Travel Demand Prediction Research Articles

How Weather Events Affect Freeway Demand Patterns

The ability to estimate freeway traffic demand during inclement weather events is critically needed for demand and supply management, level-of-service assessment, capacity verification, and other freeway operational and planning applications. Reliable travel demand prediction is also a useful input for freeway travel time reliability analyses and advanced traveler information systems. This paper investigates the effect of inclement weather on freeway demand with the use of hourly traffic and weather data collected for 5 years (2011–2015) at 15 urban freeway locations in the city of Milwaukee, Wisconsin, which represents a North American region experiencing distinct seasons with significant snowfall and cold temperatures in the winter. About 489,300 hourly traffic demand observations were used in the analysis, and more than 52,800 of those were under inclement weather. Freeway demand reduction factors were introduced for diverse weather variables, precipitation intensities, visibility levels, temperatures, weekdays and weekends, and different times of day. Variability across seasons and sites was also examined. The paper further explores traffic demand in the hours immediately preceding or following inclement weather events. Finally, freeway weather-related demand and capacity adjustment factors are compared. Lessons learned from the study along with recommendations for future Highway Capacity Manual editions are summarized.

A simulation study of the impact of the public-private partnership strategy on the performance of transport infrastructure.

The choice of investment strategy has a great impact on the performance of transport infrastructure. Positive projects such as the “Subway plus Property” model in Hong Kong have created sustainable financial profits for the public transport projects. Owing to a series of public debt and other constraints, public–private partnership (PPP) was introduced as an innovative investment model to address this issue and help develop transport infrastructure. Yet, few studies provide a deeper understanding of relationships between PPP strategy and the performance of such transport projects (particularly the whole transport system). This paper defines the research scope as a regional network of freeway. With a popular PPP model, travel demand prediction method, and relevant parameters as input, agents in a simulation framework can simulate the choice of PPP freeway over time. The simulation framework can be used to analyze the relationship between the PPP strategy and performance of the regional freeway network. This study uses the Freeway Network of Yangtze River Delta (FN-YRD) in China as the context. The results demonstrate the value of using simulation models of complex transportation systems to help decision makers choose the right PPP projects. Such a tool is viewed as particularly important given the ongoing transformation of functions of the Chinese transportation sector, including franchise rights of transport projects, and freeway charging mechanism.

Choice Set Formation Behavior: Joint Mode and Route Choice Selection Model

To improve the level of service of traffic and predict travel demand, it is essential to analyze the behavioral factors that affect transportation mode choice and route choice at the individual level. Such analysis requires detailed data on the behavior of people in the selection of different modes and routes. This paper presents a unique data collection endeavor, intended to observe the formation of the choice set from which the final alternative is selected. The revealed preference information about the routes people considered and used for their last work or study trip was targeted. Google Maps API (application programming interface), which has the capacity to calculate and return car and public transportation routes, was employed to program the survey and adaptively show the respondents the routes according to the reported origin and destination. A pilot survey was conducted with this survey tool at the University of New South Wales, Sydney, Australia, with a sample of 200 respondents. A preliminary analysis was carried out to analyze the effectiveness of the survey tool and the specification of the choice set. Three modeling structures—multinomial logit (MNL), nested logit, and mixed MNL—were used to estimate the parameters of the preliminary analysis. The results were fairly intuitive as far as the signs of the parameters were concerned, and travel time significantly influenced route choice.

