BIG DATA ANALYTICS FOR PREDICTIVE DECISION MAKING IN COMPLEX BUSINESS ENVIRONMENTS

Abstract

Big Data Analytics (BDA) has evolved as a strategic capability that allows enterprises to deal with the ever-increasing complexity and volatility of the business environment. This survey review comprehensively evaluates the impact of predictive analytics in decision-making across a variety of contexts within an organization and synthesized evidence from 109 academic and practitioner journal articles, published between 2010 and 2024. Through our extensive literature analysis, we build a systematic taxonomy of the Big Data Predictive Analytics (BDPA) applications within diverse organizational domains: industrial manufacturing, healthcare, financial services, e-commerce, strategic management, smart agriculture, and the information and communication technology (ICT) sector. This work scrutinizes the key theoretical aspects of predictive analytics, its location in the continuum from descriptive to prescriptive Analytics, and systematically analyzes various analytical techniques used for making predictions, ranging from the traditional statistical approaches to sophisticated machine learning algorithms. Key insights from our review show that those organizations that have achieved advanced Analytics maturity are well-equipped to reap several substantial benefits: reduction of unplanned downtime (20%-40%), enrichment of customer knowledge, enhancement of the financial health of the organization, and the gaining of sustainable competitive advantage. However, we acknowledge the numerous persistent challenges to implementation, such as problems with the quality of the data, deficiencies in required skills, ethical implications of the use of predictive analytics, and the resistance within the organization to the adoption of data-driven decision making. We offer an Integrated conceptual framework connecting BDA with decision processes and business outcomes, highlight key facilitators and inhibitors to effective implementation. We investigate various novel technologies like Explainable Artificial Intelligence (XAI), Edge Analytics, and Privacy-Preserving Analytics that hold the promise to overcome existing limitations. In addition to bridging the gap between academic scholarship and practitioner experience in the application of BDPA, the survey review suggests the need to develop explainable predictive models, audit for fairness, and create models for ethical AI implementation as key future research directions.