Development Services for Predictive Maintenance

Benefit from failure prognostics for your assets

Systematic development process for economically viable solutions

For many – if not most – promising predictive maintenance use cases the too simplified idea of automatically finding correlations in big data sets doesn’t keep its promises. Quite often the relevant data is not even available yet.

Significant saving potentials nevertheless justify the development of predictive maintenance solutions. We have specialised in supporting you with this in a structured process consisting of four phases: initial business case quantification, design and subsequent implementation of specific predictive maintenance principles, and finally its operational deployment.

Perform a first quick simulation of your value potentials jointly with us. Based on the viability of your use case we will be happy to offer our services for some or all of the following phases.


Our simulation approach quantifies scenarios for value potentials and de-risks your business case before launching a predictive maintenance project.


Our Hybrid Predictive Maintenance design method systematically delivers prediction models combining engineering and predictive analytics skills.


We integrate the required IoT hardware and software into your environment efficiently. Qualification and certification testing can be performed.


We support the entry-into-service of your solution with logistical support and troubleshooting. Once operative, we will optimise your maintenance strategy.

Phase 1 – Business Case Simulation

How to quantify benefits other than operational reliability?

Probabilistic agent-based simulation for quantification of benefit ranges
Seven value pools based on SAE ARP 6275

Often obvious benefits such as operational reliability are the tip of an iceberg. More often, predictive maintenance benefits for asset management are less straightforward and easy to quantify – as in the case of aircraft cabins, for example.

We have developed a probabilistic simulation able to quantify less obvious but often still significant potentials at more detailed levels. Our model in its second generation includes real context data and will compute different scenarios under your assumptions.

The SAE ARP 6275 provides guidance to the seven value pools identified by us: reduction of scheduled maintenance requirements, optimisation of spare parts pools, improvement of functional availability, amongst others.

Your business case decision for implementation of predictive maintenance candidates will be backed by quantified, realistic, and tangible benefit ranges.

Please contact us with a brief outline of your problem statement. For industries other than aerospace, we are happy to discuss an adaptation of our simulation technology.

Phase 2 – Systematic Design

How to develop predictive maintenance solutions without available data?

Hybrid Predictive Maintenance development method with five steps
Physics-based modelling combined with data-driven predictive analytics

An obvious approach to Predictive Maintenance is to perform analytics on big data sets discovering correlations. But how should you proceed if no data is available yet? And how do you define which data to acquire?

We have proven the success of a Hybrid Predictive Maintenance method combining physics-based engineering modelling with data-driven technologies. By simulation it allows to define sensor requirements and generate synthetic training data during design. Continuous machine learning will start with these and be refined once real sensor data becomes available.

Hybrid approaches are gaining increasing interest in relevant conferences such as organised by the PHM Society. Referring to ISO 13381, we are integrating physics-based and data-driven models in our method while heuristic and statistical models are used to complement informally.

Your use case will benefit from a high success confidence already during design. You’ll be able to exploit incoming data for predictions right from day one of deployment.

Contact us with a brief description of your use case in order to assess its feasibility. In case you’ve got data available already, we are of course happy to explore and build prediction models directly.

Phase 3 – Fast Implementation

How to make affordable predictive maintenance solutions real and scalable?

Pre-developed software brick with a focus on integration into your environment
Commercial IoT components for quick time-to-market solutions

Making a theoretical solution become reality is the next challenge. While constraints are different, both new developments and retrofit solutions require the implementation of hardware and software.

Our modular health management framework adapts efficiently to your use case with six layers ranging from data acquisition to prognostic assessments and potentially decision supporting advisory generation. This software is optimised for commercial Internet of Things hardware, ideally suited to affordably implement required sensors and connectivity.

The ISO 13374 functional architecture and the OSA-CBM data model form the basis of our flexible software stack for health management applications. By relying on the industry standards we secure a good level of interoperability.

Affordable implementations with quick time-to-market cycles will enable the realisation of optimised operations. It will provide you with a real competitive advantage, potentially also for your existing fleet as retrofit solutions. We can perform qualification tests for example complying with DO-160.

Discuss your use case with us either for your next new development or to explore opportunities for retrofit of your current products.

Phase 4 – Operations Support

How to finally realise savings of your operating costs?

Digital Twin concept for continuous optimisation of your maintenance operations
Outsourcing of the required computing and connectivity infrastructure

Finally operating predictive maintenance solutions in a way to realise your benefit potential is the ultimate objective. Besides operating the required infrastructure the main challenge is to derive optimal actionable decision proposals.

Using the initially created simulation model as a digital twin and feeding it with close to real-time status and prediction results, we are able to derive concrete maintenance strategy proposals.

Simulation-based optimisation is a common approach to dealing with complex nonlinear systems. It is able to compute optimal strategies by simulating different scenarios that take into account your planned mission profiles and predicted failure probabilities.

You will become able to conduct your maintenance operations in a way that best optimises your asset profitability.

Contact us to discuss your needs and how decision proposals can be integrated into your operations control centres.

Contact Us

I am interested in systematically developing predictive maintenance solutions for my use cases:

Error: Contact form not found.

Scroll to Top