MANTIS

Cyber Physical System based Proactive Collaborative Maintenance

A 50 partner strong consortium funded by ECSEL-JU. MANTIS addressed a proactive maintenance service platform architecture that allows to estimate future performance, to predict and prevent imminent failures and to schedule proactive maintenance. The proactive maintenance service platform architecture consists of distributed processing chains that efficiently transform raw data into knowledge while minimising the need for transfer bandwidth.

After three years of work, MANTIS partners achieved its objectives, namely:

1. There exists a MANTIS reference architecture, that has been published in several research papers, which is the principal content of the MANTIS Book and is already being used by a follow-up H2020 FoF project and partners that continued their own MANTIS implementations.
2. There exist a number of cyber-physical systems, cloud services and many components of different nature in the plethora of use cases that have validated the MANTIS Maintenance architecture. Those need a clear guidance for interoperability that was provided in the form of deliverables, and also as Open Access Book chapter.
3. We learnt how to select the optimal sensors that can provide the wear related information of a specific component, what type of intelligent functions are available to provide higher knowledge level and have validated them. We researched on how to communicate the sensors in challenging environments, even restricting the bandwidth they need to provide information. And more importantly, we made several processes of design, develop, and validation of new sensors to the community. These all have been documented in deliverables related to the WP3 and also as a book chapter.
4. We proposed, created, described and validated Artificial Intelligence algorithms in the form of algorithm portfolios, so the industry can similarly use or apply them, based on our experience, and the type of data we were facing with. There exist validated examples in most of the 11 use cases of MANTIS. Those apply for Root Cause Analysis, Remaining Useful Life prediction, Failure prediction, Maintenance Optimisation, and collaborative decision making. All available as a book chapter.
5. We made and published a valuable exercise on how different maintenance scenarios can be translated into Human-Machine Interfaces, representing data in different interface types (from Web UIs to AR and VR), that can be interacted with. There are a number of HMIs developed in the project that demonstrate different approaches to Maintenance interaction. All can be learnt from the book chapter.