Assessing and predicting car type choices are important for policy analysis. Car type choice models are often based on aggregate alternatives. This is due to the fact that analysts typically do not observe choices at the detailed level that they are made. In this paper, we use registry data of all new car purchases in Sweden for two years where cars are observed by their brand, model and fuel type. However, the choices are made at a more detailed level. Hence, an aggregate (observed) alternative can correspond to several disaggregate (detailed) alternatives. We present an extensive empirical study analyzing estimation results, in-sample and out-of-sample fit as well as prediction performance of five model specifications. These models use different aggregation methods from the literature. We propose a specification of a two-level nested logit model that captures correlation between aggregate and disaggregate alternatives. The nest specific scale parameters are defined as parameterized exponential functions to keep the number of parameters reasonable. The results show that the in-sample and out-of-sample fit as well as the prediction performance differ. The best model accounts for the heterogeneity over disaggregate alternatives as well as the correlation between both disaggregate and aggregate alternatives. It outperforms the commonly used aggregation method of simply including a size measure.

It has been reported that platooning has the potential of saving fuel and increasing traffic throughput. We formulate a platoon coordination problem with soft time windows as a mixed-integer linear programming problem and solve it with exact solutions. The objective function consists of operation costs, schedule miss penalties and fuel costs. In the numerical example, a Swedish highway network model is used and the computation result shows that, for 21 vehicles, the total cost can be reduced by 3.5% when the optimal preferred arrival times are chosen. A random disturbance is then added to the optimal time windows and the optimal result shows great sensitivity with respect to the disturbance. When the mean of the disturbance becomes larger than 10 minutes, more than half of the platooning benefits will be lost. The study also analyzes the change of different cost compositions as disturbance increases.

As a result of public investment, lower freight transport costs tend to translate into lower local price indices and are associated with equilibria characterized by higher output and consumption. In this paper we investigate an additional effect to these trade gains, namely the gains from better spatial matching in the labor market. We simulate a two-region Spatial OLG model in which agents are heterogeneous in terms of skill. Under repeated simulation experiments, we show that, for high household relocation frictions, the possibility of interregional commuting can be seen as an alternative way to realize the potential matching effects. For high levels of skill heterogeneity and a plausible parametric input, a steady state in which labor matching is realized through commuting can be associated with up to 10% higher per capita output, compared to the one with homogenous labor, in which only gains from trade are feasible.

During the last decades, many activity-based models have been developed in the literature. However, especially in random utility based models timing decisions are often treated poorly or inconsistently with other choice dimensions. In this paper we show how dynamic discrete choice can be used to overcome this problem. In the proposed model, trip decisions are made sequentially in time, starting at home in the morning and ending at home in the evening. At each decision stage, the utility of an alternative is the sum of the one-stage utility of the action and the expected future utility in the reached state.

The model generates full daily activity schedules with any number of trips that each is a combination of one of 6 activities, 1240 locations and 4 modes. The ability to go from all to all locations makes evaluating the model very time consuming and sampling of alternatives were therefore used for estimation. The model is estimated on travel diaries and simulation results indicates that it is able to reproduce timing decisions, trip lengths and distribution of the number trips within sample.

To explain when people perform different activities, two sets of parameters are used: firstly, the utility of being at home varies depending on the time of day; and secondly, constants determine the utility of arriving to work at specific times. This was enough to also obtain a good distribution of the starting times for free-time activities.