Matthew G Karlaftis

Short‐term traffic forecasting: Overview of objectives and methods

In the last two decades, the growing need for short‐term prediction of traffic parameters embedded in a real‐time intelligent transportation systems environment has led to the development of a vast number of forecasting algorithms. Despite this, there is still not a clear view about the various requirements involved in modelling. This field of research was examined by disaggregating the process of developing short‐term traffic forecasting algorithms into three essential clusters: the determination of the scope, the conceptual process of specifying the output and the process of modelling, which includes several decisions concerning the selection of the proper methodological approach, the type of input and output data used, and the quality of the data. A critical discussion clarifies several interactions between the above and results in a logical flow that can be used as a framework for developing short‐term traffic forecasting models.

Spatio-Temporal Short-Term Urban Traffic Volume Forecasting Using Genetically Optimized Modular Networks

Current interest in short-term traffic volume forecasting focuses on incorporating temporal and spatial volume characteristics in the forecasting process. This paper addresses the problem of integrating and optimizing predictive information from multiple locations of an urban signalized arterial roadway and proposes a modular neural predictor consisting of temporal genetically optimized structures of multilayer perceptrons that are fed with volume data from sequential locations to improve the accuracy of short-term forecasts. Results show that the proposed methodology provides more accurate forecasts compared to the conventional statistical methodologies applied, as well as to the static forms of neural networks.

COST STRUCTURES OF PUBLIC TRANSIT SYSTEMS: A PANEL DATA ANALYSIS

Abstract Results from numerous cost studies have generated conflicting results on public transit production technologies. Because prior studies have employed various sub-samples of public transit properties, the diverse results may either reflect alternative technologies or sampling differences. Based on a panel of nearly all transit systems reporting for Section 15 (US National Transit Data) purposes from 1986 to 1994, this paper explores whether public transit production technology differs by size and operating characteristics of the system. The results indicate that US transit properties are heterogeneous with different production technologies, which implies that transit cost analyses based upon a set of heterogeneous systems will generate incorrect inferences on public transit cost and production structures.

Temporal Evolution of Short‐Term Urban Traffic Flow: A Nonlinear Dynamics Approach

Recognizing temporal patterns in traffic flow has been an important consideration in short- term traffic forecasting research. However, little work has been conducted on identifying and associating traffic pattern occurrence with prevailing traffic con- ditions. We propose a multilayer strategy that first identifies patterns of traffic based on their structure and evolution in time and then clusters the pattern- based evolution of traffic flow with respect to pre- vailing traffic flow conditions. Temporal pattern iden- tification is based on the statistical treatment of the recurrent behavior of jointly considered volume and oc- cupancy series; clustering is done via a two-level neural network approach. Results on urban signalized arterial 90-second traffic volume and occupancy data indicate that traffic pattern propagation exhibits variability with respect to its statistical characteristics such as determinis- tic structure and nonlinear evolution. Further, traffic pat- tern clustering uncovers four distinct classes of traffic pat- tern evolution, whereas transitional traffic conditions can be straightforwardly identified.

Pattern-Based Short-Term Urban Traffic Predictor

Short-term urban traffic predictor is a prediction system based on artificial intelligence and advanced analysis of nonlinear dynamics of short-term traffic flow. The aim is to address: (1) the need for accurate anticipated traffic information, (2) the ability to cope with various traffic conditions in an integrated up-to-date functional structure and (3) the need for accurate and timely predictive short-term traffic information in an extended time horizon in cases of data collection malfunction. The conceptual and functional framework of short-term urban traffic predictor is presented. The above are encapsulated in a user-friendly application. Finally, some interesting results regarding computational time and accuracy of recursive predictions are presented

ITS Solutions and Accident Risks: Prospective and Limitations

Abstract This article investigates the prospective and limitations in the application of potential intelligent transport system (ITS) functions to reduce accident risks, using a cause‐treatment relationship. The main causes of road accidents are described and appropriate ITS solutions (including advanced driver assistance systems and advanced traveller information systems) are presented as countermeasures. Anticipated impacts are discussed and indicate that several ITS have the potential of improving road safety and addressing specific accident causes. However, attention is required on particular aspects of their implementation as they may trigger adverse effects by imposing behavioural adaptation risks, and overestimation and over‐reliance on system capabilities. Further, user acceptability and strategic implementation issues are paramount to the successful introduction of these systems.

