Challenges of Applying Predictive Analytics in Transport Logistics

Abstract

The field of Predictive Analytics (PA) provides the possibility to utilize large amounts of data to improve forecasting, data-driven decision-making, and competitive advantage. Especially the transport logistics sector, which is characterized by high business-related uncertainties, time-sensitivity, and volatility, highly benefits from accurate resource and production planning. While success factors and framework conditions of applying PA are well-investigated on a theoretical SCM level, findings on internal and external challenges of transport logistics organizations remain scarce. Therefore, based on a multiple case approach, this study offers in-depth insights into six real-world cases of freight forwarders, ocean carriers, and air carriers. The results uncover both internal and external challenges. From the internal perspective, the biggest challenges are related to the technical implementation including the acquisition of globally generated, internal and external data and its harmonization. In addition, stakeholder management and target setting impede the development of PA. Regarding external challenges, relational and external conditions hamper the application. Therefore, especially actions of third-party institutions in terms of standardization and security enhancements are required. This study contributes to the existing literature in various ways as the systematic identification addresses real-world issues of PA in the neglected but crucial area of transport logistics, discussing urgent research needs and highlighting potential solutions. Additionally, the results offer valuable guidance for managers when implementing PA in transport logistics.