The Best of Both Worlds : Machine Learning and Behavioral Science in Operations Management

Problem Definition: Two increasingly popular disciplines applied in operations management are (a) machine learning, such as deep learning techniques (“ML”), and (b) behavioral science, notably human-subject experiments and behavioral modeling (“BSci”). Despite the overlap in how they are used, ML and BSci are often considered disjoint fields. Instead of treating them as mutually exclusive, in this article, we discuss how ML and BSci can work as complements to solve important OM problems. Methodology/results: We first provide a summary of the objectives, strengths, and weaknesses of ML and BSci. We then propose a set of frameworks to help identify how ML and BSci can contribute to an OM problem, depending on the nature of the problem, the objective of the researcher, and the availability of data. We then detail how each step of the ML research process can benefit from BSci and vice versa. We discuss some particularly important areas for integration, revolving around algorithm aversion, ethics, and behavioral game theory. Managerial implications: Overall, we aim to explore how the integration of ML and BSci can enable researchers to solve a wide range of problems within OM, allowing future research to generate valuable insights for both managers and companies