Digital Intelligent Assistant for Predictive Maintenance as a Response to Demanding Employee Skill Requirements
The implementation of predictive maintenance (PdM) in production processes is a challenging, data-intensive task. It requires company specific maintenance strategies, which are generated by highly sought after and expensive data science experts. This, alongside pricey support software and further investment in personnel training, act as barriers in the adoption of PdM methods.
The DIAMOND project aims to lower these barriers by introducing a digital intelligent assistant for PdM. This technology will make use of artificial intelligence (AI) to support low-skilled employees in PdM tasks such as failure diagnoses, forecasting, and maintenance planning. The digital assistant will be designed in a way that allows employees to talk to it and ‘discuss’ the machines and processes they are working with.
It will be integrated to the extent that workers will be able to use it during meetings, for planning, and during shop floor tasks that require information about the health, condition, components, and maintenance history of specific machines.
The envisioned solution will take complex graphical interfaces and simplify them into a dialogue model. This will not only allow manufacturers to transform time-consuming workflows into short, voice-first interactions, but it will also create new ways PdM systems can be used. The DIAMOND project will evaluate the digital intelligent assistant through use cases at household appliance and medical device manufacturers.
This project is part of KIC EIT Manufacturing, funded by the European Union.
Santiago Ruiz Arenas
- TU Delft
- Whirlpool EMEA