Implementation of Transport Data in to the Transport Forecasting in Slovakia

The transport forecast is the unconditional part of every transport studies, LOS capacity calculations or town transport plans. It is processed for several time horizons or time intervals.The basic purpose of transportation planning and management is to match transportation supply with travel demand, which represents ‘need’. A thorough understanding of existing travel pattern is necessary for identifying and analyzing existing traffic related problems. Detailed data on current travel pattern and traffic volumes are needed also for developing travel forecasting/prediction models. The prediction of future travel demand is an essential task of the long-range transportation planning process for determining strategies for accommodating future needs.Further sustainable mobility in Slovak cities, their competitiveness, economy as well as living conditions of inhabitants are depend also on the way of transport planning, development of transport infrastructure and services.Generally the calculation of the interzonal transport relations (number of trips) is a challenge. The practical application of the transport forecast theory is not easy. Some steps are hardly implemented to the practice for various reasons. The discrepancy between needed data and available data is one of the common problem. The traffic modelers must often study the method of data gathering which use an institutions possessing of demanded data. The discrepancies are showed on different levels (for examples the borders of urban zone and statistic area units are different in principle). One of the problematic issue is the definition of work trips or business trips. The origin and destination is defined at office address. The mobility survey contains data characterized one household, people not admit usual mentioned trips. The official company address are different as transport department. The ratio of these trips in traffic flow is not negligible. The article is aimed to brief description of the transport forecast process calculation. The main part of the article describes outputs from mobility surveys of some regions of Slovakia and them implementation in transport forecast calculation process. This methodically tool is intended for traffic analysis, planned solutions and arrangements verification of problem situations solutions.The University of Žilina processed several mobility surveys. The sample covers different regions with different gravity areas. The database was detailed evaluated and compared. Two Slovak regions were modeled in the Visum software (PTV Vision). Output data were used for town transport plans. Other mobility surveys were made for analysis of interstate transport relations (Slovak republic – Hungary).

Shaping the Future: Case Studies in UK Accessibility Planning

There has been an extensive academic debate about the theory of accessibility planning but far less discussion of the practice. This paper illustrates how accessibility planning has succeeded in delivering more sustainable solutions yet failed to change mainstream transport delivery, and what this means for future transport policy directions in the UK.The best way to predict the future is to shape it. Transport planning has a poor track record in predicting travel demand, so accessibility planning emerged in the 1990s as a promising approach to help shape travel demand. It was hoped that with a focus on accessibility a more optimal balance between supply and demand in transport could be achieved, consistent with sustainability. Since then there has been a steady expansion in the number of businesses and other employers planning access for their staff and in local authorities improving access to local services and facilities. However after 20 years, mainstream accessibility plans cover only a small minority of people and organizations. The focus of delivery remains social groups who face particular problems, such as personal mobility restrictions. Responding to problems is only one motivation for action and successful approaches to widen the coverage of accessibility planning have relied on audits, funding, and legislative requirements. The paper presents a typology projects to explain the scope and limits of current UK accessibility planning. It identifies how projects can change land use plans, invest in new safe routes to shops, devise new ticketing and payment approaches for passenger transport, and develop new solutions for sharing of cars, taxis and cycles to make transport more available to more people. The experiences from practice show that there are important obstacles to cross sector working and evidence led planning. The transport sector continues to be reluctant to lead complex cross sector programmes, and there is an evolving policy debate about leadership of transport, accessibility planning and sustainable development. The future of accessibility planning is secured by the essential needs of people to reach opportunities, goods and services, but the organization of delivery needs new funding and focus.

Study on Transportation Development Mode Selection for the Growth of Employment in City Center - Case Study of the Expansion of Financial District in Beijing

Urban expansion and development has brought the growth of employment centers. How to meet the needs of employment travel activities with transport services becomes a bottleneck restricting urban development. Setting the expansion and development of Beijing financial district as an example, this article systematically sorts out the relation between urban form and selection of transportation development mode. A travel demand forecasting model is set up to predict travel demand triggered by the employment centers concentration. Through scenario analysis of traffic patterns, the article analyzes the matching transportation development direction, and proposes metro transit as the only solution to traffic demand of the financial center. On the basis of analyzing metro transit need, it puts forward a further plan of metro transit lines, and measures such as corresponding car control, and green travel guide. The research findings will provide support for the development of financial centers, and reference to more cities that facing the growth of employment centers.

Inter-temporal variation in the travel time and travel cost parameters of transport models

The parameters for travel time and travel cost are central in travel demand forecasting models. Since valuation of infrastructure investments requires prediction of travel demand for future evaluat ...

