Osteoarthritis and rotator cuff tears are two common and wide spread shoulder diseases leading to significant restrictions in the daily life of the affected. The pathogenesis is not fully understood yet. A chronic process triggered by multiple causes is assumed including inflammation, metabolic disturbances, and mechanical contributions. Several studies described as mechanical cause a statistically important impact of the scapula shape. The appearance of osteoarthritis and rotator cuff tears could be linked with specific anatomic parameters. The aim of this thesis was to improve the understanding of the mechanical contribution to the development of osteoarthritis and rotator cuff tears by the means of computational modelling. In this work, we focused on the lateral extension of the acromion and the glenoid inclination. Three numerical shoulder models were developed to analyse these two parameters. A musculoskeletal model was used to evaluate muscle and joint reaction forces acting on the glenohumeral joint. The underlying algorithm calculated joint torques for a given movement based on an inverse dynamics approach and solved the undetermined problem of muscle coordination with a nullspace optimization routine. The calculated joint forces were then used as input for a finite element cartilage model. Cartilage was modelled as incompressible hyperelastic material characterized with experimental data. The model provided values for cartilage strain and humeral head translations. The humeral head translation and supraspinatus force were used in a third model of the supraspinatus tendon. The constitutive equations for tendon tissue were derived from an anisotropic hyperelastic strain energy potential. Material parameters were identified in both fibre direction and the transverse plane. The elasticity tensor was approximated using a forward differentiation algorithm. Using the finite element method, tendon strains and tendon impingement were evaluated. The biomechanical variables cartilage strain, subacromial space, and tendon strain were analysed for different anatomic configurations. All models were validated with available clinical data. The biomechanical variables were more sensitive to changes of the acromion length than to changes of the glenoid inclination. The numerical study could explain the statistical association for osteoarthritis for short acromion by an increase in cartilage strains. A long acromion increased superior humeral head translation, which is associated to tendon impingement and related tears. We also observed increasing tendon strains close to the tendon insertion for a short acromion. Clinical studies did not correlate tendon tears to a short acromion. However, the clinical studies did not distinguish between the different types of tendon tears. At this point, further clinical studies would be necessary to clarify the numerical results. A better understanding about physiologic anatomic parameters could help to improve the treatment of osteoarthritis and rotator cuff tears. Prosthesis design and positioning during total shoulder arthroplasty, tendon repair techniques, and conservative treatment could be improved to compensate negative effects of excessive anatomic parameters.