The importance of traffic planning has, throughout the years, been in- creased, providing sustainable developments of traffic and infrastructural investments. The analysis of the current traffic situation and the evalua- tion of the effects of a future investment are crucial for the socio-economic benefits maintenance. These analyses and evaluations are most commonly done using traffic simulation models. One of the main traffic planning aims, nowadays, is to increase the number of public transport users against the number of private car users. This change in mode choice is called transition effect and could be beneficial both from an environmental and socio-economic perspective. This thesis aims to evaluate and improve the macroscopic traffic demand and transition model, used fot the city of Norrköping. Additionally, the thesis investigates if a general transition Logit model can be developed and which parameters are the most important to be included in a modal choice estimation. For the needs of this study, the traffic planning software Visum is used. The travel mode distribution is calculated by Logit models coded in Python-scripts integrated in Visum. Then, a traffic assignment is performed by Visum, computing new travel times as inputs to the Logit model and this iterative procedure continues until the system reaches an equilibrium. The thesis aims for a more reliable prediction of the transition effect by correcting the Python-scripts and estimating the parameters of the Logit model using data from surveys. The study shows that travel times is the most important factor for realistic results generation. However, the data used for the estimation of the Logit model parameters did not include sufficient information of travel times. The travel times had to be calculated, using two different methods, in order to be included in the estimation of new parameters. Although these methods could not provide any positive effects on the transition, they did prove the importance and significance travel time have when developing a traffic model. The result of the study invokes the importance to further develop the method of calculating travel times, when the input data is not sufficient, and shows that the travel time parameters are case specific.

The multinomial logit model (MNL) possesses a latent variable representation in terms of random variables following a multivariate logistic distribution. Based on multivariate finite mixture approximations of the multivariate logistic distribution, various data-augmented Metropolis-Hastings algorithms are developed for a Bayesian inference of the MNL model.

Autonomous Vehicles (AVs) are computer equipped vehicles that can operate without human driver’s active control using information provided by their sensors about the surrounding environment. Self-driving vehicles may have seemed to be a distant dream several years ago, but manufactures’ prototypes showed that AVs are becoming real now. Several car manufactures (i.e. Benz, Audi, etc.) and information technology firms (i.e. Google) have either showcased their fully AVs or announced their robot cars to be released in a few years. AVs hold the promise to transform the ways we live and travel. Although several studies have been conducted on the impacts of AVs, much remains to be explored regarding the various ways in which AVs could reshape our lifestyle. This dissertation addresses the knowledge gap in understanding the potential implications of AV technologies on travel behavior and system modeling. A comprehensive review of literature regarding AV adoption, potential impacts and system modeling was provided. Bass diffusion models were developed to investigate the market penetration process of AVs based on experience learned from past technologies. A stated preference survey was conducted to gather information from university population on the perceptions and attitudes toward AV technologies. The data collected from the Florida International University (FIU) was used to develop econometric models exploring the willingness to pay and relocation choices of travelers in light of the new technologies. In addition, the latest version of the Southeast Planning Regional Model (SERPM) 7.0, an Activity-Based Model (ABM), was employed to examine the potential impacts of AVs on the transportation network. Three scenarios were developed for short-term (2035), mid-term (2045) and long-term (2055) conditions. This dissertation provides a systematic approach to understand the potential implications of AV technologies on travel behavior and system modeling. The results of the survey data analysis and the scenario analysis also provide important inputs to guide planning and policy analysis on the impacts of AV technologies.

