ABSTRACT
This chapter introduces the virtual-to-reality big data (VRBD) methodology. Big data is commonly defined by three Vs: volume, velocity, and variety. To move this information from the virtual world to reality requires actionable and impactful changes, which is typically the fourth V, value. This chapter introduces four Rs, remove, recognize, remedy, and repeat, to create this value. Remove eliminates incorrect and useless data. Recognize searches through the remaining data to identify problems. The remedy step provides hints and suggestions for how to solve the problem. Repeat ensures that each of these steps is fast and can generate numerous instances of improvement. This VRBD methodology is implemented at a manufacturing facility. In this case, artificial intelligence/machine learning methods, such as MapReduce, isolation forests, principal components analysis, and random decision trees, efficiently handle the four Vs and Rs. The facility now rapidly identifies and fixes predatory profit-eating problems that are broadly classified as wolves, pythons, crocodiles, and sharks.
