Multi-step prediction of experienced travel times using agent-based modeling

This paper develops an agent-based modeling approach to predict multi-step ahead experienced travel times using real-time and historical spatiotemporal traffic data. At the microscopic level, each agent represents an expert in a decision-making system. Each expert predicts the travel time for each time interval according to experiences from a historical dataset. A set of agent interactions is developed to preserve agents that correspond to traffic patterns similar to the real-time measurements and replace invalid agents or agents associated with negligible weights with new agents. Consequently, the aggregation of each agent’s recommendation (predicted travel time with associated weight) provides a macroscopic level of output, namely the predicted travel time distribution. Probe vehicle data from a 95-mile freeway stretch along I-64 and I-264 are used to test different predictors. The results show that the agent-based modeling approach produces the least prediction error compared to other state-of-the-practice and state-of-the-art methods (instantaneous travel time, historical average and k-nearest neighbor), and maintains less than a 9% prediction error for trip departures up to 60 min into the future for a two-hour trip. Moreover, the confidence boundaries of the predicted travel times demonstrate that the proposed approach also provides high accuracy in predicting travel time confidence intervals. Finally, the proposed approach does not require offline training thus making it easily transferable to other locations and the fast algorithm computation allows the proposed approach to be implemented in real-time applications in Traffic Management Centers.     

Dynamic travel time prediction using data clustering and genetic programming

The current state-of-practice for predicting travel times assumes that the speeds along the various roadway segments remain constant over the duration of the trip. This approach produces large prediction errors, especially when the segment speeds vary temporally. In this paper, we develop a data clustering and genetic programming approach for modeling and predicting the expected, lower, and upper bounds of dynamic travel times along freeways. The models obtained from the genetic programming approach are algebraic expressions that provide insights into the spatiotemporal interactions. The use of an algebraic equation also means that the approach is computationally efficient and suitable for real-time applications. Our algorithm is tested on a 37-mile freeway section encompassing several bottlenecks. The prediction error is demonstrated to be significantly lower than that produced by the instantaneous algorithm and the historical average averaged over seven weekdays (p-value <0.0001). Specifically, the proposed algorithm achieves more than a 25% and 76% reduction in the prediction error over the instantaneous and historical average, respectively on congested days. When bagging is used in addition to the genetic programming, the results show that the mean width of the travel time interval is less than 5 min for the 60–80 min trip.     

A bayesian dynamic linear model approach for real-time short-term freeway travel time prediction

This paper presents a Bayesian inference-based dynamic linear model (DLM) to predict online short-term travel time on a freeway stretch. The proposed method considers the predicted freeway travel time as the sum of the median of historical travel times, time-varying random variations in travel time, and a model evolution error, where the median is employed to recognize the primary travel time pattern while the variation captures unexpected supply (i.e. capacity) reduction and demand fluctuations. Bayesian forecasting is a learning process that revises sequentially the state of a priori knowledge of travel time based on newly available information. The prediction result is a posterior travel time distribution that can be employed to generate a single-value (typically but not necessarily the mean) travel time as well as a confidence interval representing the uncertainty of travel time prediction. To better track travel time fluctuations during non-recurrent congestion due to unforeseen events (e.g., incidents, accidents, or bad weather), the DLM is integrated into an adaptive control framework that can automatically learn and adjust the system evolution noise level. The experiment results based on the real loop detector data of an I-66 segment in Northern Virginia suggest that the proposed method is able to provide accurate and reliable travel time prediction under both recurrent and non-recurrent traffic conditions.     

Experienced travel time prediction for congested freeways

Travel time is an important performance measure for transportation systems, and dissemination of travel time information can help travelers make reliable travel decisions such as route choice or departure time. Since the traffic data collected in real time reflects the past or current conditions on the roadway, a predictive travel time methodology should be used to obtain the information to be disseminated. However, an important part of the literature either uses instantaneous travel time assumption, and sums the travel time of roadway segments at the starting time of the trip, or uses statistical forecasting algorithms to predict the future travel time. This study benefits from the available traffic flow fundamentals (e.g. shockwave analysis and bottleneck identification), and makes use of both historical and real time traffic information to provide travel time prediction. The methodological framework of this approach sequentially includes a bottleneck identification algorithm, clustering of traffic data in traffic regimes with similar characteristics, development of stochastic congestion maps for clustered data and an online congestion search algorithm, which combines historical data analysis and real-time data to predict experienced travel times at the starting time of the trip. The experimental results based on the loop detector data on Californian freeways indicate that the proposed method provides promising travel time predictions under varying traffic conditions.     