Travel Mode Choice Modeling: A Comparison of Bayesian Networks and Neural Networks

In recent years, Bayesian networks and neural networks have been widely applied to the travel demand prediction area. However, their prediction performance is rarely directly compared. By experimental tests conducted using the same dataset, a Bayesian network model and a neural network model are compared for the travel mode analysis for the first time in this paper. It is found that the fully Bayesian network model tends to overfit the training set when the network itself is considerable complicated. The TAN structure otherwise has a better generalization performance and can achieve a better and more stable prediction performance, for its prediction accuracy 75.4%±0.63%, compared to the BP neural network model ,which prediction accuracy is 72.2%±3.01%. Experiment and statistical tests demonstrate the superiority of Bayesian networks and we propose using Bayesian networks, especially TAN, instead of neural networks in the travel mode choice prediction field.

Modeling households activity participation decisions in a rule-based system of travel demand

This paper describes an empirical derivation of an activity participation choice model at the household level decisions taking into account the allocated activities and joint activity participation of household heads in discretionary activities. The households that we consider here are two-heads households; each is either a worker or non-worker. Attributes of households, such as, for example the presence of young children, attributes of the work activities and space-time settings are considered as explanatory variables. To deal with this large set of attributes and account for non-linear relationships between the variables, a decision tree induction method - CHAID - is used to derive a decision tree model. We show how the decision tree model can be used as a component in an activity-scheduling model, ALBATROSS, to predict travel demand in an activity-based-micro-simulation system. The model shows a satisfactory performance as indicated by its goodness-of-fit on validation data.

Analysis of Structural Equation Model for the Student Pleasure Travel Market: Motivation, Involvement, Satisfaction, and Destination Loyalty

Greater understanding of the model that plays a role in attracting certain visitor segments thus allows travel businesses to develop marketing strategies. This research analyzes an empirical model including push motivations, pull motivations, cognitive involvement, affective involvement, satisfaction, and loyalty to students' top domestic and international destinations. Results suggest push motivations are a good predictor of pull motivations, which in turn are considered a good predictor of cognitive involvement. Cognitive involvement is a good predictor of both affective involvement and satisfaction. Affective involvement is a good predictor of satisfaction. This study also confirms the significant association of satisfaction with destination loyalty. Hence, the proposed model provides a good quality of theoretical framework to positively predict travel demand toward domestic and international destinations.

Impact of Data Integration on Some Important Travel Behavior Indicators

Travel surveys are one of the most important ways of obtaining the critical information needed for transportation planning and decision making today. Reliable and quality data from household travel surveys also demand large sample sizes. These surveys are notoriously expensive, however, and highly time-consuming, and they are faced with a high response burden and subsequent low response rates. Although methodological and technological survey techniques have become increasingly refined, high unit costs and public resistance will continue to plague future survey efforts. This paper investigates the impact of combining survey data from different sources on some important travel behavior indicators. Given the availability of other types of surveys such as time use surveys, which tend to collect a great deal of important data in regard to people's travel, a wealth of information can be obtained through combining these data with the available travel survey data. Initially, survey data were weighted on the basis of census data on some important socio-demographic characteristics that had been shown to have an impact on travel. Hereby, the fact that census data had the required sample size but generally did not have the required information on travel was exploited. The Flemish travel survey data were then combined with the Flemish time use survey data. The resultant combined data set offered a larger and more representative sample of the population, which gave more reliable travel information on the population. The larger sample was valuable in the prediction of travel demand and can also be used as a base for simulating travel data.

