Human dexterity is a complex phenomenon associated to physiological and cognitive factors that affect the execution of precise movements. Dexterity is strongly linked to upper limb (UL) functionality and performance, and its study is important for clinical analysis, ergonomics, sports biomechanics, design, rehabilitation, and human-machine interactions. However, its understanding is quite limited. Dexterity is commonly assessed through time-dependent dexterity tests that can determine the successful completion of tasks on test boards paced in front of the participant. However, such tests cannot inform about participant performance in other regions of the corresponding UL workspace volume (WV), and they can only collect data related to specific tasks, and therefore, cannot predict UL performance for the execution of other tasks. This thesis establishes a time-independent novel method for the characterisation of UL workspace with respect to dexterity; the “Dexterity Analysis Method” (DAM), which is based on the manipulability analysis method (used in robotics to quantify robot manipulability). The DAM is flexible, versatile, and scalable. It can be used to analyse real and virtual individuals or populations using direct measures or statistical data. Moreover, the DAM allows adding human factors, and to assigning their weights for adjustment and calibration. Hence, the DAM is a powerful tool that can help to evaluate performance, assess healthiness, optimise implants and prosthetics, design ergonomic workplaces and homes, develop assistive devices, and conduct pre- and post-surgery evaluations. Moreover, this work, as implemented in the DAM, promotes the use of WV as an objective reference to map performance, healthiness, and dexterity. Finally, the DAM contributes to closing the knowledge gaps on the understanding and quantification of UL motion, workspace, and dexterity. However, the DAM still needs to be fully validated as the experimental results obtained in this research with such purpose were not conclusive. A real-life application of the DAM is illustrated in Chapter 7 of this work, which analyses the effects of reverse shoulder arthroplasty (RSA) on WV and dexterity. The results indicate that WV for healthy people can be around 32% larger than those for people with RSA. However, it was found that greater WV do not necessarily translate into larger high dexterity regions, and the effects of reductions in ROM on WV depend on the extreme at which such reductions occur. For instance, a decrease of 15° in elbow extension reduces 2% of 2-D reachability, whereas a decrease of 15° in elbow flexion only reduces it by 10.8%. Therefore, surgeons should carefully consider such factors when making decisions during joint surgery, reconstruction, and implant position optimisation.