The focus of the material presented in this PhD thesis is the development of stochastic Direct and Inverse Monte-Carlo-based models for drug dissolution. Drug dissolution from different carriers is a complex phenomenon. Limited knowledge is available on some of the underlying constituent processes, which restricts development of mechanistic models. Monte Carlo techniques permit the treatment of certain structures and events in a probabilistic manner. The thesis examines a number of possible ways of using Monte Carlo both (1) to explore modelling for the dissolution of Drug Delivery Systems and (2) to reconstruct general system behaviour during dissolution, using noisy drug delivery data. Further, important investigations on the determination of factors responsible for noise and quantification of noise levels, are reported. In the first part of the thesis, an investigation of MC-based methods in the field of Drug Delivery is given, with the complexity of drug dissolution and design explored and the contribution of the MC approach reported. The use of Direct MC and Stochastic Cellular Automata models in the simulation of dissolution from pharmaceutical compacts or related phenomena are discussed, together with various features and requirements. The principal objective here is to extend use of Direct Monte Carlo techniques in simulating drug delivery from compacts of complex composition, taking into consideration special features of the dissolution in an in vitro environment. After examining the existing MC models for drug delivery, the need for more sophisticated models is described. Exploratory modelling is proposed in order to address the problems of dissolution related to certain drug carriers with complex internal morphology and difficult-to-predict dissolution profiles. Phenomena such as local interactions of dissolving components, development of wall-roughness at the solid-liquid interface, diffusion through occlusions and pores and moving concentration boundary layers were examined and directly accounted for in the model. As a result, new models have been developed for: i) matrix soluble drug carriers and ii) bioerodible polymeric micro- and nanospheres for controlled release of proteins. The simulations provide results in acceptable agreement with different drug release profiles obtained during laboratory experiments. The novelty of this work consists in including new features of experimental system complexity in the frame of simple and user-friendly Direct MC models, indicating that the Direct MC technique can be very helpful in exploring design parameters in the field of drug delivery. The other major axis of the thesis investigates use of Monte Carlo in data reconstruction and noise quantification. The problem posed was whether it is possible to extract detailed dynamic distributional knowledge about a dissolving pharmaceutical system composed from many small entities, when the researcher is provided with insufficient experimental data. A model based on Inverse Monte Carlo simulations was designed to exploit Bayesian principles in retrieving the desired features, such as particle size distribution. Importantly, this work demonstrates that Inverse Monte Carlo methods are capable of reconstructing underlying characteristics of drug carriers involved, even when dissolution profiles available rely on sparse data sets. The models proposed in this thesis are currently being incorporated in a largescale project in collaboration between the DCU research team and the Hospital for Special Surgery, New York. The project focuses on developing therapeutic implants with controlled drug release, specifically designed for the regeneration of severely damaged tissues.