Investigation and analysis of evidence of asymmetric churn in travel demand models

There is a large amount of research work that has been devoted to the understanding of travel behaviour and for the prediction of travel demand and its management. Different types of data including stated preference and revealed preference, as well as different modelling approaches have been used to predict this. Essential to most travel demand forecasting models are the concepts of utility maximisation and equilibrium, although there have been alternative approaches for modelling travel behaviour. In this paper, the concept of asymmetric churn is discussed. That is travel behaviour should be considered as a two way process which changes over time. For example over time some travellers change their mode of travel from car to bus, but more travellers change their mode from bus to car. These changes are not equal and result in a net change in aggregate travel behaviour. Transport planners often aim at producing this effect in the opposite direction. It is important therefore to recognise the existence of churns in travel behaviour and to attempt to develop appropriate policies to target different groups of travellers with the relevant transport policies in order to improve the transport system. A data set collected from a recent large survey, which was carried out in Edinburgh is investigated to analyse the variations in departure time choice behaviour. The paper reports on the results of the investigation.

Before and After the Inclusion of Intervention Events: An Evaluation of Alternative Forecasting Methods for Tourist Flows

This article investigates the intervention impacts on tourist flows and evaluates the accuracy of various forecasting techniques to predict travel demand before and after the inclusion of intervention events. The forecasting methods used in this study include (1) Naïve 1, (2) Naïve 2, (3) Holt-Winter's model, (4) Seasonal Autoregressive Integrated Moving Average (SARIMA) model, and (5) Artificial Neural Networks (ANN). The Holt Winter's and Naïve models are included for comparison purposes to ensure that minimum performance standards are being met. Data on air transport passengers including international arrivals and domestic air transport flows of the US (from January 1990 to June 2003) were obtained from the US Bureau of Transportation Statistics. This study focuses firstly on the importance for forecasting accuracy of allowing for intervention events in the modeling process. SARIMA models are therefore estimated both with and without intervention effects (the September 11th events). These models are used to generate forecasts for 2002 and the first part of 2003, and forecast accuracy is assessed using mean absolute percentage error and root mean square percentage error. The second focus of the study is to examine the impacts on tourism demand of the major crises that occurred during the period 2001–2003.

An Integrated GIS Approach to Accessibility Analysis

AbstractAccessibility is an important element in evaluating existing land‐use patterns and transportation services, predicting travel demands and programming transportation investments in urban transportation planning. This paper presents an integrated GIS approach to accessibility analysis, which provides a general framework for integrated use of GIS, travel impedance measurement tools and accessibility measures to support the accessibility analysis process, including formulating the concept of accessibility, selecting or developing accessibility measures, specifying the accessibility measures, deriving the accessibility values using the selected or developed accessibility measures, and presenting and interpreting the accessibility values. An integrated GIS tool, ACCESS, designed for implementing the integrated approach is also described and illustrated.

TRIP DISTRIBUTION MODELLING USING FUZZY LOGIC AND A GENETIC ALGORITHM

This article examines possibilities for the application of soft computing techniques for the prediction of travel demand. The model, based on fuzzy logic and a genetic algorithm, successfully solves the trip distribution problem. The possibilities of using the proposed model in solving trip generation, modal split and route choice problems have also been indicated. The model has been tested on a real numerical example. Exceptionally good correspondences between estimated and real values of passenger flows have been obtained.

Combined activity/travel choice models : time-dependent and dynamic versions

To fully understand and predict travel demand and traffic flow, it is necessary to investigate what drives people to travel. The analysis should examine why, where and when various activities are engaged in, and how activity engagement is related to the spatial and institutional organization of an urban area. In view of this, two combined activity/travel choice models are presented in this paper. The first one is a time-dependent (quasi-dynamic) model for long-term transport planning such as travel demand forecasting, while the other one is a dynamic model for short-term traffic management such as instantaneous flow analysis. The time-dependent model is formulated as a mathematical programming problem for modeling the multinomial logit activity/destination choice and the user equilibrium route choice behavior. It can further be converted to a variational inequality problem. On the other hand, the dynamic model is aimed to find a solution for equilibrium activity location, travel route and departure time choices in queuing networks with multiple commuter classes. It is formulated as a discrete-time, finite-dimensional variational inequality and then converted to an equivalent “zero-extreme value” minimization problem. Solution algorithms are proposed for these two models and numerical example is presented for the latter. It is shown that the proposed modeling approaches, either based on time-dependent or dynamic traffic assignment principles, provide powerful tools to a wide variety of activity/travel choice problems in dynamic domain.

