Blog

S[&]T corporation: Your partner in smart manufacturing

Science and Technology BV (S[&]T) is a SME developing cutting edge technology for complex systems. The company has built track record in space industry and since several years it is applying its innovation knowledge in smart manufacturing. Examples are central use of predictive analytics for real-time optimization and automated event handling with intelligent database and self-learning algorithm enabling impact analysis and decision support.

Below are some examples of previous projects that helped build the knowledge base:

Decision Support Systems

Time is a scarce and expensive resource aboard challenging scientific environments such as the International Space Station (ISS). The training and preparation of astronauts for onboard missions can take up a large amount of this resource, as do the actual maintenance, operation and troubleshooting involved with such missions. ESA has been researching more efficient, effective, and easy ways to realize human operations. S[&]T works on the following projects: ETECA (Expert Tool to Enhance Crew Autonomy), MECA (Mission Execution Crew Assistant) and TIDE (Technology for Information, Decision and Execution superiority). These technologies form the basis of decision support development in MANTIS.

GUI of Mission Execution Crew Assistant
GUI of Mission Execution Crew Assistant

System Health Management

SHM optimizes operation of complex systems by analyzing their health using either model-based methods i.e., using online sensor data and knowledge of normative system behaviour, or data-driven techniques, i.e., system behaviour is learned and extracted from large data volumes. Typical functionality includes: fault detection, diagnosis of system failures, and prediction of system failures. S[&]T has been involved in: ESA’s Future Launchers Preparatory Programme (FLPP), the real-time multiple sensor array LOFAR, and the development of a health management system for a Reusable Space Transportation System.

Rocket propellant with graphical analysis of the system health management
Rocket propellant with graphical analysis of the system health management
An example of system health management
Rocket propellant with graphical analysis of the system health management

Aim of S[&]T is to further develop its digital factory tools in a modular way. In this way specific analysis- and decision-support solutions can be created, dedicated per client.

Emerging Requirements for Future Manufacturing: the role of the new IoT/CPS based technologies

Internet of Things Applications in Future Manufacturing

The book chapter “Internet of things Applications in Future Manufacturing” is part of the 2016 IERC-European Research Cluster on the Internet of Things book (Digitising the Industry – Internet of Things Connecting the Physical, Digital and Virtual Worlds). This book is published by “River Publishers Series in Communications” that is a series of comprehensive academic and professional books which focus on communication and network systems. The book chapter is the result of a collaboration between John Soldados (Athens Information Technology), Sergio Gusmeroli (Politecnico di Milano) and MANTIS partner: Pedro Malo and Giovanni Di Orio (UNINOVA). You can download the book chapter in the Dissemination section of the web as well as the entire book by using the following link:

http://www.internet-of-things-research.eu/pdf/Digitising_the_Industry_IoT_IERC_2016_Cluster_eBook_978-87-93379-82-4_P_Web.pdf

 

Introduction — Future manufacturing is driven by a number of emerging requirements including:

  • The need for a shift from capacity to capability, which aims at increasing manufacturing flexibility towards responding to variable market demand and achieving high-levels of customer fulfillment.
  • Support for new production models, beyond mass production. Factories of the future prescribe a transition from conventional make-to-stock (MTS) to emerging make-to-order (MTO), configure-to-order (CTO) and engineer-to-order (ETO) production models. The support of these models can render manufacturers more demand driven. For example, such production models are a key prerequisite for supporting mass customization, as a means of increasing variety with only minimal increase in production costs.
  • A trend towards profitable proximity sourcing and production, which enables the development of modular products based on common plat- forms and configurable options. This trend requires also the adoption of hybrid production and sourcing strategies towards producing modular platforms centrally, based on the participation of suppliers, distributors and retailers. As part of this trend, stakeholders are able to tailor final products locally in order to better serve local customer demand.
  • Improved workforce engagement, through enabling people to remain at the heart of the future factory, while empowering them to take efficient decisions despite the ever-increasing operational complexity of future factories. Workforce engagement in the factories of the future is typically associated with higher levels of collaboration between workers within the same plant, but also across different plants.