In this thesis, we develop a SKU level market basket model and apply the model to investigate cross-category reference price effects. This research extends previous work on the category-level multivariate logit model (Russell and Petersen 2000). Our model is a generalization of the multivariate logit model which allows for both complementarity and substitution effects at the brand level. The modeling effort in this thesis allows us to use conditional probability distributions of individual items to construct the final joint-distribution of all possible basket selections. The resulting model is very flexible and accommodates a large variety of market structure patterns. The model structure implies that the changes in brand-level marketing variables directly affect category incidence (by altering category attractiveness) and indirectly determine market basket composition. Because the model can be written in a closed form manner, we can easily study the pattern of brand price competition by computing a matrix of cross-price elasticities. We use scanner panel data for the yogurt category to demonstrate the structural flexibility of the model. The results from this application reveal asymmetric competition consistent with price-tier competition literature. We use this model to investigate how consumers' responses to reference prices within a category spillover into their choices across multiple categories. The notion is that a consumer's subjective judgment of the fairness of the price levels in one category influences the choice decisions of related items in other categories. We begin with building within-category SKU-level model based on previous findings from single category reference price models (i.e., internal versus external reference prices, asymmetric response due to loss aversion, and heterogeneity in response across consumers). We then develop four alternative model specifications for cross-category spillover effects and test competing theories about those effects. Using scanner panel data for detergent and softener categories, we discover valuable implications for reference price effects. First, SKU-level reference price effects exist and improve forecasting ability. Second, those reference price effects influence category attractiveness, but do no spillover across categories. Finally, category-level reference dependent evaluation may exist but not be important in forecasting.

Customer overall satisfaction towards a public transport system depends mainly on two factors: how satisfied they are with different aspects that make up the service and how important each of the service aspects is to the customer. Traditionally, researchers use revealed preference surveys and ordered probit/logit models to estimate the contribution of each service attribute towards the overall satisfaction. This paper aims to verify the possibility of replacing the traditional method with the more cost-effective best-worst case 1 method, using a customer survey recently conducted in Santander, Spain. The results show that the satisfaction level obtained from these alternative methods are remarkably similar. The relative importance of each attribute delivered by the two methods differ, with the Best-Worst approach showing more intuitive and consistent results with the literature on public transport customer satisfaction. A regression method is developed to derive customer satisfaction with each service attribute from Best-Worst modelling results.

This thesis proposes a planning framework for Autonomous Electric Vehicle (AEV) charging. The framework is intended to help transportation decision-makers determine Electric Vehicle (EV) charging facility locations and capacities for the mixed fleet of Autonomous Vehicle (AV) and Human-driven Vehicle (HDV). The bi-level nature of the framework captures the decision-making processes of the transportation agency decision-makers and travelers, thereby providing solid theoretical and practical foundations for the EV charging network design. At the upper level, the decision-makers seek to determine the locations and operating capacities of the EV charging facilities, in a manner that minimizes total travel time and construction costs subject to budgetary limitations. In addition, the transportation decision-makers provide AV-exclusive lanes to encourage AV users to reduce travel time, particularly at wireless-charging lanes, as well as other reasons, including safety. At the lower level, the travelers seek to minimize their travel time by selecting their preferred vehicle type (AV vs. HDV) and route. In measuring the users delay costs, the thesis considered network user equilibrium because the framework is designed for urban networks where travelers route choice affects their travel time. The bi-level model is solved using the Non-Dominated Sorting Genetic Algorithm (NSGA-II) algorithm.

In a lot of big cities, the traffic network is overloaded, with congestion and unnecessary emissions as consequence. Therefore, different traffic control methods are useful, especially in case of an incident. One key problem for traffic control is traffic prediction and the aim of this thesis is to develop, calibrate and evaluate a route flow model using only observed travel times and travel demand as input. The route flow model was used to calculate the metric link flow destination distribution, that presents to which destinations the travelers on a link are going in percentage.

