Urban decision-making involves choices made over different time horizons,from short-term daily activity and travel scheduling to long-term residential location and land-use planning. These decisions occur at multiple levels,including individuals, households,and public authorities,and are shaped by interactions both within and across these levels.Short-term choices often involve intra-household interactions,where individuals coordinate their daily activities in response to shared constraints and preferences.Long-term decisions reflect broader spatial interactions,such as those between residents' location and mobility choices and urban policies.These choices also highlight the interplay between transport and land-use systems,underscoring the mutual influence of mobility and spatial development.This thesis contributes to a new generation of travel models by advancing disaggregate approaches,namely activity-based models,and offering aggregate alternatives when such models become overly complex.The work spans both methodological innovation and practical application. First, the thesis focuses on short-term decisions;daily activity and travel scheduling, capturing intra-houshold dependencies in daily activity scheduling. This thesis introduces the household-level OASIS (Optimisation-based Activity Scheduling with Integrated Simultaneous choice dimensions),an integrated household scheduling framework based on mixed-integer utility optimisation. Household schedules are simulated such that the household utility is maximised, subject to individual and household constraints and preferences.Multiple scheduling dimensions-activity participation, location, timing, duration, participation mode (solo or joint), and transport mode-are simulated within a single framework. Intra-household interactions are embedded through constraints and objective function, capturing vehicle allocation, task sharing, escorting, joint participation, and ride-sharing.This integrated simulation framework explicitly reflects complex intra-household interactions in disaggregated results,illustrating how such interactions can lead individuals to deviate from schedules planned in isolation.This is a general framework applicable to different household compositions and available resources. Second,the framework is operationalised via an estimation process that estimates the parameters of the household schedule utility function from data.Household-level choice sets are generated with a Metropolis Hastings algorithm, and parameters are estimated by a maximum likelihood estimation procedure.Applied to a sampled dataset,the method reproduces realistic distributions of daily activities and yields significant, behaviourally interpretable parameters.Model calibrated at the household level match observed activity patterns better than individual-level model. Third,the analysis extends to long-term urban dynamics using an aggregated multi-year model that links transport and housing through feedbacks,explicit time lags, and hierarchical decisions.Fast processes (e.g. mode choice),medium-horizon relocation,and slow infrastructure development are integrated in one framework.A cross-national case study demonstrates the model's practical use.By tracing the evolution of key indicators,the framework serves as an indicative tool to understand system behaviour,anticipate future trends,and assess the long-term impacts of policy interventions,while remaining computationally efficient for strategic use.