A comparison of three different approaches to tourist arrival forecasting

Forecasting plays a major role in tourism planning. The promotion of tourism projects involving substantial sums of money requires an estimate of future demand and market penetration. The commitment to developing tourism would be much easier if it were possible to analyse current and past tourist traffic and predict the nature of changes in tourism demand. These extrapolative approaches to forecasting require historical data. This paper investigates the application of three time-series forecasting techniques, namely exponential smoothing, univariate ARIMA, and Elman's Model of Artificial Neural Networks (ANN), to predict travel demand (i.e. the number of arrivals) from different countries to Hong Kong. Exponential smoothing and ARIMA are two commonly used statistical time series forecasting techniques. The third approach, Neural Networks, is an artificial intelligence technique derived from computer science. According to the analysis presented in this paper, Neural Networks seems to be the best method for forecasting visitor arrivals, especially those series without obvious pattern.

Tourism Forecasting and its Relationship with Leading Economic Indicators

This article investigates the application of three time-series forecasting techniques, namely, exponential smoothing, univariate Autoregressive Integrated Moving Average (ARIMA), and adjusted ARIMA to predict travel demand (i.e., the number of arrivals) from different countries to Hong Kong. The third approach, adjusted ARIMA, is an enhancement of univariate ARIMA. This uses influential economic indicators that are highly correlated with travel demand to adjust univariate ARIMA. According to the analysis, adjusted ARIMA with economic indicators seems to be the best forecasting method for Japan, whereas univariate ARIMA is the best predictor for the United States and United Kingdom. Univariate ARIMA and adjusted ARIMA behave similarly for countries like Taiwan, Singapore, and Korea. Among the three forecasting methods, exponential smoothing is the least accurate. This shows that univariate ARIMA and adjusted ARIMA are more suitable and can be applied to forecast the fluctuating series of visitor arrivals.

Use of Fuzzy Inference for Modeling Prediction of Transit Ridership at Individual Stops

Fuzzy inference has become a popular approach to modeling systems in which uncertainties associated with human perception and decision are present. Its use for the modeling framework of travel demand prediction is promising. The possibility of applying fuzzy inference to the problem of predicting bus ridership at individual stops is examined. The required properties for such a model are identified, and the mathematics of fuzzy inference is examined. The factors that may cause transit use at individual stops are identified, and their relationships to ridership are modeled using the hierarchically structured fuzzy rule basis. This fuzzy rule–based model is similar to that of the cross-classification approach, with the boundaries of the classes being fuzzy. The artificial neural network and regression methods are used to model the same problem, and the results are compared with those of the fuzzy inference method. The data used for calibration are obtained from a study of actual bus stops in Delaware. Although no conclusion is drawn as to which approach is superior, fuzzy inference provides an alternative to the traditional regression approach in which the phenomenon’s causality is complicated yet the behavior of the output must be consistent with commonsense knowledge. In particular, it is suited for problems in which the actual volume to be predicted fluctuates widely. Such problems require strength of logical explanatory power and understanding of the pattern rather than accuracy in fitting the data.

Formulating a methodology for modelling revealed preference discrete choice data—the selectively replicated logit estimation

There are a number of studies on modelling with Revealed Preference (RP) data. It is a traditional technique and it is based on actual market data. The method has been extensively used in transportation as a tool for predicting travel demand. Although the method constitutes a relevant analysis on the process of modelling, it suffers from limitations, mainly associated with the lack of control over the experiment, that sometimes overwhelm the model results. This work proposes and tests a methodology for estimating a more efficient binary RP sample set. The objective is to develop and test a methodology that identifies and eliminates potentially irrational choices made. Responses are evaluated according to the set of trade-offs in values of time. Having identified these individuals they are eliminated from the original sample and a new sample is created, the selectively replicated (SR) sample. Original and SR samples are then re-estimated in a tree nested logit structure.