Li-ion batteries are currently the energy storage of choice for a wide range of applications reaching from consumer electronics and power tools up to battery electric vehicles (BEV) and stationary storage. While the requirements for the different products may differ, or even depend on the individual usage pattern, cell and system development are often not closely interlinked and are realized in consecutive steps. Hence, battery usage can be suboptimal for specific applications by having oversized safety margins or, contrary, overstraining the battery and causing a rapid decay in performance or safety. At the same time, application specific physical prototyping from cell to system scale is time-consuming and costly, in particular for BEVs, such that cell development is often limited to cell scale and standardized operation protocols and lacking broad coverage of usage scenarios. This issue can be alleviated by considering holistic multiscale simulation frameworks that allow for simultaneous simulation of system and battery performance. As a result, key requirements and potential limitations can be identified under distinct usage profiles. In this work, we are presenting a two-way coupling of a region- and usage-dependent vehicle simulation with a model for the traction battery. Individual cell behaviour is described by considering an equivalent circuit model or a pseudo-two-dimensional electrochemical model. The resulting BEV model allows to utilize a geo-referenced database of environmental information around the globe and various driver profile characteristics in order to perform virtual test drives under realistic driving conditions. Our modelling approach focuses on a high computational efficiency achieving a simulation to real-time ratio significantly smaller than 1%. As a result, it enables rapid evaluation of system performance and large scale statistical analysis to support early stages of vehicle development.