Industrial engineering graduates need to be familiar with artificial intelligence (AI) due to its transformative impact on modern manufacturing and production processes. AI technologies, such as machine learning and predictive analytics, optimize resource allocation, enhance efficiency, and streamline operations. Proficiency in AI equips graduates to innovate, automate tasks, and address complex industrial challenges effectively. Predictive models are typically taught in one or more Industrial Engineering courses, such as Operations Planning and Control at Colorado State University Pueblo. It is beneficial that students learn the general implications of using predictive models. The models utilize various AI algorithms, where an algorithm learns from a retrospective dataset that comprises samples and features to make predictions. An AI algorithm trained on a balanced dataset produces good results, while an algorithm trained on a biased dataset may lead to unfavorable outcomes. Some dataset features can be eliminated as insignificant, and the AI algorithm is deployed only on the significant features. There are many reasons why features are insignificant beyond using just biased datasets. However, it is worthwhile to investigate the effects of these insignificant features, as detailed analyses can reveal damaging or positive consequences of the omitted features. In this paper, three publicly accessible datasets are used to present subjective analyses of insignificant features beyond the general recommendation of an AI algorithm.
Are you a researcher? Would you like to cite this paper? Visit the ASEE document repository at peer.asee.org for more tools and easy citations.