Artificial intelligence in managerial decision-making: A systematic review of applications, constraints, and emerging tensions

Abstract

The use of artificial intelligence (AI) in managerial decision making is now common practice in strategic, tactical, and operational decision making and literature on this topic is scattered over fields, industries, and decision contexts. The aim of this systematic literature review is to demonstrate how previous scientific research studies have played a part in the managerial decision-making process, its outcome, and the organizational dynamics of the process, and to highlight the remaining challenges that hinder the effective use of AI. The review focuses on the use of machine learning, predictive analytics, and decision-support systems in the past decade, based on peer-reviewed research articles, to improve forecasting accuracy, combat cognitive bias, and facilitate complex managerial judgments (Jarrahi, 2018; Davenport & Ronanki, 2018). Meanwhile, the results show persistent issues with data bias, transparency, explainability, and the excessive dependence on the algorithmic output by the manager, which hinders accountability and trust in the decisions made with the help of AI tools (Rai et al., 2019; Shrestha et al., 2021). This review builds on a thematic framework that synthesizes management, information systems, and organizational theory perspectives relating AI capabilities to decision levels and human–AI interaction modes.