Smart Maintenance

Resource Intelligent Maintenance Strategy for the Production of the Future

The research project "Smart Maintenance" was launched at the beginning of September 2014 as part of the FTI initiative "Production of the Future", which is dedicated primarily to the central questions of the material goods producing industry. The project consortium: Montan University Leoben - Chair of Economics and Management Sciences (coordinator), Software Competence Center Hagenberg, Messfeld as well as BMW Motoren and BRP-Powertrain. This brings together the leading national experts in their fields to research a resource-intelligent anticipatory maintenance approach and provide appropriate solutions for industry.


The aim of this research project - funded by the FFG - is to create an improved basis for maintenance strategy decisions. Maintenance costs are to be reduced while plant availability is to be further increased.


Manufacturing plants are subject to wear and tear. In the worst case, without countermeasures this can lead to component failure and unplanned shutdowns of the plant, but can also lead to a reduction in product quality, restricted operating conditions or increased energy consumption.

Wear and tear therefore requires an appropriate maintenance strategy with a package of reactive or preventive measures to minimise economic and ecological consequences. Preventive measures include e.g. condition monitoring systems.

The Innovative Approach of "Smart Maintenance"

For the first time, the project provides approaches for the derivation of a cost-optimized asset management strategy, taking into account

  • information from selective condition monitoring data,
  • a combinatorial data analysis, i.e. of current machine data (e.g. current consumption, power consumption) and stored plant history data (fault database), process and product data (e.g. quality features, number of faulty parts, fault characteristics), and
  • fault analysis and prediction.

Technical possibilities and limits are identified and the industrial application in a production system is evaluated economically.

The use of condition monitoring is optimized by using evaluation and decision models to analyse whether the use of condition monitoring technologies for system-critical plant components is at all technologically feasible or makes economic sense. In principle, both technological limits and business management aspects play an important role in the sense of risk and life cycle analyses of critical investments.

In addition to condition monitoring, data analysis methods (data mining of machine, process and product data) are used to identify possible patterns for machine failure behavior (e.g. an accumulation of plant failures occurs with a certain product group, temperature, time of day...). These patterns are subsequently used to develop methods for predicting faults, from which improved planning rules (algorithms) for maintenance strategies can be derived.

Added Value for Project Partners

Industrial surveys show that traditional methods (current consumption measurement, lubricant analysis, temperature measurement) are particularly well established, e.g. acoustic measurement methods are even less frequent.

Through the combinatorial application of these methods, Smart Maintenance's approach goes far beyond the traditional established methods and is also part of a superordinate asset management strategy.

The industrial partners BMW Motoren and BRP-Powertrain benefit directly from the concrete implementation and application of the developed methods and models. In addition, the developed models can be better validated by using practical parameters.

Project Data

Duration: 01.09.2014 - 31.12.2017
Budget: ca. 1.417.000 € total costst and ca. 938.000 € funding contribution
Partners :

  • Montan University Leoben - Chair of Economics and Management Sciences (coordinator)
  • Software Competence Center Hagenberg
  • Messfeld GmbH
  • BRP-Powertrain GmbH & Co KG
  • BMW Motoren GmbH

Funding partner: FFG, Production of the Future - 4th Call


Bernhard Freudenthaler

Freudenthaler Bernhard

Area Manager Data Science
Phone: +43 50 343 850