User tasks are mapped to tools that manipulate data on the server (see Fig. 1.). This shortens the learning curve for users by reducing the complexity of each tool to the essential functionality required for the task the tool was designed to support. Fig. 1. Overview of the CREATE system. Fig. 2. The Layout Tool with the model library to the left, from which objects can be dragged into the scene. The virtual prototype is constructed by dragging and dropping objects from the project context’s model library into a 3D scene (see Fig. 2.). It is also possible to model entirely new objects within the Layout Tool itself. This is particularly useful early in the design process when basic shapes may be used to indicate regions or identify the future locations of specific objects that have not yet been modeled. The room or building can be modeled or extended using the Layout Tool to add walls and openings for door and windows. Parametric objects are available to easily add door and stairways. Alternatively, an existing 3D model of a room or building can be imported into the system via the Model Bank Tool. Fig. 4. Evaluation of label legibility, showing the height of the text and calculated range of legibility. The label size tool (Fig. 4.) is used to measure the height of text and width of characters, and visualizes the distance within which a label is considered readable. It is used to check the legibility of labels on control panels and other text in the control room and surrounding environment. For example, it can be used to verify that alarm information can be read from specific locations. Fig. 5. A manikin with a view cone and line of sight (broken by a computer display) to the left and another manikin with its reach envelope shown to the right. Fig. 6. Virtual control room from the first study. Twelve subjects with control room verification and validation experience, but no VR experience, participated in the experiment. Five review tasks totaling fifty guidelines were selected from NUREG-0700, and were carried out by the subjects in both conditions. The performance measures were task correctness, in comparison with reference scores agreed by an independent expert panel, and task completion time. Subjective usability was evaluated via a questionnaire.