I-Form Recommender System can:

• Present recommendations by reasoning about detected anomalies - either as individual or patterns of events. This allows for multiple factors to be considered before advising an operator.
• As the software is vendor independent, it has the advantage of being able to capture and evaluate data from different sources, including from AM systems supplied by different manufacturers, thus facilitating improved data exchange from different sources and in different formats.
• The ‘plug and play’ Recommender System architecture allows it to operate independently of both the input data source, the forms of data analysis applied, as well as the selection of rules it uses for reasoning.

Please find the video link below:

Background

Custom/vendor specific software are often not compatible with other software tools. This work decouples several aspects of the implementation in order to allow application of the recommender system across different hardware and domains.

In addition, homogenizing data reduces the development workload as analysis routines need only be implemented for the homogenized data. These are implemented as a black- box. These take data from data-streams provided by the manufacturing platform and report their results back as events. Several different analysis routines can simultaneously report their results as events.

Please contact the academic lead on this project Dr Cathal Hoare (cathal.hoare@i-form.ie) or info@i-form.ie to trial the tool and get more details.