Manufacturing at scale has always relied on data analytics in combination with the manufacturing engineer’s intuition borne out of many years, if not decades, of experience. As the complexity of manufacturing processes increases data analytical techniques and software tools have also increased in complexity. As an example univariate models, first done on graph paper and then on computers, gave way to multivariate statistical models. The shift was made possible because of increased computation power that could analyze the datasets being generated from products with an increased number of manufacturing steps and increased number of complex machines. The adoption of these techniques resulted in increases in top and bottom-line revenues. Those manufacturers that did not adopt fell by the wayside. An example of this from the semiconductor industry is the rise of South Korea’s semiconductor manufacturing. South Korea semiconductor fabs quickly adopted Statistical Process Control in the 90s, where as Japan was slow in adoption. As a result South Korea dominated over the Japanese manufacturers within a decade. Now, just like in the past, manufacturing has continued the trend of complexity. This complexity arises due to the ever-increasing requirements from customers for products to perform multiple function. In parallel the equipment being used have become a lot more complex and the rise of adoption of IoT sensors and devices have furthered the complexity. Standard data analytical techniques are failing the intuition of engineers resulting in production delays and missed opportunities and hence reducing manufacturer’s profit margins. New data analytical techniques such as Artificial Intelligence (AI) and Machine Learning (ML) can enable engineers to overcome the complexity. Next a general definition of AI/ML is given followed by how to best use these algorithms and finally limitations of these algorithms and their solutions.