Activity-based models offer the potential for a far deeper understanding of daily mobility behaviour than trip-based models. Based on the fundamental assumption that travel demand is derived from the need to do activities, they are flexible tools that aim to put individuals and multidimensional interactions at the centre of the analysis. Due to their complexity and combinatorial nature, activity-based models used in research and practice have often relied on assumptions, predefined rules and modelling structures, which tend to oversimplify the scheduling process and limit the behavioural accuracy of the outputs. Specifically, the sequential approach used to model activity-travel decisions coupled with arbitrary model specifications and parameters significantly hinder the potential of these models. In this thesis, we introduce OASIS, an integrated framework to simulate activity schedules for given individuals based on utility maximisation under time and space constraints. In OASIS, all choice dimensions (activity participation, location, start time, duration and transportation mode) are considered simultaneously into a single optimisation problem. The fundamental behavioural principle behind our approach is that individuals schedule their day to maximise their overall derived utility from the activities they complete, according to their individual needs, constraints, and preferences. Constraints are a critical component in explaining activity-travel behaviour and are explicitly accounted for in OASIS. By combining multiple choices into a single optimisation problem, and considering both the influence of constraints and preferences, our framework can capture trade-offs between scheduling decisions (e.g. spending less time in an activity to ensure enough time for another one or choosing locations where multiple activities can be performed). We present a methodology to estimate the parameters of the schedule utility function from historical data to generate realistic and consistent daily mobility schedules. The estimation process has two main elements: (i) choice set generation, using the Metropolis-Hasting algorithm, and (ii) estimation of the maximum likelihood estimators of the parameters. As the initial analyses of the framework were conducted considering only one day of activities, we extend the single-day framework to include intrapersonal interactions influencing longer-term decisions (e.g. weekly scheduling). Finally, we present two successful practical implementations of OASIS, showcasing its versatility and potential for contributions in different research domains.