Automated infrastructure: Integration of diverse data
The degradation monitoring of infrastructure, such as roads and railways, in the US faces challenges due to the vastness of the network and limited resources for continuous inspection. Artificial Intelligence (AI) can play a vital role by predicting infrastructure degradation by leveraging large-scale data sets, including information such as sensor readings, traffic patterns, and weather conditions. AI-powered predictive models can provide valuable insights, enabling timely maintenance and cost-effective infrastructure management to ensure safer and more efficient transportation systems.
Rutgers researcher Dr. Xiang Liu has developed a Machine Learning based system to forecast infrastructure degradation of transportation systems such as rail and roadways. This system provides support for optimal infrastructure asset management decisions using automated infrastructure data processing, integration of diverse data, segmentation, predictive analytics, visualization and reporting. The infrastructure owner or operator only needs to identify and import raw data (files) and the system automatically performs a pipeline of machine learning processes and generates predictions. This system of infrastructure prediction enables data-focused asset management, ensuring safety despite scarce resources.
- The machine learning technique can process large data sets from multiple sources and streamline the data science pipeline.
- Provides the ability to identify infrastructures most likely to experience failures in advance of occurrence of problems.
- Adaptable to be used with most modes of linear transportation infrastructure such as rail, bridges & roadways.
- Railway asset monitoring
- Rail/ road network maintenance solutions
- Infrastructure optimization
Intellectual Property & Development Status: Patent number WO 2022/159565 A1. Available for licensing and/or research collaboration. For any business development and other collaborative partnerships contact firstname.lastname@example.org