Evaluating the feasibility of a passive travel survey collection in a complex urban environment: Lessons learned from the New York City case study

The combination of increasing challenges in administering household travel surveys and advances in global positioning systems (GPS)/geographic information systems (GIS) technologies motivated this project. It tests the feasibility of using a passive travel data collection methodology in a complex urban environment, by developing GIS algorithms to automatically detect travel modes and trip purposes. The study was conducted in New York City where the multi-dimensional challenges include urban canyon effects, an extreme dense and diverse set of land use patterns, and a complex transit network. Our study uses a multi-modal transportation network, a set of rules to achieve both complexity and flexibility for travel mode detection, and develops procedures and models for trip end clustering and trip purpose prediction. The study results are promising, reporting success rates ranging from 60% to 95%, suggesting that in the future, conventional self-reported travel surveys may be supplemented, or even replaced, by passive data collection methods.     

A recursive, cell processing model for predicting freeway travel times

Travel time information influences driver behaviour and can contribute to reducing congestion and improving network efficiency. Consequently many road authorities disseminate travel time information on road side signs, web sites and radio traffic broadcasts. Operational systems commonly rely on speed data obtained from inductive loop detectors and estimate travel times using simple algorithms that are known to provide poor predictions particularly on either side of the peak period. This paper presents a new macroscopic model for predicting freeway travel times which overcomes the limitations of operational ‘instantaneous’ speed models by drawing on queuing theory to model the processing of vehicles in sections or cells of the freeway. The model draws on real-time speed, flow and occupancy data and is formulated to accommodate varying geometric conditions, the relative distribution of vehicles along the freeway, variations in speed limits, the impact of ramp flows and fixed or transient bottlenecks. Field validation of the new algorithm was undertaken using data from two operational freeways in Melbourne, Australia. Consistent with the results of simulation testing, the validation confirmed that the recursive model provided a substantial improvement in travel time predictions when compared to the model currently used to provide real-time travel time information to motorists in Melbourne.     

Identifying trip ends from raw GPS data with a hybrid spatio-temporal clustering algorithm and random forest model: a case study in Shanghai

ABSTRACT

Smartphones have been advocated as the preferred devices for travel behavior studies over conventional surveys. But the primary challenges are candidate stops extraction from GPS data and trip ends distinction from noise. This paper develops a Resident Travel Survey System (RTSS) for GPS data collection and travel diary verification, and then uses a two-step method to identify trip ends. In the first step, a density-based spatio-temporal clustering algorithm is proposed to extract candidate stops from trajectories. In the second step, a random forest model is applied to distinguish trip ends from mode transfer points. Results show that the clustering algorithm achieves a precision of 96.2%, a recall of 99.6%, mean absolute error of time within 3?min, and average offset distance within 30 meters. The comprehensive accuracy of trip ends identification is 99.2%. The two-step method performs well in trip ends identification and promotes the efficiency of travel survey systems.     

Real-time trip purpose prediction using online location-based search and discovery services

The use of smartphone technology is increasingly considered a state-of-the-art practice in travel data collection. Researchers have investigated various methods to automatically predict trip characteristics based upon locational and other smartphone sensing data. Of the trip characteristics being studied, trip purpose prediction has received relatively less attention. This research develops trip purpose prediction models based upon online location-based search and discovery services (specifically, Google Places API) and a limited set of trip data that are usually available upon the completion of the trip. The models have the potential to be integrated with smartphone technology to produce real-time trip purpose prediction. We use a recent, large-scale travel behavior survey that is augmented by downloaded Google Places information on each trip destination to develop and validate the models. Two statistical and machine learning prediction approaches are used, including nested logit and random forest methods. Both sets of models show that Google Places information is a useful predictor of trip purpose in situations where activity- and person-related information is uncollectable, missing, or unreliable. Even when activity- and person-related information is available, incorporating Google Places information provides incremental improvements in trip purpose prediction.     

