A mixed integer knapsack model for allocating funds to highway safety improvements

This paper presents a methodology for allocating funds to highway safety improvements. Besides the commonly used binary variables that represent discrete interventions at specific points of a highway, continuous variables are introduced to represent the lengths of a highway over which continuous improvements, such as pavement resurfacing or lighting, are implemented. The problem is formulated as a mixed integer knapsack model with linear multiple choice constraints. Some insight into its solution properties is provided and an efficient branch and bound algorithm is proposed for its solution. A case study that illustrates the application of the model is also presented.     

A practical model for transfer optimization in a transit network: Model formulations and solutions

This paper studies the transit network scheduling problem and aims to minimize the waiting time at transfer stations. First, the problem is formulated as a mixed integer programming model that gives the departure times of vehicles in lines so that passengers can transfer between lines at transfer stations with minimum waiting times. Then, the model is expanded to a second model by considering the extra stopping time of vehicles at transfer stations as a new variable set. By calculating the optimal values for these variables, transfers can be better performed. The sizes of the models, compared with the existing models, are small enough that the models can be solved for small- and medium-sized networks using regular MIP solvers, such as CPLEX. Moreover, a genetic algorithm approach is represented to more easily solve larger networks. A simple network is used to describe the models, and a medium-sized, real-life network is used to compare the proposed models with another existing model in the literature. The results demonstrate significant improvement. Finally, a large-scale, real-life network is used as a case study to evaluate the proposed models and the genetic algorithm approach.     

Latent variables and route choice behavior

In the last decade, a broad array of disciplines has shown a general interest in enhancing discrete choice models by considering the incorporation of psychological factors affecting decision making. This paper provides insight into the comprehension of the determinants of route choice behavior by proposing and estimating a hybrid model that integrates latent variable and route choice models. Data contain information about latent variable indicators and chosen routes of travelers driving regularly from home to work in an urban network. Choice sets include alternative routes generated with a branch and bound algorithm. A hybrid model consists of measurement equations, which relate latent variables to measurement indicators and utilities to choice indicators, and structural equations, which link travelers’ observable characteristics to latent variables and explanatory variables to utilities. Estimation results illustrate that considering latent variables (i.e., memory, habit, familiarity, spatial ability, time saving skills) alongside traditional variables (e.g., travel time, distance, congestion level) enriches the comprehension of route choice behavior.     

An aggregate demand model for intercity passenger travel in Sri Lanka

Inspite of the inherent weaknesses in aggregate demand models, they continue to be used in everyday applications, especially in developing countries. The largely data intensive disaggregate model preclude its application in many cases. This paper attempts the formulation and calibration of an aggregate total demand model for estimating inter-district passenger travel by public transport in Sri Lanka. In its process, an investigation is made of the common problems in the aggregate approach while examining possible remedial measures to improve the accuracy and (hence) the usability of the aggregate model. It is argued that commonly used variables and functional forms are inappropriate for making accurate estimates in developing countries. Consequently, the model calibration is shown to incorporate variables representing urbanisation, under-development, transfers, a mode-abstract cost function and intrinsic features. The necessity for functional form for each variable to be based on behavioral assumptions that are tested using the Box-Cox transformation for ensuring the best fit of the data is also observed. Although, the model form was calibrated for Sri Lanka, the model is generalised in order for its applications to other countries as well as, both, inter-district and intercity travel demand estimation.     

Regimes in social–cultural events-driven activity sequences: Modelling approach and empirical application

In this study we propose and apply a Bayesian-network model to predict and analyse the factors that influence activity-travel sequences that are triggered by social–cultural events. The study is motivated by the intention to examine the wider context in which activity-travel decisions are made and to model such decisions under longitudinal time horizons. We assume that social events trigger a series of interrelated activities and corresponding trips. Data about events and related activities are collected using a month-diary and involving a large sample of households in the Eindhoven region, The Netherlands. A learning algorithm is applied to derive a Bayesian-network model from the event diary. The results show that indeed many travel choices are influenced by particular events, that these influences vary by socio-demographic variables and that the learned Bayesian-network model is able to represent these interdependencies among all these variables. We demonstrate how the model can be used to predict event-driven activity-travel sequences in a micro-simulation.     

