Nowadays, due to the widespread access to information, modern consumers acquire products after comparing different alternatives. As a result, there is a need to design and manufacture products with improved performance and increased functionality at reduced price. Examples of mechatronic system cover a broad range such as home appliances, consumer goods, modern automotive and aerospace vehicles, manufacturing machines, industrial robots, machine tools and many others. A typical mechatronic system comprises several sub-systems. Integrating these sub-systems and achieving system level functionality and performance specifications is very challenging due to limited understanding of complex interdependencies and interactions between these different sub-systems. Traditional or conventional system engineering methods used in the modern industry segment do not allow to push the existing limits. To this end, model centric approaches are gaining popularity in the mechatronic industry in order to overcome the existing shortcomings and accelerate the design and development process. This doctoral research focuses on combining modeling and model-based control techniques to enhance model based system engineering for modern complex mechatronics systems. In order to demonstrate the potential of model based approaches, three different contributions relevant to mechatronic industry are made. The first contribution proposes to use multi-fidelity models for the virtual prototyping of complex mechatronic system. To this end, a complex model for an industrial 3-axes industrial machine is made to demonstrate the potential of virtual prototyping. The model comprises a 3D flexible multibody model, a 1D drive line model and a controller. The virtual design and analysis help to reduce the number of design and prototyping iterations, and consequently avoids delays, cost over runs, brand damages and even customer dissatisfaction. The second contribution is a system-level design approach to push the performance of a mechatronic system to its limits. Conventionally, in order to design and develop a complete mechatronic system, engineers adopt a sequential approach. Often, a control engineer cannot meet the desired specifications because of the poorly designed plant. To this end, an efficient concurrent design of optimal plant and feedback controller is proposed. The proposed technique falls under the category of nested optimization approach and can deal with the shortcomings of the existing techniques available in the literature. The developed co-design approach is validated for an industrial gantry machine and an electric motor. In industry, engineers continuously try to find discrepancies in the design before rolling out a new product to the market. Therefore, service load simulation and component testing such as qualification tests, durability tests, and endurance tests are an integral part of the overall design and development process. A substantial amount of these tests is usually performed on multi-axial hydraulic test rigs in the laboratory. The third contribution is an alternative strategy for the service load simulation process that can deal with the shortcomings of the existing techniques available in the literature. This new strategy has been validated both numerically using linear and non-linear models of a suspension test rig and experimentally using an industrial multi-axial hydraulic CUBE Shaker.