Globally, the transport system faces a paradigmatic shift where, in addition to increased local traffic problems, climate change and depletion of fossil oil reserves will foster a successive transition to renewable fuels and a need for more resource-efficient mobility management and communication alternatives. Foresighted countries, cities or companies taking the lead in adapting to these tougher conditions might well not only solve those problems, but also turn the problems into business advantages. This thesis is based on six studies that attempt to develop future strategies based on rigorous principled emission and energy efficiency targets and to modulate the impact of travel policies, technical components and behaviours in economically advantageous ways. The modelling frameworks developed throughout the thesis build on a target-orientated approach called backcasting, where the following general components are applied: (1) target description at a conceptual level i.e. the potential for sustainable energy systems, emissions, costs, behavioural patterns, preferences, etc.; (2) mapping of the current situation in relation to target description; and (3) modelling of alternative sets of policies, technologies, behaviours and economic prerequisites to arrive at target achievement. Sustainable travel strategies are analysed from two main viewpoints. The first four studies focus on company travel planning, where behavioural modelling proved to be an important tool for deriving targetorientated travel policies consistent with employee preferences. The latter two studies focus on strategies and preconditions to meet future emission targets and energy efficiency requirements at a macroscopic regional level by 2030. Backcasting’s role as a generic methodology for effective strategic planning is discussed.

Travel demand forecasting is essential for many decisions, such as infrastructure investments and policy measures. Traditionally travel demand modelling has considered trip frequency, mode, destination and route choice. This thesis considers two other choice dimensions, hypothesised to have implications for travel demand forecasting. The first part investigates how the increased possibilities to overcome space that ICT (information and communication technology) provides, can be integrated in travel demand forecasting models. We find that possibilities of modelling substitution effects are limited, irrespective of data source and modelling approach. Telecommuting explains, however, a very small part of variation in work trip frequency. It is therefore not urgent to include effects from telecommuting in travel demand forecasting. The results indicate that telecommuting is a privilege for certain groups of employees, and we therefore expect that negative attitudes from management, job suitability and lack of equipment are important obstacles. We find also that company benefits can be obtained from telecommuting. No evidences that telecommuting gives rise to urban sprawl is, however, found. Hence, there is ground for promoting telecommuting from a societal, individual and company perspective.

The second part develops a departure time choice model in a mixed logit framework. This model explains how travellers trade-off travel time, travel time variability, monetary and scheduling costs, when choosing departure time. We explicitly account for correlation in unobserved heterogeneity over repeated SP choices, which was fundamental for accurate estimation of the substitution pattern. Temporal constraints at destination are found to mainly restrict late arrival. Constraints at origin mainly restrict early departure. Sensitivity to travel time uncertainty depends on trip type and intended arrival time. Given appropriate input data and a calibrated dynamic assignment model, the model can be applied to forecast peak-spreading effects in congested networks. Combined stated preference (SP) and revealed preference (RP) data is used, which has provided an opportunity to compare observed and stated behaviour. Such analysis has previously not been carried out and indicates that there are systematic differences in RP and SP data.