulation model, an algorithm based on Kalman filtering was developed and applied to forecast travel times for specific origin-destination pairs over different periods. Prediction accuracy was evaluated by the simulation model. Results show that the developed travel-time predictive model demonstrates satisfactory performance.

Telecommunications- and Information Technology–Inspired Analyses: Review of an Intelligent Transportation Systems Experience

2017 ◽  
Vol 2658 (1) ◽  
pp. 44-55 ◽  
Author(s):  
Yupo Chan
Keyword(s):  
Information Technology ◽  
Big Data ◽  
Real Time ◽  
Traffic Management ◽  
Transportation Systems ◽  
Lessons Learned ◽  
Connected Vehicles ◽  
Ongoing Research ◽  
Real Time Information ◽  
Time Information

This paper reviews both the author’s experience with managing highway network traffic on a real-time basis and the ongoing research into harnessing the potential of telecommunications and information technology (IT). On the basis of the lessons learned, this paper speculates about how telecommunications and IT capabilities can respond to current and future developments in traffic management. Issues arising from disruptive telecommunications technologies include the ready availability of real-time information, the crowdsourcing of information, the challenges of big data, and the need for information quality. Issues arising from transportation technologies include autonomous vehicles and connected vehicles and new taxi-like car- and bikesharing. Illustrations are drawn from the following core functions of a traffic management center: ( a) detecting and resolving an incident (possibly through crowdsourcing), ( b) monitoring and forecasting traffic (possibly through connected vehicles serving as sensors), ( c) advising motorists about routing alternatives (possibly through real-time information), and ( d) configuring traffic control strategies and tactics (possibly though big data). The conclusion drawn is that agility is the key to success in an ever-evolving technological scene. The solid guiding principle remains innovative and rigorous analytical procedures that build on the state of the art in the field, including both hard and soft technologies. The biggest modeling and simulation challenge remains the unknown, including such rapidly emerging trends as the Internet of things and the smart city.

Dynamic Value Pricing on I-15 in San Diego: Impact on Travel Time and Its Reliability

2003 ◽  
Vol 1839 (1) ◽  
pp. 45-54 ◽  
Author(s):  
Janusz Supernak ◽  
Christine Kaschade ◽  
Duane Steffey
Keyword(s):  
Travel Time ◽  
San Diego ◽  
Time Profile ◽  
Travel Times ◽  
Case Scenario ◽  
Worst Case ◽  
San Diego State University ◽  
Time Changes ◽  
Value Pricing ◽  
Express Lanes

Selected results are presented of the Traffic Study, one of 12 studies conducted by San Diego State University for the I-15 Congestion (Value) Pricing Project in San Diego, a 3-year demonstration. The focus is on the project's impact on travel times and their distribution on both the main lanes and the express lanes of I-15 for both ExpressPass and FasTrak phases of the project. Specifically addressed is the issue of reliability of on-time arrival enjoyed by the FasTrak subscribers and the high variability of travel times for the I-15 travelers who use only main lanes of I-15 for their commute. Examination of the ramp and freeway delays shows that in the worst-case scenario, FasTrak subscribers who use express lanes can save up to 20 min avoiding delay on the I-15 main lanes. This finding agrees with the drivers’ perceptions about their time savings when using FasTrak. Travel-time changes during the duration of the project also are examined. There were substantial year-to-year changes in travel times along the I-15 main lanes and the I-8 lanes used as control. The travel-time profile along the I-15 main lanes differed significantly from the profile along I-8, the control corridor, in both a.m. and p.m. peak periods.

Minimum Sampling Size of Floating Cars for Urban Link Travel Time Distribution Estimation

2019 ◽  
Vol 2673 (3) ◽  
pp. 24-43 ◽  
Author(s):  
Meiping Yun ◽  
Wenwen Qin
Keyword(s):  
Sample Size ◽  
Travel Time ◽  
Real Time ◽  
Time Distribution ◽  
Ground Truth ◽  
Sample Size Determination ◽  
Minimum Sample Size ◽  
Travel Time Distribution ◽  
Minimum Sample ◽  
Link Travel Time

Despite the wide application of floating car data (FCD) in urban link travel time estimation, limited efforts have been made to determine the minimum sample size of floating cars appropriate to the requirements for travel time distribution (TTD) estimation. This study develops a framework for seeking the required minimum number of travel time observations generated from FCD for urban link TTD estimation. The basic idea is to test how, with a decreasing the number of observations, the similarities between the distribution of estimated travel time from observations and those from the ground-truth vary. These are measured by employing the Hellinger Distance (HD) and Kolmogorov-Smirnov (KS) tests. Finally, the minimum sample size is determined by the HD value, ensuring that corresponding distribution passes the KS test. The proposed method is validated with the sources of FCD and Radio Frequency Identification Data (RFID) collected from an urban arterial in Nanjing, China. The results indicate that: (1) the average travel times derived from FCD give good estimation accuracy for real-time application; (2) the minimum required sample size range changes with the extent of time-varying fluctuations in traffic flows; (3) the minimum sample size determination is sensitive to whether observations are aggregated near each peak in the multistate distribution; (4) sparse and incomplete observations from FCD in most time periods cannot be used to achieve the minimum sample size. Moreover, this would produce a significant deviation from the ground-truth distributions. Finally, FCD is strongly recommended for better TTD estimation incorporating both historical trends and real-time observations.

Advanced Techniques for Travel Time Data Collection

1996 ◽  
Vol 1551 (1) ◽  
pp. 51-58 ◽  
Author(s):  
Shawn M. Turner
Keyword(s):  
Data Collection ◽  
Travel Time ◽  
Real Time ◽  
Cellular Phone ◽  
License Plate ◽  
Time Data ◽  
Video Imaging ◽  
Advantages And Disadvantages ◽  
Travel Time Data ◽  
Time Information

Travel time information is becoming more important for applications ranging from congestion measurement to real-time travel information. Several advanced techniques for travel time data collection are discussed, including electronic distance-measuring instruments (DMIs), computerized and video license plate matching, cellular phone tracking, automatic vehicle identification (AVI), automatic vehicle location (AVL), and video imaging. The various advanced techniques are described, the necessary equipment and procedures are outlined, the applications of each technique are discussed, and the advantages and disadvantages are summarized. Electronic DMIs are low in cost but typically limited to congestion monitoring applications. Computerized and video license plate matching are more expensive and would be most applicable for congestion measurement and monitoring. Cellular phone tracking, AVI, and AVL systems may require a significant investment in communications infrastructure, but they can provide real-time information. Video imaging is still in testing stages, with some uncertainty about costs and accuracy.

Sign in / Sign up

Export Citation Format

Share Document