Sight Machine together with Microsoft and NVIDIA have developed a tool for high-speed automated data labelling of manufacturing plant data. The tool, Sight Machine Blueprint, is designed to make it possible for manufacturers to analyse all their plant data, improving outcomes in throughput, quality and sustainability. The tool allows manufacturers to make use of up to 100 times more plant data than current tagging methods are yielding.

Current Methods for manufacturing data labelling require significant time and subject matter experts to label data manually. Typically tagging schemas are typically unintelligible and difficult to contextualise, producing tags such as 268-IE-A-2ND-FFA-LC. As a result, many manufacturers are using only a fraction of the data available; in some cases, only one percent, creating significant missed opportunities for driving value through analytics.

“The true value of Blueprint is that it makes automated data labelling feasible in a way that it simply was not previously,” said Kurt DeMaagd, Chief AI Officer and Co-Founder, Sight Machine. “Without Blueprint, mapping data to the right machine is possible, but it is so slow and resource-intensive that it is prohibitive. With Blueprint’s ability to automate data labelling, manufacturers can realise unprecedented processing speed acceleration. Blueprint empowers manufacturers to prepare 100x more data for analysis, and from there, rapidly and smartly uncover new data insights to optimise plant operations.”

“With Blueprint’s robust machine learning technology, automated data labelling becomes possible,” said Piyush Modi, Global Business Development Leader and Chief Strategist of the Industrial Sector at NVIDIA. “The GPU-accelerated NVIDIA AI platform powers the development and deployment of highly performant AI pipelines to learn mapping between data sources and is ideal for crunching large volumes of data efficiently from edge to cloud – making this technology ideal to complement Blueprint and the impact it drives.”

“Microsoft Azure Machine Learning, combined with Sight Machine and advanced technology from NVIDIA, provides the infrastructure to easily scale GPU-based machine learning pipelines,” said Indranil Sircar, CTO, Manufacturing at Microsoft. “This combination in Sight Machine Blueprint will help eliminate manufacturers’ massive time drain from manually labelling data, and enable them to tap into the full wealth of data at their fingertips for business impact through true analytics-driven decision-making.”