Impact of telecommuting on spatial and temporal patterns of household travel

A spatial and temporal analysis of travel diary data collected during the State of California Telecommuting Pilot Project is performed to determine the impacts of telecommuting on household travel behavior. The analysis is based on geocoded trip data where missing trips and trip attributes have been augmented to the extent possible. The results confirm the earlier finding that the Pilot Project telecommuters substantially reduced travel; on telecommuting days, the telecommuters made virtually no commute trips, reduced peak-period trips by 60%, total distance traveled by 75%, and freeway miles by 90%. The spatial analysis of the trip records has shown that the telecommuters chose non-work destinations that are closer to home; they exhibited contracted action spaces after the introduction of telecommuting. Importantly, this contraction took place on both telecommuting days and commuting days. The telecommuters distributed their trips, over the day and avoided peak-period travel on telecommuting days. Non-work trips, however, show similar patterns of temporal distribution on telecommuting days and commuting days. Non-work trips continued to be made during the lunch period and late afternoon and evening hours.     

A genetic algorithm-based method for improving quality of travel time prediction intervals

The transportation literature is rich in the application of neural networks for travel time prediction. The uncertainty prevailing in operation of transportation systems, however, highly degrades prediction performance of neural networks. Prediction intervals for neural network outcomes can properly represent the uncertainty associated with the predictions. This paper studies an application of the delta technique for the construction of prediction intervals for bus and freeway travel times. The quality of these intervals strongly depends on the neural network structure and a training hyperparameter. A genetic algorithm–based method is developed that automates the neural network model selection and adjustment of the hyperparameter. Model selection and parameter adjustment is carried out through minimization of a prediction interval-based cost function, which depends on the width and coverage probability of constructed prediction intervals. Experiments conducted using the bus and freeway travel time datasets demonstrate the suitability of the proposed method for improving the quality of constructed prediction intervals in terms of their length and coverage probability.     

Modeling travel time reliability of freeways using risk assessment techniques

Travel time reliability is considered to be one of the key indicators for the performance of transport systems and is measured in various ways. This paper synthesizes both reliability concepts: traffic breakdown, the indicator of the instability of travel times, is treated as the risk, whereas travel time variability, the indicator of the uncertainty of travel times, is considered as the consequence of this risk. An analytical formula, using risk assessment technique, explicitly expresses the cost of travel time unreliability as the sum of the products of the consequences (i.e. variability) and the corresponding probabilities of breakdown. It provides a novel measure of travel time reliability and is applicable in network performance evaluations. An empirical example based on a large dataset of freeway traffic flow data from loop detectors shows that the developed travel time reliability measure is both intuitively logical and consistent.     

A discrete optimization approach for locating Automatic Vehicle Identification readers for the provision of roadway travel times

This paper develops an algorithm for optimally locating Automatic Vehicle Identification tag readers by maximizing the benefit that would accrue from measuring travel times on a transportation network. The problem is formulated as a quadratic 0–1 optimization problem where the objective function parameters represent benefit factors that capture the relevance of measuring travel times as reflected by the demand and travel time variability along specified trips. An optimization approach based on the Reformulation–Linearization Technique coupled with semidefinite programming concepts is designed to solve the formulated reader location problem. To illustrate the proposed methodology, we consider a transportation network that is comprised of freeway segments that might include merge, diverge, weaving, and bottleneck sections. In order to derive benefit factors for the various origin–destination pairs on this network, we employ a simulation package (INTEGRATION) in combination with a composite function, which estimates the travel time variability along a trip that is comprised of links that include any of the four identified sections. The simulation results are actually recorded as generic look-up tables that can be used for any such section for the purpose of computing the associated benefit factor coefficients. Computational results are presented using data pertaining to a freeway section in San Antonio, Texas, as well as synthetic test cases, to demonstrate the effectiveness of the proposed approach, and to study the sensitivity of the quality of the solution to variations in the number of available readers.     

