Big data-based deterioration prediction models and infrastructure management: towards assetmetrics

Abstract

In the past decades, infrastructure management has been performed based on implicit knowledge, consisting experience and knowledge of professional engineers. The objective of assetmetrics is to convert such decision processes based on implicit knowledge and experience into systematic decision processes based on formal knowledge. The presented research and development policy is a practical approach that tries to create methodologies based on data obtained through daily and periodic inspections. Moreover, the authors point out that in the field of asset management, technical knowledge related to the analysis of existing data is more important than hardware technologies for obtaining new data. The authors also discuss these ideas as pertaining to the concept of big data. Finally, by presenting examples of advanced researches on assetmetrics, the authors give an overview of monitoring methodologies and comprehensive risk management in the relevant field.

Keywords:

• Big data
• infrastructure
• inspections
• knowledge management
• decision-making
• Markov Chain Monte Carlo method
• assetmetrics