Project-Level Life-Cycle Benefit–Cost Analysis Approach for Evaluating Highway Segment Safety Hardware Improvements

This paper presents a methodology for a benefit–cost analysis of improving the conditions of highway segment safety hardware over its life cycle. In the framework, a safety index is established for assessing the risks of vehicle crashes on a highway segment. These risks are associated with the safety-related attributes of the segment, which include traffic volume, segment length, standards and consistency of geometric design, pavement surface condition, safety hardware condition, and roadside features. The annual potential for safety improvements (PSI) associated with safety hardware improvements is calculated as reductions in fatal, injury, and property-damage-only crashes on the highway segment on the basis of the without- and with-improvement safety indices. Then, by monetizing and extrapolating the annual PSI over the longest hardware service life, the maximum life-cycle benefits of safety hardware improvements for the highway segment are established. To demonstrate the application of the methodology, data on 193 highway segments in Ozaukee County, Wisconsin, are collected and analyzed and the results are validated using the root-mean-square error test, chi-square test, Spearmans rank correlation test, and the Mann–Whitney test. A direction for future research is recommended to simplify the methodology and to prepare guidelines for enhancing its practical implementation.

Advanced Computational Approaches for Predicting Tourist Arrivals: the Case of Charter Air-Travel

Tourism is one of the major industries profiting various sectors of the economy, such as the transportation, accommodation, entertainment and so on. According to the World Tourism Organization (2008), international tourism grew at around 5% during the first four months of the year 2008. Fastest growth was observed in the Middle East, North-East and South Asia, and Central and South America. Even though, uncertainty over the global economic situation is affecting consumer confidence and could hurt tourism demand, for 2008 as a whole, UNWTO maintains a cautiously positive forecast. Moreover, international trends show that tourists are becoming more discerning in their choice of destinations and, therefore, becoming less predictable and more spontaneous in terms of their consumption patterns (Burger et al. 2001). Air transportation is probably the most important mode for international travel and leisure. A typical characteristic of air tourism in Europe is the extensive use of nonscheduled/charter flights and the existence of low-cost carriers in the leisure travel market, that account for 8% of passengers and 3% or revenues in the aviation industry (Dresner 2006). Non-scheduled demand is typical in Mediterranean countries where connections are essentially touristic and characterized by non-scheduled services. In this type of air travel, the ability to accurately predict tourist arrivals is of importance in the successful management and operation of the airport facilities, as well as the adjacent transportation network. Yet, the literature has little to offer in modeling demand stemming from non-scheduled flights, as such series exhibit seasonality, intense variability and inherent unpredictability. This paper develops and tests advanced computational approaches in order to predict nonscheduled/charter international tourist demand. The computational challenges that may arise in such a problem are twofold: first, to treat seasonal and stochastic characteristics of non-scheduled tourist demand, and, second, to develop models that consider past tourist demand characterists. This paper focuses on developing ARFIMA models that consider both non-stationarity and long-term memory effects in the auto-regressive process and temporal neural networks with advance genetically optimized characteristics that treat both nonlinearity and non-stationarity.

Developing Public Transport Network systems: The DIANA approach

In this paper we introduce the project DIANA, which deals with the development of innovative algorithms and decision support systems for the design of public transport network systems. The project aims to design transportation networks with conventional and electric vehicle types, with the objectives of maximizing total welfare, including the minimization of system emissions. Evolutional algorithms and reinforcement learning methods are being developed for solving the associated transit route network design problem. Finally, a web-based decision support system is under development utilizing state of the art GIS technology.

An Information Fusion Framework of Traffic Counts Forecasts Based on Concepts from Fuzzy Set Theory.

Abstract Reliable surveillance of urban networks coupled with techniques for information acquisition on traffic states provides the basis for the deployment of Advanced Transportation Management and Information Systems (ATMIS). Since information can be collected from various sources, a gamut of approaches for the fusion of available data has been utilized in traffic control centers. This paper focuses on a special paradigm of data fusion that combines information on forecasted traffic volume obtained from a variety of alternative approaches and provides a novel forecasting scheme that treats uncertainty by adopting concepts from fuzzy set theory and expert knowledge.