A latent variable representation of count data models to accommodate spatial and temporal dependence: Application to predicting crash frequency at intersections

This paper proposes a reformulation of count models as a special case of generalized ordered-response models in which a single latent continuous variable is partitioned into mutually exclusive intervals. Using this equivalent latent variable-based generalized ordered response framework for count data models, we are then able to gainfully and efficiently introduce temporal and spatial dependencies through the latent continuous variables. Our formulation also allows handling excess zeros in correlated count data, a phenomenon that is commonly found in practice. A composite marginal likelihood inference approach is used to estimate model parameters. The modeling framework is applied to predict crash frequency at urban intersections in Arlington, Texas. The sample is drawn from the Texas Department of Transportation (TxDOT) crash incident files between 2003 and 2009, resulting in 1190 intersection-year observations. The results reveal the presence of intersection-specific time-invariant unobserved components influencing crash propensity and a spatial lag structure to characterize spatial dependence. Roadway configuration, approach roadway functional types, traffic control type, total daily entering traffic volumes and the split of volumes between approaches are all important variables in determining crash frequency at intersections.     

A statistical model for estimating oxides of nitrogen emissions from light duty motor vehicles

Motor vehicle emission rate models for predicting oxides of nitrogen (NOx) emissions are insensitive to vehicle modes of operation such as cruise, acceleration, deceleration, and idle, because they are based on average trip speed. Research has shown that NOx emissions are sensitive to engine load; hence, load-based variables need to be included in emissions models. Ongoing studies attempting to incorporate these `modal' variables have experienced difficulties with: (1) incomplete and/or non-representative data sets of emissions test data vis-a-vis the modal operating profiles of the tested vehicles; (2) lack of information for predicting on-road operating parameters of vehicles; and (3) non-representative vehicles recruited for emissions tests.The objective of this research was to develop a statistical model for predicting NOx emissions from light-duty gasoline motor vehicles. The primary end use of this model is forecasting, rather than explanation of the factors that affect NOx emissions, which brings to bear different requirements from the statistical model. The three challenges noted above are addressed by: (1) analyzing a data set of more than 13 000 hot-stabilized laboratory treadmill tests on 19 driving cycles (specific speed versus time testing conditions), and 114 variables describing vehicle, engine and test cycle characteristics; (2) making the models compatible with empirical data on how vehicles are being operated in-use; and (3) developing statistical weights to account for the differences in model year distributions between the emissions testing database and the current national on-road fleets.The NOx emissions model is estimated using ordinary least-squares regression techniques, with transformed response variable and regression weights. Tree regression is employed as a tool for mining relationships among variables in the data, with particular focus on identifying useful interactions among discrete variables. Details of the model development process are presented, as well as results for the final model showing the predicted emissions algorithm for the current motor vehicle fleet in Atlanta, GA metropolitan region.     

The location selection problem for the household activity pattern problem

In this paper, an integrated destination choice model based on routing and scheduling considerations of daily activities is proposed. Extending the Household Activity Pattern Problem (HAPP), the Location Selection Problem (LSP–HAPP) demonstrates how location choice is made as a simultaneous decision from interactions both with activities having predetermined locations and those with many candidate locations. A dynamic programming algorithm, developed for PDPTW, is adapted to handle a potentially sizable number of candidate locations. It is shown to be efficient for HAPP and LSP–HAPP applications. The algorithm is extended to keep arrival times as functions for mathematical programming formulations of activity-based travel models that often have time variables in the objective.     

A latent segmentation based multiple discrete continuous extreme value model

We examine an alternative method to incorporate potential presence of population heterogeneity within the Multiple Discrete Continuous Extreme Value (MDCEV) model structure. Towards this end, an endogenous segmentation approach is proposed that allocates decision makers probabilistically to various segments as a function of exogenous variables. Within each endogenously determined segment, a segment specific MDCEV model is estimated. This approach provides insights on the various population segments present while evaluating distinct choice regimes for each of these segments. The segmentation approach addresses two concerns: (1) ensures that the parameters are estimated employing the full sample for each segment while using all the population records for model estimation, and (2) provides valuable insights on how the exogenous variables affect segmentation. An Expectation–Maximization algorithm is proposed to address the challenges of estimating the resulting endogenous segmentation based econometric model. A prediction procedure to employ the estimated latent MDCEV models for forecasting is also developed. The proposed model is estimated using data from 2009 National Household Travel Survey (NHTS) for the New York region. The results of the model estimates and prediction exercises illustrate the benefits of employing an endogenous segmentation based MDCEV model. The challenges associated with the estimation of latent MDCEV models are also documented.     