The advent of future internet technologies, including cloud computing and the Internet of Things (IoT), provides essential support to fulfilling these requirements and enhancing the efficiency and performance of factory processes. Indeed, nowadays manufacturers are increasingly deploying Future Internet (FI) technologies (such as cloud computing, IoT and Cyber-Physical Systems (CPS) in the shop floor. These technologies are at the heart of the fourth industrial revolution (Industrie 4.0) and enable a deeper meshing of virtual and physical machines, which could drive the transformation and the optimisation of the manufacturing value chain, including all touch-points from suppliers to customers. Furthermore, they enable the inter-connection of products, people, processes and infrastructures, towards more automated, intelligent and streamlined manufacturing processes. Future internet technologies are also gradually deployed in the shopfloor, as a means of transforming conventional centralized automation models (e.g., SCADA (Supervisory Control and Data Acquisition), MES (Manufacturing Execution Systems), ERP (Enterprise Resource Planning)) on powerful central servers) towards more decentralized models that provide flexibility in the deployment of advanced manufacturing technology.

The application of future internet technologies in general and of the IoT in particular, in the scope of future manufacturing, can be classified in two broad categories:

  • IoT-based virtual manufacturing applications, which exploit IoT and cloud technologies in order to connect stakeholders, products and plants in a virtual manufacturing chain. Virtual manufacturing applications enable connected supply chains, informed manufacturing plants comprising informed people, informed products, informed processes, and informed infrastructures, thus enabling the streamlining of manufacturing processes.
  • IoT-based factory automation, focusing on the decentralization of the factory automation pyramid towards facilitating the integration of new systems, including production stations and new technologies such as sensors, Radio Frequency Identification (RFID) and 3D printing. Such integration could greatly boost manufacturing quality and performance, while at the same time enabling increased responsiveness to external triggers and customer demands.

Within the above-mentioned categories of IoT deployments (i.e. IoT in the virtual manufacturing chains and IoT for factory automation), several IoT added-value applications can been supported. Prominent examples of such applications include connected supply chains that are responsive to customer demands, proactive maintenance of infrastructure based on preventive and condition-based monitoring, recycling, integration of bartering processes in virtual manufacturing chains, increased automation through interconnection of the shopfloor with the topfloor, as well as management and monitoring of critical assets. These applications can have tangible benefits on the competitiveness of manufacturers, through impacting production quality, time and cost. Nevertheless, deployments are still in their infancy for a number of reasons including:

  • Lack of track record and large scale pilots: Despite the proclaimed benefits of IoT deployments in manufacturing, there are still only a limited number of deployments. Hence manufacturers seek for tangible showcases, while solutions providers are trying to build track record and reputation.
  • Manufacturers’ reluctance: Manufacturers are rather conservative when it comes to adopting digital technology. This reluctance is intensified given that several past deployments of digital technologies (e.g. Service Oriented Architectures (SOA), Intelligent Agents) have failed to demonstrate tangible improvements in quality, time and cost at the same time.
  • Absence of a smooth migration path: Factories and production processes cannot change overnight. Manufacturers are therefore seeking for a smooth migration path from existing deployments to emerging future internet technologies based ones.
  • Technical and Technological challenges: A range of technical challenges still exist, including the lack of standards, the fact that security and privacy solutions are in their infancy, as well as the poor use of data analytics technologies. Emerging deployments and pilots are expected to demonstrate tangible improvements in these technological areas as a prerequisite step for moving them into production deployment.

In order to confront the above-listed challenges, IoT experts and manufacturers are still undertaking intensive R&D and standardization activities. Such research is undertaken within the IERC cluster, given that several topics dealt within the cluster are applicable to future factories. Moreover, the Alliance for IoT Innovation (AIOTI) has established a working group (WG) (namely WG11), which is dedicated to smart manufacturing based on IoT technologies. Likewise, a significant number of projects of the FP7 and H2020 programme have been dealing with the application and deployment of advanced IoT technologies for factory automation and virtual manufacturing chains. The rest of this chapter presents several of these initiatives in the form of IoT technologies and related applications. In particular, the chapter illustrates IoT technologies that can support virtual manufacturing chains and decentralized factory automation, including related future internet technologies such as edge/cloud computing and BigData analytics. Furthermore, characteristic IoT applications are presented. The various technologies and applications include work undertaken in recent FP7 and H2020 projects, including FP7 FITMAN, FP7 ProaSense, ECSEL MANTIS, H2020 BeInCPPS, as well as the H2020 FAR-EDGE initiative. The chapter is structured as follows: The second section of the chapter following this introduction illustrates the role of IoT technologies in the scope of EU’s digital industry agenda with particular emphasis on the use of IoT platforms (includ- ing FITMAN and FIWARE) for virtual manufacturing. The third section is devoted to decentralized factory automation based on IoT technologies. A set of representative applications, including applications deployed in FP7 and H2020 projects are presented in the fourth section. Finally, the fifth section is the concluding one, which provides also directions for further research and experimentation, including ideas for large-scale pilots.

