For some decades now, transport researchers have put considerable efforts into developing what is called activity- based approaches for modelling urban travel demand (for overviews see Arentze and Timmermans, 2000; Ettema and Timmermans, 1997; Jovicic, 2001; McNally, 2000). The basic idea is that travel demand is derived from people's desires to take part in different activities. In particular, the interrelationships among different activities with respect to temporal and spatial constraints are in focus. It means that an activity-based model typically, or ideally, attempts to derive the travel demand from the desire or need to perform activities, that it focuses on households and interactions among them rather than on individuals, that it handles interdependences among activities across the day, and that it observes the spatial and temporal constraints that the individuals of the households have to obey. The Stockholm Integrated Model System (Algers et al 1995) - presented also at the 1996 PTRC summer annual meeting - is an early attempt to model household interactions using a tour based approach. The model system, still being used, consists not only of the usual nested logit models for mode, destination and trip frequency choice, but includes also trip chaining and a number of household interactions such as allocation of the car and allocation of trips to individuals. Since then it has become desirable to focus more on modelling activity scheduling (and thus to develop the model in terms of activity combinations) and extensions to departure time choices. Such extensions can be thought of as adding levels to the choice nests, thereby increasing the complexity of the model. The increased complexity can be quite substantial, as in the case of the Portland model (Bowman et al 1998) where the addition of choices of activity combinations implied 114 alternatives at the highest level (compared to just a few trip purpose trip frequency models), for each of which logsums were carried up from lower levels. Especially in the US, much work has been devoted to finding simplified ways of calculating large choice nests. In this context, the possibility to use micro simulation (Vovsha et al 2004) has turned out to be quite useful. This means that instead of calculating choice probabilities for the whole choice set, simulation techniques can be used to calculate choices, which then condition choices at lower levels. This saves a lot of calculations, allowing for modelling extensions that increase the behavioural realism. In a research project funded by the Swedish research funding agency Vinnova, an opportunity to implement the SIMS model in a micro simulation setting has emerged. The purpose is to form a base for further model enhancements. This paper reports on the micro simulation implementation of the SIMS model, and makes comparisons with the traditional implementation. For the covering abstract please see ITRD E135207.
Abstract