Link-based day-to-day network traffic dynamics and equilibria

A general dynamical system model with link-based variables is formulated to characterize the processes of achieving equilibria from a non-equilibrium state in traffic networks. Several desirable properties of the dynamical system model are established, including the equivalence between its stationary state and user equilibrium, the invariance of its evolutionary trajectories, and the uniqueness and stability of its stationary points. Moreover, it is shown that not only a link-based version of two existing day-to-day traffic dynamics models but also two existing link-based dynamical system models of traffic flow are the special cases of the proposed model. The stabilities of stationary states of these special cases are also analyzed and discussed. In addition, an extension is made to the case with elastic demand. The study is helpful for better understanding the day-to-day adjustment mechanism of traffic flows in networks.     

Modeling interdependencies between vehicle transaction,residential relocation and job change

This paper introduces a vehicle transaction timing model which is conditional on household residential and job relocation timings. Further, the household residential location and members’ job relocation timing decisions are jointly estimated. Some researchers have modeled the household vehicle ownership decision jointly with other household decisions like vehicle type choice or VMT; however, these models were basically static and changes in household taste over time has been ignored in nearly all of these models. The proposed model is a dynamic joint model in which the effects of land-use, economy and disaggregate travel activity attributes on the major household decisions; residential location and members’ job relocation timing decisions for wife and husband of the household, are estimated. Each of these models is estimated using both the Weibull and log-logistic baseline hazard functions to assess the usefulness of a non-monotonic rather than monotonic baseline hazard function. The last three waves of the Puget Sound Panel Survey data and land-use, transportation, and built environment variables from the Seattle Metropolitan Area are used in this study as these waves include useful explanatory variables like household tenure that were not included in the previous waves.     

Household trip generation choice—Alternative empirical approaches

Trip generation models have generally received less attention than other aspects of travel decision making. This article presents some explorations into the structure of trip decisions for shopping, using data from weekly shopping diaries. The paper compares alternative formulations of the naturally-ordered choice model used by Sheffi (1979) to avoid the problems inherent in multinomial logit models. Firstly, imposing cross-alternative restrictions on some of the coefficient values is termed the constrained model. Secondly, when no such restrictions are imposed, the model decomposes into a sequence of binary models, and this is termed the unconstrained model, which can be used to test the validity of the restrictions. The variables used include both shopping expenditures and locational factors, both of which are found to play a key role in shopping travel decisions, as well as more conventional socio-economic variables. A clearer understanding of the role of car ownership in travel decisions is obtained.     

Do microscopic merging models reproduce the observed priority sharing ratio in congestion?

A classical way to represent vehicle interactions at merges at the microscopic scale is to combine a gap-acceptance model with a car-following algorithm. However, in congested conditions (when a queue spills back on the major road), outputs of such a combination may be irrelevant if anticipatory aspects of vehicle behaviours are disregarded (like in single-level gap-acceptance models). Indeed, the insertion decision outcomes are so closely bound to the car-following algorithm that irrelevant results are produced. On the one hand, the insertion decision choice is sensitive to numerical errors due to the car-following algorithm. On the other hand, the priority sharing process observed in congestion cannot be correctly reproduced because of the constraints imposed by the car-following on the gap-acceptance model. To get over these issues, more sophisticated gap-acceptance algorithms accounting for cooperation and aggressiveness amongst drivers have been recently developed (multi-level gap-acceptance models). Another simpler solution, with fewer parameters, is investigated in this paper. It consists in introducing a relaxation procedure within the car-following rules and proposing a new insertion decision algorithm in order to loosen the links between both model components. This approach will be shown to accurately model the observed flow allocation pattern in congested conditions at an aggregate scale.     

