Predictive Maintenance

Optimising production and new service models

Predictive Maintenance

Predictive Maintenance is one of the central applications within Industrie 4.0.

Analysing vast amounts of data collected from a network of connected sensors installed in production facilities, enables companies to make reliable predictions on how the conditions of a machine or facility will develop over time and when maintenance will be required. The condition of production facilities, however, exerts a direct influence on the quality of the final product.

Therefore a close nexus between Predictive Maintenance and Predictive Quality can be established. Last not least, these new technological scenarios provide for opportunities of developing innovative service models as well, enabling machine manufacturers to set new standards with regard to customer relationship managemen.

Model and Infrastructure

The prediction model is at the core of any predictive maintenance scenario: modelling starts with identification of relevant parameters, such as temperature, pressure, vibration or visual characteristics. The basis is the historical data. By applying the model on historical data, the model can be tested for aptitude an accuracy of predictions can be fine-tuned. Machine Learning technology can support this process, making the models continuously “smarter” and steadily increasing their prognostic power.

As a prerequisite, the IT infrastructure as well as the networks must be capable to handle high volumes of data. Internet of Things and Big Data are the main keywords in this regard. The harmonization of different types of data is of crucial importance to uncover hidden correlations between measured values and the propensity of defect. In this context Cloud technology offers some central advantages such as high scalability and ubiquitous accessibility via Internet.

Reply Value

With Brick Reply, Reply offers a comprehensive Manufacturing Operations Management Platform that integrates all equipment on the shopfloor and makes their data accessible for a wide range of appliances. Through the integration of Senseye, the condition monitoring data from the equipment can easily be analysed for predictive maintenance purposes.

Reply also supports industrial Enterprises through targeted innovation and service design, by developing new business models based on latest technology and through management consulting to successfully transform enterpises along the new paradigms of Industrie 4.0.

Reply can leverage extensive expertise and experience in all areas related to Industrie 4.0, such as Machine Learning, Cloud Computing, Data Science, IoT and Architecture, Cybersecurity, Rapid Prototyping, Augmented Reality and Smart Logistics.

Predictive Maintenance in practice

The higher the demands in terms of quality towards a product, the less tolerable deviations of production parameters become. A metalworking production site producing high precision components for automotive, pharma, chemical or medical industries can predict material flaws with high accuracy by closely monitoring the production conditions and the state of the production facility.

These insights allow for quick adjustments to ongoing production process and to suspend subsequent production steps to save energy if needed. On the other hand, correlations between machine performance and defect propensity become apparent. Maintenance can be scheduled accordingly to ensure compliance with previously defined thresholds. This proactive maintenance avoids cost-intensive, unplanned downtimes and contributes to predictive quality assurance.

The main benefit for the manufacturing industry consists in a higher Overall Equipment Effectiveness (OEE):

  • Higher availability factor due to more efficient planning of maintenance;
  • Increased product quality through faster identification and elimination quality deviations;
  • Reduction of warranty cases and rejections thanks to higher product quality.

New Service Models on the way to an integrated industry

For machine manufacturers the deployment of connected sensors opens up entirely new service models. Data exchange via the Internet of Things allows for predictions on the breakdown of wear parts. Service teams can be dispatched route-optimized before the breakdown occurs at the client site. This proactive service has some distinct advantages over the classical reactive approach: at no time has the client reason to be dissatisfied. Based on this, pay-per-performance service models can be developed – a customer centric approach for manufacturers to distinguish themselves in a competitive environment. The benefits for the machine manufacturer are:

  • Real-time optimized warehousing;
  • Increased efficiency in planning and carrying out service tasks;
  • Satisfied customers thanks to proactive service;
  • Reliable data basis for product improvements and innovation processes;

Interconnecting Manufacturer and end-users through predictive maintenance scenarios illustrates how to make a huge step towards an integrated industry. Data collected from the production process that are needed for predictive maintenance, offer further chances of advanced digitalization of processes in production and service management. This creates added value for manufacturer and user.