Deep Learning Decision Support for Sustainable Asset Management

Sustainable decisions can be thwarted by a plethora of conflicting influences and information, yet the need to prioritise sustainability in management has never been greater. In an era of high- and ultra-high dimensional data, deep learning models offer the scalability required to extract good representations of significant features from raw data. With automatic learning at several levels of abstraction, deep learning can support sustainable asset management by learning complex functions mapping of systems. By processing data directly from input to output, clear and concise information can support visionary asset management whilst exposing hidden insights, detecting anomalies or predicting future states. This research will look at the necessity of applying deep learning in sustainable asset management and reveal some of the challenges that exist.