The use of Bayesian statistics to predict patterns of spatial repeatability

Statistical spatial repeatability (SSR) is an extension to the well known concept of spatial repeatability. SSR states that the mean of many patterns of dynamic tyre force applied to a pavement surface is similar for a fleet of trucks of a given type. A model which can accurately predict patterns of SSR could subsequently be used in whole-life pavement deterioration models as a means of describing pavement loading due to a fleet of vehicles. This paper presents a method for predicting patterns of SSR, through the use of a truck fleet model inferred from measurements of dynamic tyre forces. A Bayesian statistical inference algorithm is used to determine the distributions of multiple parameters of a fleet of quarter-car heavy vehicle ride models, based on prior assumed distributions and the set of observed dynamic tyre force from a ‘true’ fleet of one hundred simulated models. Simulated forces are noted at 16 equidistant pavement locations, similar to data from a multiple sensor weigh-in-motion site. It is shown that the fitted model provides excellent agreement in the mean pattern of dynamic force with the originally generated truck fleet. It is shown that good predictions are possible for patterns of SSR on a given section of road for a fleet of similar vehicles. The sensitivity of the model to errors in parameter estimation is discussed, as is the potential for implementation of the method.     

Urban travel demand: The impact of Box-Cox transformations with nonspherical residual errors

We develop an algorithm for solving regression models with Box-Cox transformations on both the dependent and independent variables, while simultaneously taking into account corrections for serial correlation of several orders and for heteroscedasticity. The latter correction is of a general form which contains as special cases most specifications of heteroscedasticity found in practice. We apply the procedure to three urban travel demand functions, two of which are currently used in their linear form by the Montreal Transit Authority, and analyze more than 100 specifications. Our results show that taking into account nonsphericalness of the residuals has a major impact on model parameter estimates, notably on those which determine the functional form of the model, and that, conversely, modifications of the functional form have strong implications for both the structure of autocorrelation and the importance of heteroscedasticity; moreover, we find interactions between autocorrelation and heteroscedasticity structures. We introduce a special measure of elasticity for variables which contain zero observations, particularly dummy variables. Moreover, we find that elasticities of demand and implicit values of time depend to a large extent on the stochastic specification of the model.     

Geographical unit based analysis in the context of transportation safety planning

A wide array of spatial units has been explored in macro-level modeling. With the advancement of Geographic Information System (GIS) analysts are able to analyze crashes for various geographical units. However, a clear guideline on which geographic entity should be chosen is not present. Macro level safety analysis is at the core of transportation safety planning (TSP) which in turn is a key in many aspects of policy and decision making of safety investments. The preference of spatial unit can vary with the dependent variable of the model. Or, for a specific dependent variable, models may be invariant to multiple spatial units by producing a similar goodness-of-fits. In this study three different crash models were investigated for traffic analysis zones (TAZs), block groups (BGs) and census tracts (CTs) of two counties in Florida. The models were developed for the total crashes, severe crashes and pedestrian crashes in this region. The primary objective of the study was to explore and investigate the effect of zonal variation (scale and zoning) on these specific types of crash models. These models were developed based on various roadway characteristics and census variables (e.g., land use, socio-economic, etc.).It was found that the significance of explanatory variables is not consistent among models based on different zoning systems. Although the difference in variable significance across geographic units was found, the results also show that the sign of the coefficients are reasonable and explainable in all models.Key findings of this study are, first, signs of coefficients are consistent if these variables are significant in models with same response variables, even if geographic units are different. Second, the number of significant variables is affected by response variables and also geographic units.Admittedly, TAZs are now the only traffic related zone system, thus TAZs are being widely used by transportation planners and frequently utilized in research related to macroscopic crash analysis. Nevertheless, considering that TAZs are not delineated for traffic crash analysis but they were designed for the long range transportation plans, TAZs might not be the optimal zone system for traffic crash modeling at the macroscopic level. Therefore, it recommended that other zone systems be explored for crash analysis as well.     