User-friendly interfaces: Treating data sets like recommendations from a friend

Data from smart, connected devices and related external data may come in an array of formats, such as sensor readings, locations, temperatures and history. In order to understand the patterns, the spreadsheets and database tables need to be suited to user’s needs.

Ideally, the interface would not require the user to undergo training in order to be able to use it. As far as text is concerned, it is crucial that the messages avoid ambiguity, inaccuracy, inconsistency and inadequacy in order to ensure more safety, fewer errors and higher productivity.

To capture and display maintenance-relevant information, MANTIS is designing user-friendly industry dashboards for control rooms combined with mobile extensions on the go. As MANTIS use cases vary from railway maintenance to healthcare equipment maintenance, strongly dependent on individual technical constraints, there is no common, specific MANTIS Human Machine Interface (HMI) design.

However, as central feature of most design use cases is displaying data of monitoring process and responding to alarms, parallels can be drawn and best use practices can be used in MANTIS.

E.g. presenting information such as asset maintenance history, the overall condition, sensor data and analysis was a task assigned to XLAB in several other past and current projects. We describe how feedback gained from two successful past projects – the spin off company Sentinel Marine Solutions and FINESCE smart buildings – helped draw up guidelines. Special emphasis was placed on user experience (UX), creating a relationship with the user over time and on how he feels during the whole lifecycle of the product.

The IMPACT-accelerated startup and FIWARE technology implementer Sentinel Marine Solutions follows the overall state of the boat and allows real time access to sensor data to avoid e.g. batteries losing charge. Crucial boat parameters, water level, temperature readings, battery voltage, bilge pump, are presented on a user-friendly interface – tablet, smartphone or dashboard in the control room or cabin (Image below).

Monitoring is done continuously, providing rich information about one or several vessels’ position, voltage, temperature – where ever the user may be, on land or off shore. That is why Sentinel keeps it simple, to avoid cluttering and confusion, and complements it with a visually pleasing design for sustained usage of application.

HMI battery status maintenance
HMI battery status maintenance
hmi sensors maintenance
hmi sensors maintenance

Work on previous research projects confirms that it is worth investing a significant part of the application development into how the system presents data to the end users. Within the FINESCE project, the top part of the presentation layer was designed by XLAB. The interface connects data on energy generation and consumption in smart factories and smart buildings. It had to be optimized for quick retrieval and display in order to identify deviations and enable fast responses.

The application has many views showing specific aspects of the data, but each in a context that is relevant to the workflows of end users – below an example of the Smart building consumption Overview tab.

Last but not least, the aim of the design of these applications was to present the data to the users with a pleasing appearance, while at the same time offer crucial visual information on the current and past energy consumption.

Early detection of fissures in industrial structures

Industrial sectors such as automation, energy and mechanical manufacturing are more increasingly in the need of including predictive maintenance techniques and processes, in order to improve their product quality and reduce their maintenance costs.

In such scenarios, fatigue is a cumulative phenomenon that appears when material is subjected to repeated loading and unloading. When this happens and the target is stressed beyond its critical threshold, microscopic cracks begin to form, which will eventually end fracturing the structure. Thus, smart sensors must be placed in the critical zones, in order to early monitor the growth and evolution of the cracks.

Solution approaches

As it is described within WP3’s use case 1.3 in MANTIS project, which focuses on developing a framework for smart sensing and data acquisition technologies, there are several detection techniques for direct crack measurements, such us regular strain gauges and crack gauges. These last elements are made of aligned grids which are disconnected one by one with the propagation of the crack. The drawback comes in terms of placement, as the location of the failure is not always predictable and therefore it would require a complex multi-gauge installation, in order to cover a large area of sensorization and anticipate the formation of cracks.

Another approach which is being analyzed and tested within use case 1.3, is the utilization of conductive inks, which could give more flexibility in terms of placement and the design of the sensorization area to be monitored.

