Stefan Riesen
ERNI Switzerland
Medical device complexity is increasing due to an increased amount of features, including ones that are not directly related to the medical diagnostic function, e.g. connectivity, predictive maintenance, and optimisation algorithms. Additionally, the market expects continuous improvements and innovation as well as short reaction times in case of unexpected system behaviours over the full lifecycle of a product. This increased complexity raises the product cost in development, production and maintenance. In order to keep the cost increase as low as possible, a good understanding is required of the systems of all parties involved in the development, production and support of a product, based on consistent, accessible and reliable information.
In a traditional document-centric setting, the increased system complexity brings some shortcomings in regard to the aforementioned information requirements. Firstly, the information in the documents is not linked and spread among multiple documents, thus being potentially inconsistent due to a missing single source of truth. Secondly, it is usually based on natural human language, which is prone to misinterpretation. To eliminate these shortcomings, Model-Based Systems Engineering (MBSE) comes into play.
MBSE in a nutshell
MBSE is a systems engineering methodology that focuses on creating and exploiting domain models as the primary means of information exchange, rather than document-based information exchange. This model acts both as a single source of truth and as a collaboration platform, as it is shared between all involved parties across various R&D disciplines, manufacturing operations, marketing and field support.
At the core, the system model is created in a formalised modelling language, such as SysML or mathematical models, thus overcoming the potential misinterpretation of information written in a natural language.
Further, this model contains all the information, starting at the intended use and requirements on all levels; followed by the corresponding design data, like SW architecture, interface specifications, or electromechanical design and manufacturing data; and concluded by verification and validation data such as test cases and clinical validation data. As not all the information is needed for every activity, views on the model are created that show only the currently relevant information. These views may, for example, be all user workflows so that they can be discussed within the UX Team or the CNC data of a part so that it can automatically be sent to manufacturing. Therefore, there is no inconsistent information.
The right time to start with MBSE
Although MBSE has some significant advantages, it comes at an initial setup cost. This includes the training of all involved parties, the setup of the infrastructure (including the validation of the tool chain), and the adaptations of processes. Additionally, in the MedTech domain, the appropriate views must be created that can be exported to documents to be submitted to the notified bodies and regulators. Therefore, the biggest return on investment is given when MBSE is introduced at the beginning of a product lifecycle. Also, with each product that is added to the MBSE-related portfolio of a company, the initial costs are lowered. However, the gains outweigh the investments in the long run due to the advantages of MBSE related to a faster time-to-market and the reuse of data, thus giving a positive return on investment.
What cannot be neglected is the fact that the model needs to be populated. But the effort to create a model is considered equivalent to creating the corresponding document in a first step. For example, if the user workflows shall be defined, the UX and requirement engineering teams still need to discuss with potential system users what their needs and desires are and what a workflow could look like. The creation of the model is the smaller part of that activity. However, a model gives the UX and RE teams the appropriate tools to design the workflows. Further, it provides the platform for the SW engineering teams to link their designs and the V&V teams to link their testing activities directly to the workflow previously defined by the UX team.
As illustrated in this example, MBSE can start relatively small within the R&D department between the design input, the design, and the V&V team in order to create a basis to onboard additional departments in the course of the project.
As a conclusion, one can derive that MBSE is an appropriate methodology in a MedTech environment when applied from the beginning of a product lifecycle and brings the following advantages over a traditional document-centric approach:
- Lower long-term product lifecycle costs
- Better collaboration between all involved parties
- Better system understanding from all parties based on a single source of truth
- Faster time-to-market
How we as ERNI can provide support and what business benefits the customer gets
ERNI can support you in the implementation and use of MBSE as our Requirement Engineers, UX Experts, SW Developers and Testing teams are trained in formal modelling languages (next to their core expertise in the respective fields) and can easily support your product development in an MBSE context. Our Business Analysts and Project Leaders are experts in setting up the appropriate development environment in order to facilitate an implementation of Model-Based Systems Engineering.