Estimation of bus dwell time using univariate time series models

A significant proportion of bus travel time is contributed by dwell time for passenger boarding and alighting. More accurate estimation of bus dwell time (BDT) can enhance efficiency and reliability of public transportation system. Regression and probabilistic models are commonly used in literatures where a set of independent variables are used to define the statistical relationship between BDT and its contributing factors. However, due to technical and monetary constraints, it is not always feasible to collect all the data required for the models to work. More importantly, the contributing factors may vary from one bus route to another. Time series based methods can be of great interest as they require only historical time series data, which can be collected using a facility known as automatic vehicle location (AVL) system. This paper assesses four different time series based methods namely random walk, exponential smoothing, moving average (MA), and autoregressive integrated moving average to model and estimate BDT based on AVL data collected from Auckland. The performances of the proposed methods are ranked based on three important factors namely prediction accuracy, simplicity, and robustness. The models showed promising results and performed differently for central business district (CBD) and non‐CBD bus stops. For CBD bus stops, MA model performed the best, whereas for non‐CBD bus stops, ARIMA model performed the best compared with other time series based models. Copyright © 2014 John Wiley & Sons, Ltd.     

Accessibility and transport infrastructure improvement assessment: The role of borders and multilateral resistance

The market potential indicator is a commonly used tool in transport planning for evaluating the potential economic effects derived from improvements in transport infrastructures. The general assumption is that exports from a given region will rise with increased accessibility, thus benefiting economic activities. However, the specification of the market potential model is typically very simple and ignores both the impact of competing rivals and the role of international borders, which leads to unrealistic results. Spatial interaction models on bilateral trade have already proved that international trade is affected by multilateral resistance, borders, adjacency, language or currency. Nevertheless, apart from some recent analyses that simply calibrate the distance decay parameter from trade datasets, these variables have hardly been integrated into research on market potential. This paper sets out to demonstrate that more realistic results are obtained by calibrating the distance-decay parameter and introducing the impact of competing rivals and border effects into the market potential formulation. The proposed model is then applied to the assessment of the accessibility impacts of new road transport infrastructure in the European Union between 2001 and 2012, which shows that the greatest improvements in accessibility were experienced by peripheral countries with high road infrastructure investment.     

Network-based real option models

Building on earlier work to incorporate real option methodologies into network modeling, two models are proposed. The first is the network option design problem, which maximizes the expanded net present value of a network investment as a function of network design variables with the option to defer the committed design investment. The problem is shown to be a generalized version of the network design problem and the multi-period network design problem. A heuristic based on radial basis functions is used to solve the problem for continuous link expansion with congestion effects. The second model is a link investment deferral option set, which decomposes the network investment deferral option into individual, interacting link or project investments. This model is a project selection problem under uncertainty, where each link or project can be deferred such that the expanded net present value is maximized. The option is defined in such a way that a lower bound can be solved using an exact method based on multi-option least squares Monte Carlo simulation. Numerical tests are conducted with the classical Sioux Falls network and compared to earlier published results.     

Park-and-ride lot choice model using random utility maximization and random regret minimization

This research aims to understand the park-and-ride (PNR) lot choice behaviour of users i.e., why PNR user choose one PNR lot versus another. Multinomial logit models are developed, the first based on the random utility maximization (RUM) concept where users are assumed to choose alternatives that have maximum utility, and the second based on the random regret minimization (RRM) concept where users are assumed to make decisions such that they minimize the regret in comparison to other foregone alternatives. A PNR trip is completed in two networks, the auto network and the transit network. The travel time of users for both the auto network and the transit network are used to create variables in the model. For the auto network, travel time is obtained using information from the strategic transport network using EMME/4 software, whereas travel time for the transit network is calculated using Google’s general transit feed specification data using a backward time-dependent shortest path algorithm. The involvement of two different networks in a PNR trip causes a trade-off relation within the PNR lot choice mechanism, and it is anticipated that an RRM model that captures this compromise effect may outperform typical RUM models. We use two forms of RRM models; the classical RRM and µRRM. Our results not only confirm a decade-old understanding that the RRM model may be an alternative concept to model transport choices, but also strengthen this understanding by exploring differences between two models in terms of model fit and out-of-sample predictive abilities. Further, our work is one of the few that estimates an RRM model on revealed preference data.