As one of the major trends in the automotive industry, assisted and automated driving has a significant impact on the way people travel in automobiles today and in the future. In addition to the development of Advanced Driver Assistance Systems (ADAS) and Automated Driving Systems (ADS), the integration and testing of the systems with the objective of giving a release requires a lot of resources in the vehicle development process. Integration and testing include iterative calibration, verification, and validation of the systems. Various testing approaches are applied to perform the processes efficiently and to keep the required resources within a feasible extent. These testing approaches include testing in prototype vehicles, shadow testing, data-replay testing and virtual testing in simulation, which features flexibility in test case definition and the automated execution of test cases in virtual driving tests. From the requirements of the ADAS/ADS integration and testing process and the connection to other testing approaches, use cases of virtual ADAS/ADS calibration and validation are defined in this dissertation. The objective of the dissertation is to develop efficient, virtual ADAS/ADS calibration and validation methods to cover the defined use cases. The basic components of the methods are combined in a modular, virtual ADAS/ADS testing framework, which features scalability to different systems under test and use cases. For virtual ADAS/ADS calibration with a large set of calibration-relevant scenarios, a novel multi-scenario-level method is presented. In order to consider a large set of scenarios in calibration, these scenarios are divided into different levels in which the data set of the ADAS/ADS is iteratively optimized. By including the optimization results of one scenario level to initialize the optimization in the following scenario level, the efficiency of the virtual calibration is significantly increased. Within the dissertation, the multi-scenario-level method is also applied to other use cases of virtual ADAS/ADS calibration, such as calibration for different system modes, customer groups, markets, vehicle derivatives and sub-areas of the systems Operational Design Domain. The data sets determined in virtual calibration serve as a starting point for fine-tuning in prototype vehicles. In addition, a novel virtual validation method is presented, in which the boundaries between critical and non-critical test cases are iteratively identified in the scenario parameter space of a logical scenario. Due to the iterative, adaptive sampling of test cases, the method features a high efficiency increase compared to conventional methods and is suitable for efficient comparison of different hardware and software versions as well as the generation of criticality indicators in the context of virtual ADAS/ADS validation.