Installation requirements

If the fissure detection is performed with conductive inks a necessary requirement must be taken into account. As the structures that are monitored are electrically conductive, an insulation layer must be deposited between the structure and the conductive ink. For the efficient detection of fissures, this insulation layer should break with the structure, but it must not brittle with time, temperature, humidity, etc. Even more, it should be easily deposited, as the structure may be located in a difficult to access place.

According to the conductive ink, it should also be of easy deposition, with low resistivity and withstand high temperatures without breaking.

Current tests

In the preliminary tests, a Magnesia paste based insulation layer has been used and on top of it, a vinyl mask layer has been stuck for the definition of the conductive layer structure. As conductive ink, a low resistivity silver ink has been used, defining two structures: a gage structure, Figure 1, and continuous conductive line.

Conductive ink structures
Conductive ink structures

During the fissure measurements, the gage structure has been supplied and measured, recording both current and voltage. As it is shown in Figure 2, as the lines of the gage structure break, an increase in the measured voltage and resistance is recorded.

Fissure detection measurements
Fissure detection measurements

Thus, properly deposited and structure adapted conductive inks could stand as a solution for the early detection of fissures in industrial structures.

Nowcasting and Forecasting tecniques for railway switches maintenance

A large proportion of the costs of the European railway Infrastructure is related to maintenance and the current situation of railway maintenance .

The increasing usage of the railway infrastructure, due to a growing frequency of passenger and freight trains on it, and environmental and safety regulations, the increasing of maintenance requirements cannot be met without a substantial shift in maintenance strategies.

In this prospective, the new “Proactive/Predictive Maintenance”  approach will involve/have an impact all the main railway stakeholder on long-term preservation of assets (expressed in RAMS requirements) minimizing life cycle costs, while for the railway operators, the proactive maintenance will improve the railway infrastructure availability and reliability.

Switches and Crossings (S&C) are fundamental infrastructure assets that allow efficient routing of trains on the network. Assessing their current status and ensuring their proper functionality is obviously a key requirement to guarantee the railway infrastructure availability for the railway operators. For these reasons, Ansaldo STS tries to discover in the Mantis project, references concerning railway S&C condition monitoring based on a wide range of methodologies from the world of statistics, data mining, time series analysis, machine learning, and filtering.

Given the examined literature, it seems clear that the use of the word “nowcasting” can be associated with: a shorter timeframe respect to “forecasting”, a different approach or algorithm for performing the estimation of the value of interest, the fact that the data used for the estimation is imprecise, uncertain, incomplete or is only indirectly related to the phenomenon of interest, and, finally, with the purpose of providing an alert for a sudden event or a possible anomaly.

A reasonable simple definition could be adopted in the framework of the Mantis project for differentiating “forecasting” and “nowcasting” processes:

Forecasting:

The process of exploiting past and present data to make deductions about the future.

Nowcasting:

The process of exploiting past and present uncertain or incomplete data to make deductions about the present.

Many of the failures are not detected until the asset is being operated by the interlocking system when trying to lock the train route. This means that for some failures, nowcasting cannot be done before operating the unit. Frequent test procedures could solve this issue but it has other disadvantages like the introduction of increased wear, cost and reduced inherit capacity. Another solution for this problem is to introduce additional condition monitoring systems for detection of different states of critical components in the S&Cs.

Estimating the remaining component lifetime – a key activity of of MANTIS

A key activity of the MANTIS WP4 work on analysis and decision making functionalities has focus on identification and development of methods for forecasting the remaining useful life of an asset by modelling it’s deterioration and by extrapolating this into the future. When the extrapolated deterioration reaches a certain threshold that marks the end-of-life of the asset, and that point in time can be used as the forecasted remaining useful life (RUL). Simultaneously, the extrapolated deterioration gives an indication of the expected wear and tear, which can be compared with the observed wearing out to monitor the behavior of an asset.

Based on the predicted remaining lifetime, suitable maintenance tasks can be planned, in order to prevent unscheduled system down time.

 

Overall working strategy

The approach followed by MANTIS to address the estimation challenge is as follows:

  • Develop algorithms and train models to estimate the RUL or predict the wear and tear of an asset.
  • Monitored data (from embedded sensors or from quality assurance checks) is streamed to the cloud based storage system, where it will be investigated using time-series analysis or temporal data mining approaches.
  • These methods will be used to discover revealing relationships between parameters to identify significant patterns, trends and anomalies.
  • Based on this analysis of the historical data, a diagnostic model will be made self-learning, comparing predicted and actual data.
  • Flexible modelling techniques will be used on diverse information, such as condition parameters, data patterns, sparse data and/or existing expertise to learn more complex relationships between the different parameters or train models which allow prediction of future performance in terms of expected failures.