Space and time related determinants of public transport use in trip chains

This research aims at gaining a better understanding about time and space related determinants, which are generally acknowledged to be important factors in the choice of transport mode. The effect of trip chaining is taken into account to improve the insight in the relation between the choice of transport mode and time factors. The data source is the first large scale Belgian mobility survey, carried out in 1998–1999, complemented with a newly created database, containing for each trip a calculated public transport trip. This allows comparing for each trip the actual travel time with the calculated travel time by public transport. Using elasticities and regression techniques the relation between travel time components and public transport use is quantified. On trip level, a clear relation is found between waiting and walking time and public transport use. On trip chain level, travel time variables for the whole trip chain such as the maximum and the range in the travel time ratio provide a significant improvement to the explanatory power of the regression model. The results contain parameters for model input and recommendations to public transport companies on information provision, intermodality and supply.     

Methodology for Validating Dynamic Origin–Destination Matrix Estimation Models with Implications for Advanced Traveler Information Systems

This paper describes a methodology for validating online dynamic O–D matrix estimation models using loop detector data in large-scale transportation networks. The simulation procedure focuses on travel aspects related to the collective trip structure of users, including the amount and duration of trips between O–D pairs, trip departure rates, average travel time from each origin and combinations of them. The analysis identifies emerging systematic patterns between these factors and issues related to the model performance, including network scale effects. This procedure aims to enhance the usage of prior O–D information based on, e.g. travel surveys, that are typically used in the estimation process. Moreover, it seeks to integrate the validation of dynamic O–D matrix estimation models with strategies for identifying target population groups for online planning and assessment of real-time travel information services within the context of Advanced Traveler Information Systems (ATIS).     

Link travel time inference using entry/exit information of trips on a network

This paper studies link travel time estimation using entry/exit time stamps of trips on a steady-state transportation network. We propose two inference methods based on the likelihood principle, assuming each link associates with a random travel time. The first method considers independent and Gaussian distributed link travel times, using the additive property that trip time has a closed-form distribution as the summation of link travel times. We particularly analyze the mean estimates when the variances of trip time estimates are known with a high degree of precision and examine the uniqueness of solutions. Two cases are discussed in detail: one with known paths of all trips and the other with unknown paths of some trips. We apply the Gaussian mixture model and the Expectation–Maximization (EM) algorithm to deal with the latter. The second method splits trip time proportionally among links traversed to deal with more general link travel time distributions such as log-normal. This approach builds upon an expected log-likelihood function which naturally leads to an iterative procedure analogous to the EM algorithm for solutions. Simulation tests on a simple nine-link network and on the Sioux Falls network respectively indicate that the two methods both perform well. The second method (i.e., trip splitting approximation) generally runs faster but with larger errors of estimated standard deviations of link travel times.     

Time use in travel surveys and time use surveys – Two sides of the same coin?

An in-depth understanding of travel behaviour determinants, including the relationship to non-travel activities, is the foundation for modelling and policy making. National Travel Surveys (NTS) and time use surveys (TUS) are two major data sources for travel behaviour and activity participation. The aim of this paper is to systematically compare both survey types regarding travel activities and non-travel activities. The analyses are based on the German National Travel Survey and the German National Time Use Survey from 2002.The number of trips and daily travel time for mobile respondents were computed as the main travel estimates. The number of trips per person is higher in the German TUS when changes in location without a trip are included. Location changes without a trip are consecutive non-trip activities with different locations but without a trip in-between. The daily travel time is consistently higher in the German TUS. The main reason for this difference is the 10-min interval used. Differences in travel estimates between the German TUS and NTS result from several interaction effects. Activity time in NTS is comparable with TUS for subsistence activities.Our analyses confirm that both survey types have advantages and disadvantages. TUS provide reliable travel estimates. The number of trips even seems preferable to NTS if missed trips are properly identified and considered. Daily travel times are somewhat exaggerated due to the 10-min interval. The fixed time interval is the most important limitation of TUS data. The result is that trip times in TUS do not represent actual trip times very well and should be treated with caution.We can use NTS activity data for subsistence activities between the first trip and the last trip. This can potentially benefit activity-based approaches since most activities before the first trip and after the last trip are typical home-based activities which are rarely substituted by out-of-home activities.     

