Secondary ion mass spectrometry (SIMS) is a technique that has evolved to be one of the most powerful techniques for the analysis of organic samples. Modern instruments are capable of obtaining three-dimensional information with high spatial resolution of a material with information as rich as a full mass spectrum at every voxel of the 3D structure, thus generating very large and complex datasets. Multivariate analysis (MVA) methods are used within the SIMS community, however, the absence of MVA in the software packages of instrument manufacturers together with constant increase in data and data analysis complexity demands practical data analysis solutions that are accessible to scientists of diverse backgrounds. This thesis aims to expand the applicability of three major MVA methods to complex SIMS datasets: Principal component analysis (PCA), non-negative matrix factorisation (NMF) and k-means clustering. This is achieved by establishing and validating existing and novel methodologies for the processing of large and complex datasets. Furthermore, it presents the development of a software that encompasses these methodologies and provide accessible and flexible analysis and data visualisation tools. Finally, it presents the application of the software to a series of experiments carried out at The Surface Analysis Laboratory of the University of Surrey in which data processing enabled deeper interpretation of the results and helped to achieve insights towards scientific and industrial problem solving.