 

Current status of work

An overview of relevant deterioration models and methods to estimate the remaining useful life (RUL) has been made. Also, for each use case, the present status with respect to available data, deterioration models and methods for estimation of RUL has been collected. In particular, for each of the use cases, the following information is relevant:

  • Description of data typically available
  • Description of relevant deterioration processes
  • Description of models used for estimation of RUL

 

Next steps

During the next 6 months, the general models and methods will be analysed and compared with the present status for the individual use case (available data, deterioration process, RUL estimation models). Following this analysis, specific models and methods will be selected and unified as much as possible, and thereafter applied and validated on the use cases during the second half of the MANTIS project.

This will be achieved by using a bottom-top-bottom approach:

  • Collecting input for individual use cases
  • Seeking for commonalities and unification into a generic approach
  • Applying the generic approach on the individual use cases

Maintenance data analytics; How to deal with free text

Problem description

For analyzing maintenance problems, such as determining the root cause, in general three sources of information are available:

  • sensor data
  • machine generated logging messages
  • service and/or maintenance engineer reports

In contrast to sensor data and machine generated logging messages, of which the format is determined at design time, the service and maintenance reports have a free format, making it hard to analyze automatically.

Problem reporting
Problem reporting

Possible solutions

Before going into possible solutions, a better description of the problem is needed. Free format is this case means that a report can contain

  • unstructured text
  • ambiguous descriptions e.g. mixing commercial and technical identification
  • multiple languages e.g. using local language
  • spelling errors, ad hoc abbreviations, missing data etc.

For easy analysis it would be good if it is possible to bring free text back into the predefined format realm. Ways to solve this are

  • using predefined forms with e.g. dropdown lists for selecting options
  • using intelligent text editors that can recognize the vocabulary needed for describing the specific problems
  • combinations of the above
  • less stringent is specifying rules people have to abide by when entering maintenance logs (typo’s are still possible then)

As it is impossible to predict all problems that will be encountered during the lifetime of a machine, there still needs to be the option to enter free text, but it should be clear that this is only to be used when other options do not cover the problem at hand.

Conclusion

It is clear that free text will always be used in reporting, but in order to facilitate data analysis of service and/or maintenance reports computer aid or rules can make life a lot easier.

Research in WP5: HMI design and development

WP5 in MANTIS-project focuses to study on implementing an efficient intelligent human machine interaction to deal with the intelligent optimization of the production processes through the monitoring and management of its components. The overall objectives of WP5 is to develop HMI capable to contribute to:

  • Enhanced monitoring of shop-floor conditions
  • Automatic self-adaptation of control strategies
  • User-friendly, ergonomic and intuitive interaction between workers and machines

Work in WP5 started with the identification of human machine interaction scenarios from each eleven use-cases. When defining requirements for HMI from different use-case point of views, collaborative and proactive aspects were highlighted. Even though the use-cases are very different, common elements and functionalities from the requirements for HMI were identified. Figure 1 presents the overall approach towards WP5 when designing and developing HMI prototype.

Overall approach towards WP5
Overall approach towards WP5

At the moment work in WP5 focuses on researching different platform solutions that would serve the HMI needs specified in use-cases. The objective is to develop a platform solution that will serve the common HMI needs identified from use-cases. Initial discussions among the WP5 participants relates to building a Web-based maintenance information portal that will act as a prototype HMI, which could be tested in selected use-cases. WP5 partners have discussed the potential approaches over the last period and the implementation plan have just been agreed on. Figure 2 is an example view what maintenance information portal could look like consisting of user specific content elements.

Example view of maintenance information portal
Example view of maintenance information portal

We need to have in mind that the information portal should behave as a support for the users not as their replacement. Maintenance information portal could be:

  • a single gateway to a company’s information and knowledge base
  • a framework for integrating information, people and processes across organizational boundaries
  • a secure unified access point
  • designed to aggregate and personalize information

Discussions about importance of three concepts in HMI-development has also been active. These concepts are collaborative, proactive and context-aware functionalities. Collaborative functions in HMI could be the possibility to communicate through the portal, share information and UI-views with different actors involved in certain maintenance tasks. These actors can be within the organization or external service providers. Proactive functions can be automated, model based calculations that predicts maintenance needs early enough that required maintenance actions can be carried out to prevent unplanned down time. Context-aware functions could for instance alter user defined UI-views in certain scenarios such as emergency situations or changes in the state of the production process. UI-view could also change depending on physical location of the portable mobile device connected to MANTIS-platform.

