Project description
Novel data analysis tool helps businesses detect anomalies in real time
The EU-funded Streem.ai project could transform how businesses handle large volumes of time series data in real time. The new tool utilises an unsupervised machine learning algorithm that helps users detect patterns and abnormalities in all new data associated with a product. In particular, it helps reduce the time spent on data analysis by 85 %, analyses all of the available data and improves the detection rate of previously unknown anomalies. In the automotive industry, for example, Streem.ai could detect faulty components for car manufacturers before they make it to the market. Given the COVID-19 pandemic, the tool could make virus testing more efficient by detecting irregular diagnostic device measurements.
Objective
Streem.ai is an AI anomaly detection tool for large volumes of time series data in real time. Its unsupervised machine learning enables identification of “new anomalies” without the need for annotated data.
Our unique selling point is that Streem.ai identifies outliers and unusual behaviour using algorithms that are selected through our proprietary benchmarking algorithm. As a result, our users can benefit from an up to 85% reduction in time spent on data analysis, analysis of 100% of the available data and improved detection rate of previously unknown anomalies.
The gap between the rate at which data is being collected and how fast it can be analysed is increasing exponentially, contributing to business insight latency amongst analysts and engineers in sectors as varied as manufacturing to healthcare. E.g. in the automotive industry, anomalies that go undetected can result in defective components going unnoticed until after market launch. In 2016 alone, defective components cost the car industry a record $22 billion through car recalls. The market size for testing in the automotive sector is now expected to grow at a CAGR of 5.02% from €15.9B in 2017 to €24.3B in 2023.
Given the COVID-19 pandemic that the world is facing, Streem.ai’s core technology can aid with more quality and efficient virus testing by monitoring diagnostic machines and analysing test results to identify patterns for research. One of our successfully completed pilot projects was with Roche diagnostics, the project involved monitoring the Analyzer machine data.
This project is being undertaken by Streem.ai GmbH based in Berlin. The CEO & one of the co-founders previously founded Plastelina which peaked 1.5m unique monthly visitors in 2000 representing 0.6% of all internet users then. The team has 12 employees of which 4 have PhDs. Through the Streem.ai project the company expects to create 18 new jobs and generate cumulative first 5 year profits of €18.84 million with a ROI of 493%.
Fields of science
- natural sciencescomputer and information sciencesdata science
- medical and health scienceshealth sciencespublic healthepidemiologypandemics
- engineering and technologymechanical engineeringvehicle engineeringautomotive engineering
- medical and health scienceshealth sciencesinfectious diseasesRNA virusescoronaviruses
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
Keywords
Programme(s)
Funding Scheme
SME-2 - SME instrument phase 2Coordinator
10999 Berlin
Germany
The organization defined itself as SME (small and medium-sized enterprise) at the time the Grant Agreement was signed.