Development of a new type of intelligent grease gun to improve manual lubrication processes and lubricant management and the associated cloud-based software solution with implementation of smart maintenance algorithms.

Industry
Dosing technology

Services
Scalable maintenance DB
Machine Learning
Decision Support
Predictive Maintenance
Software Demonstrator
Cloud backend


Quick Facts

2
Duration of the R&D project in years
3
Number of research partners
97,9 %
Precision K-Nearest Neighbours
10
Employees in the project

The challenge

In the maintenance of large industrial plants, it is state of the art to automatically and precisely meter lubrication quantities, plan lubrication processes as part of smart maintenance and document them in a legally compliant manner. In contrast, lubrication in workshops is still carried out manually, based on visual judgement and rigid lubrication schedules.

This research approach aims to bridge this gap. An intelligent lubrication system is being developed that is connected to a smart maintenance cloud via NB-IoT. The aim is to make modern lubrication processes widely available.


The solution

Analysis and system design
Analysing functional and technical requirements, taking into account data protection and IT security. Design of a distributed and scalable overall system and the interaction of the components involved. Selection and evaluation of suitable technologies, programming languages, standards and frameworks.
Development of a cloud-based IoT platform + Milvus
A customised cloud-based IoT platform with an extended security architecture was developed, REST APIs for communication between the IoT components were designed and developed and a scalable framework for machine learning, predictive maintenance and decision support was set up. The user interface was implemented according to the latest UI/UX standards.
Machine learning and smart maintenance
To develop a machine learning algorithm for decision support for predictive maintenance, the annotated data was processed, the data quality was evaluated and experiments were conducted with normalisation, standardisation, feature selection and other transformation methods. After determining the baseline, the various algorithms were implemented and evaluated in terms of precision and performance compared to the test data.

Technologies used

About the research partners

With over 2,500 students from around 80 nations, more than 120 partner universities and one of the largest alumni networks in Europe, ESB Business School at Reutlingen University is one of the leading faculties for international business administration and industrial engineering. With its successful double degree programmes, it set international standards in executive education at an early stage. Since 2019, ESB Business School has been accredited by the Association to Advance Collegiate Schools of Business (AACSB), the world’s largest accreditation organisation.

Axiss Achsen- und Dosiersysteme GmbH from Keltern is a medium-sized company with many years of experience and develops dosing systems for various applications with material-specific customisation, maximum dosing accuracy and simple integration, operation and maintenance.

Research programme

Axiss Achsen- und Dosiersysteme GmbH from Keltern is a medium-sized company with many years of experience and develops dosing systems for various applications with material-specific customisation, maximum dosing accuracy and simple integration, operation and maintenance.


Do you have any questions? We will be happy to help you!

Do you have a project that you would like to talk to us about? You are welcome to contact us at any time. We look forward to getting to know you!

Marc Ponschab
Marc Ponschab Head of Technology Lab / R&D