Traditional engineering methodologies like the waterfall model rely on a sequential order of work. As is now commonly accepted, this methodology inhibits an organization’s agility because it does not easily accommodate or tolerate change. Yet changes are a common and necessary part of every development project.
That is why the Agile approach, revolving around continuous and concurrent short-term work in sprints, has risen in popularity. To enable an Agile approach, engineers and others within an engineering organization must be able to work from the same information (or better yet, the same model) simultaneously.
MBSE is an Agile Prerequisite
Collaborating and communicating about a system is difficult unless everyone involved is working from a single model. As such, collaborative Model-Based Systems Engineering (MBSE) is the prerequisite for agile, iterative engineering because it provides a common system model or set of system models. More simply, it makes the system model accessible and editable by engineers and others from different domains of expertise. It forms the basis to express ideas, make decisions and ask questions of the model. This eliminates the need to start from the top, so to speak, continually requesting input and information from design engineers.
Enabling a Collaborative Approach
A system model enabling an Agile approach is essentially the construction of different perspectives on one model in which it’s easy to follow relationships between the views and modeling items. To achieve this, organizations build up a network of model elements and perspectives that relate to other artifacts and are organized in these four layers:
This network makes it possible to eliminate many of the traceability artifacts that are manually created and maintained within the traditional layered, sequential approach to engineering:
MBSE in Action
Instead of going through the traditional project workflow to identify and validate ideas and changes, stakeholders work from a network of elements, either in the model itself or within a sandbox of the model. Changing elements in the network allows them to see if those changes work in correlation with the model and how they impact the rest of the system architecture. At any time, a stakeholder can determine if an idea works in conjunction with existing model elements, and even add to the model.
The MBSE approach works because it provides a way for project stakeholders to easily see the relationships in and between all four layers in the model. It’s no wonder we are seeing more organizations (including PTC) and universities exploring new methodologies for working with this four-layer approach. One such methodology is known as Software Platform Embedded Systems (SPES). According to a SPES project sponsored by the German Research Ministry, “Scientific studies show that the use of end-to-end model-based techniques, as developed in the SPES consortium, can significantly reduce product development costs (on average 27% - 34%).” That’s because this approach and these techniques can help avoid common problems such as under-specification of system behavior or contradictory specifications.