Strong presence of predictive maintenance technologies in Bilbao’s international machine-tool definition

Held in Bilbao from 28th May to 1st June, the 29th edition of the Machine-Tool Biennial (BIEMH) gathered a total of 40,000 people, representing a clear success and a gradual industrial recovery indicator. Predictive maintenance has constituted a trending topic (as Industry 4.0 pillar) within the fair, specifically threated by MIC (Maintenance Innovation Conference), playing several MANTIS partners (FAGOR Arrasate, IK4-Ikerlan, IK4-Tekniker) a key role during the BIEMH. Specifically related to maintenance, FAGOR Arrasate presented the wide range of applications and services derived from the advanced connectivity their products bring in (e.g. check of operating functions and potential machine utilization improvements).

FAGOR ARRASATE connected manufacturing lines
FAGOR ARRASATE connected manufacturing lines

Moreover, DANOBAT Group (belonging to MONDRAGON Corporation) presented Dynamics Active Stabilizer, a device capable of actively increasing the dynamic rigidity of the machine tool, reducing the risk of chatter during the machining process and thus increasing the cutting capacity by up to 300%. This represents a clear example of how process data real-time processing (and analysing) can have a significant impact on productivity and production costs. DAS (which received the Quality Innovation of the Year award, organized by the Finish Quality Association) improves 100% capacity through the complete workpiece volume, increases productivity up to 300%, improves workpiece surface quality, extends the tool life and operates in real-time.

It was eventually announced that next BIEMH will take place on May 28-June 1, 2018 (Monday to Friday), which means it will be one day shorter than this year’s event.

DANOBAT GROUP DAS system
DANOBAT GROUP DAS system

On the whole, ICT enabled predictive maintenance states one of the most promising revenue streams machine tool builders have (hence its strong presence within this international fair). However, there are still several technological and non-technological challenges to be faced, being most of them tackled by MANTIS.

Usage of the MANTIS Service Platform Architecture for Health equipment maintenance optimization.

Healthcare Imaging Systems of Royal Philips N.V. are essential for the diagnosis and treatment of patients in hospitals and private clinics. Due to the large costs involved it is not economically feasible to implement backup systems. Therefore system uptime has to be maximized, planned downtime has to be minimized and unplanned shutdown has to be prevented. To cope with the exploding cost of healthcare, the cost of ownership has to be reduced, which also implies that maintenance budgets are under pressure. In response Philips Healthcare has developed maintenance services for hospitals based on remote monitoring of their systems.

Unplanned system shutdown has a large impact on patients and hospital staff
Health equipment used during patient treatment

The main challenge is to retrieve, store and analyse large amounts of data from globally distributed systems such that predictive maintenance can be offered instead of maintenance at fixed time intervals. Furthermore an alerting system is necessary when the online big data analysis detects a threat of shutdown.

Due to the large purchase cost and the cost of housing unplanned shutdown has a large impact on patients who do not get the care and on the hospital. The Healthcare Imaging Systems of Royal Philips N.V. will use the MANTIS Architecture for equipment asset optimization, thereby aiming to move from a reactive to proactive and predictive maintenance.

Main challenge: getting from large amounts of data to accurate and precise failure detection and prediction.

The objective is to accurately predict upcoming failures by mining large amounts of data from heterogeneous systems distributed globally, such that maintenance can be timely scheduled or in urgent cases the responsible person can be alerted.

Graphic depiction of Health equipment maintenance use case
Graphic depiction of Health equipment maintenance use case

Every Healthcare Imaging Systems of Royal Philips N.V. contains many sensors and generates large log files daily. Since these systems are heterogeneous by nature the first challenge to address is to optimize logging such that data mining success can be optimized (anamnesis). The next challenge is to make all data available worldwide in the cloud. Once the data is centrally available it has to be translated to behavioral models and consolidated in a limited set of relevant parameters (translation). This translation requires significant computing power and storage space (infrastructure). Next, the obtained parameters have to be analysed with respect to the maintenance challenges (analytics) and the results have to be visualized for end-users (visualization).