Under the Basel accords for measuring regulatory capital requirements, the set of credit risk parameters probability of default (PD), exposure at default (EAD) and loss given default (LGD) are measured with own estimates by the internal rating based approach. The estimated parameters are also the foundation of understanding the actual risk in a banks credit portfolio. The predictive performance of such models are therefore interesting to examine. The credit risk parameter LGD has been seen to give low performance for predictive models and LGD values are generally hard to estimate. The main purpose of this thesis is to analyse the predictive performance of a multi-gene genetic programming approach to symbolic regression compared to three benchmark regression models. The goal of multi-gene symbolic regression is to estimate the underlying relationship in the data through a linear combination of a set of generated mathematical expressions. The benchmark models are Logit Transformed Regression, Beta Regression and Regression Tree. All benchmark models are frequently used in the area. The data used to compare the models is a set of randomly selected, de-identified loans from the portfolios of underlying U.S. residential mortgage-backed securities retrieved from International Finance Research. The conclusion from implementing and comparing the models is that, the credit risk parameter LGD is continued difficult to estimated, the symbolic regression approach did not yield a better predictive ability than the benchmark models and it did not seem to find the underlying relationship in the data. The benchmark models are more user-friendly with easier implementation and they all requires less calculation complexity than symbolic regression.

<p>Wildlife species generate value through their consumptive and non-consumptive uses. Consumptive uses of these species include hunting and trapping, while wildlife watching is an example of a non-consumptive use. Understanding the value of various wildlife is imperative for public agencies’ management decisions regarding different wildlife areas (e.g., nature preserves, state forests/parks, reservoirs, county/city parks). Individuals’ values for wildlife interactions on public lands can depend on the context in which these interactions occur as well as the probability of an interaction occurring. We utilize a stated preference choice experiment to estimate Indiana residents’ willingness to pay (WTP) for a marginal increase in the chance of seeing white-tailed deer and five furbearing species (bobcat, coyote, river otter, raccoon, red fox) while engaging in their favorite activities at Indiana recreational areas. Our WTP estimates are analogous to value of statistical life (VSL) calculations, and hence we refer to them as the “value of a statistical species” (VSS). We find that the VSS of a bobcat ranges from $22.73 to $41.30, the VSS of a coyote ranges from -$1.94 to $9.74, the VSS of a raccoon ranges from $5.25 to $21.69, the VSS of a red fox ranges from $43.31 to $62.52, the VSS of a white-tailed deer ranges from $22.70 to $27.00, and the VSS of a river otter ranges from $23.18 to $45.98. Our analysis suggests that individuals’ values for wildlife depend on the activity they are undertaking when they see the wildlife.  </p>

<p>This research explores the impact of road infrastructure on deforestation, the role of forest governance and a carbon tax/credit mechanism in mitigating the effect on land use change and subsequent GHG emissions, with application to the states of Pará and Mato Grosso in Brazil. Few studies have addressed how policies to protect forested land affect the rate of deforestation associated with road and infrastructure improvement. This research makes three main contributions to the literature of roads and deforestation: 1) the concept of cost of access to the “closest” market in terms of time (expressed in person hours per ten ton load) is introduced to reflect variations in the road network infrastructure; 2) development of empirical evidence of the role of forest governance in diminishing the rate of deforestation linked to roads, using data from Brazil; and 3) and assessment of the efficacy of a carbon tax/credit scheme for mitigating the impact of infrastructure investment on land use and resultant changes in GHG emissions. Access cost ranged between 0.01 and 3084 person hours per load, however 80 percent of the pixels measured less than 784 person hours across the three years analyzed (2003, 2013, and 2018). This measure facilitated a contrast in spatial accessibility due to road infrastructure across pixels within the same year and across years on a same pixel. The use of a fractional logit model allowed the incorporation of proportions of different land uses within a same pixel at the same resolution of other <a></a>variables not available at the same fine scale. Strong forest governance reduced up to 25% the elasticities on forest lands with respect to access cost; in other words, the impact of roads on deforestation is reduced by one fourth when forest governance is strengthened. These larger impacts occur at the frontier where most of the efforts need to be addressed. Finally, provided a shock in road infrastructure, a carbon tax/credit level of $82/tCO2e permitted to abate an additional amount of GHG emissions estimated in 244 million tons of CO2e released due to changes in carbon stocks and flow emissions from agricultural activities induced from changes in road infrastructure. More importantly, this research provided insights of a proportion of GHG emissions that could be abated at different levels of a carbon tax/credit.</p>