Door-to-door travel times in RP departure time choice models: An approximation method using GPS data

A common way to determine values of travel time and schedule delay is to estimate departure time choice models, using stated preference (SP) or revealed preference (RP) data. The latter are used less frequently, mainly because of the difficulties to collect the data required for the model estimation. One main requirement is knowledge of the (expected) travel times for both chosen and unchosen departure time alternatives. As the availability of such data is limited, most RP-based scheduling models only take into account travel times on trip segments rather than door-to-door travel times, or use very rough measures of door-to-door travel times. We show that ignoring the temporal and spatial variation of travel times, and, in particular, the correlation of travel times across links may lead to biased estimates of the value of time (VOT). To approximate door-to-door travel times for which no complete measurement is possible, we develop a method that relates travel times on links with continuous speed measurements to travel times on links where relatively infrequent GPS-based speed measurements are available. We use geographically weighted regression to estimate the location-specific relation between the speeds on these two types of links, which is then used for travel time prediction at different locations, days, and times of the day. This method is not only useful for the approximation of door-to-door travel times in departure time choice models, but is generally relevant for predicting travel times in situations where continuous speed measurements can be enriched with GPS data.     

Travel demand (modal split) estimation by hierarchical measurement

This paper develops a procedure for travel demand estimation via the Saaty method of Analytic Hierarchy. A stratification of the travel demand by trip-making and trip attributes has been represented more inclusively in a hierarchy system. Various elements and dimensions of the hierarchy have been hypothesized as different levels of decisions made by trip-makers. The elements contained in a set of specified matrices of travel attributes have been weighed utilizing a ratio scale, in a process of mapping transportation systems (modal) attributes with the characteristic trip-making behavior in a hierarchical demand structure. The principal output of this procedure is an estimate of the trip distribution by mode, or modal split. The estimate closely approximates the observed modal split pattern for the inter city travel problem simulated. This procedure is proposed for travel demand forecasting and planning.     

Ramp meters on trial: Evidence from the Twin Cities metering holiday

Ramp meters in the Twin Cities have been the subject of a recent test of their effectiveness, involving turning them off for eight weeks. This paper analyzes the results with and without ramp metering for several representative freeways during the afternoon peak period. Seven performance measures: mobility, equity, productivity, consumers’ surplus, accessibility, travel time variation and travel demand responses are compared. It is found that ramp meters are particularly helpful for long trips relative to short trips. Ramp metering, while generally beneficial to freeway segments, may not improve trip travel times (including ramp delays). The reduction in travel time variation comprises another benefit from ramp meters. Non-work trips and work trips respond differently to ramp meters. The results are mixed, suggesting a more refined ramp control algorithm, which explicitly considers ramp delay, is in order.     

The rational locator reexamined: Are travel times still stable?

The rational locator hypothesis posits that individuals can, if they choose, maintain approximately steady journey-to-work travel times by adjusting their home and workplace. This hypothesis was coupled with the observation of long-term stability in drive alone journey-to-work times in metropolitan Washington (those times were unchanged from 1957 through 1968 to 1988). Despite the increase of average commuting distance and congestion, trip duration remained constant or even declined when controlling for travel purpose and travel mode because of shifting a share of traffic from slow urban routes to faster suburban routes. This observation has significance, as it is important to know for travel demand analysis if there is an underlying budget, or even a regularity, as this helps us determine whether our forecasts are reasonable. To re-test the underlying rationale for the hypothesis that travel times are stable, intra-metropolitan comparisons of travel times are made using Washington DC data from 1968, 1988, and 1994, and Twin Cities data from 1990 and 2000. The results depend upon geography. For the larger Washington DC region, keeping the same geography shows little change in commute times, but using the larger 1994 area suggests an increase in commute times. However, the Twin Cities, starting from a much shorter commute time, shows a marked increase over the decade, using either the smaller or the larger geography. Despite the remarkable continuing observation of stability in drive alone commuting times in metropolitan Washington, we reject the theory of personal commuting budgets, as we find that not only are commuting times not generally stable over time at the intra-metropolitan area, but that commuting time clearly depends on metropolitan spatial structure.