Digital twins are software systems that facilitate the means to monitor, understand, and optimize the functions of all physical entities, living as well as non-living. In fact, it is likely that by 2021, half of large industrial companies will use digital twins. Expensive machinery, delicate parts, and complex systems can be digitally modeled in real time: paired with predictive analytics and forecast models, data collected by studying digital twins can transform a company’s production capabilities. These systems are arguably as complex in their algorithms, data, and software as the physical systems they model, but they offer an effective way to mitigate risks and improve performance without significant loss of time or money. In specific applications, digital twins can be used to support model-predictive control of systems to improve safety, efficiency, profitability, and performance.