Deep learning for sustainable asset management decision-making

Asset management requires a more sophisticated approach for sustainable data-driven decision-making. Organisations operate in dynamic, interconnected and uncertain environments, in which swift, optimal decisions are thwarted by a myriad of conflicting influences and information. Yet the need to prioritise sustainable asset management has never been greater, as organisations transform in the face of a new Covid world that is already in crisis. Deep learning helps to understand key features from data laden procedures for greater stability and agility, in order to transform opportunities across distinctive functions. By processing data directly from input to output, clear and concise information can support decision-making while uncovering insights, detecting anomalies or predicting future states. In this paper, the underlying concepts of deep learning are explored, as inputs for decision-making. Issues and challenges are articulated from several viewpoints, expressly from the perspective of algorithmic constraints and also in a context of holistic management of organisational assets. Advances in deep learning can expose nascent opportunities for visionary firms to explore. By focusing on up-to-date applied research, this paper concludes with a summary of sustainable deep learning insights that lead to transformational asset management decisions with far-reaching organisational consequence. The result is a set of future-based predictions for the use of deep learning in asset management for sustainable decision support.