About the Prognostics Data Library

The Prognostics Data Library is a web-based data management system for use by researchers and industry around the world to access suitable datasets for testing prognostic models. The PDL is primarily built for three groups: researchers needing datasets, industry and academics with datasets.

A crucial difference in the PDL over other databases is the emphasis on capturing meta-data about the asset and its maintenance and operating environments.

Ultimately, the aim of this Prognostics Data Library (PDL) is to collate and enable sharing of high quality datasets across the entire prognostics community.

For more information about the Prognostics Data Library please see our paper here.

Our Vision

The Prognostics Data Library aims to link the best modelling minds around the world with the very best prognostic datasets collected from experimental equipment and real systems.

Consistent Format

Wherever possible, datasets will be provided in a standard csv format for simple utilisation.

Simple Web Interface

With our simple web interface, you will be able to access the data library from anywhere, at any time, on any device.

Comprehensive Metadata

Datasets will be accompanied by a comprehensive file of metadata so you know as much as possible about where the data came from.

Corporate Anonymity

Any corporate identifiers that could be used for commercial gain will be removed from datasets and their meta data.

Reliable Sources

Datasets will be provided by researchers who have already published using the data, or by industry sharing realistic data from operating equipment. References to cite, or details about how the data was collected, will always be supplied.

Aggregate Statistics

In addition to meta data, we'll provide you with basic aggregate statistics for each column of data in the dataset. This way you can quickly verify your algorithms and get back to modelling.

Why use the data library?

Academics with datasets

  • Generate citations for your prior work.
  • Benchmark your work against others using the same datasets
  • Be recognised for sharing your data when applying for grants and promotions.

Industry with datasets

  • Maintain corporate confidentiality.
  • Allow top researchers to work on your data without significant additional expenditure to business.
  • Find out when users have developed new and improved models for your datasets.

Researchers needing datasets

  • Find quality datasets easily and quickly.
  • Spend your time improving your models rather than collecting data.
  • Downloading datasets is free.
  • View the metadata and summary statistics prior to downloading a dataset.
  • Benchmark your research against other modelling experts using the same datasets.
  • Look at what issues prior users have had with the datasets.