What is Semantic PMI?

As more and more manufacturing organizations are implementing or considering model-based definition (MBD) initiatives, they are concerned about their return on investment. And it’s no wonder, as this effort carries costs, both in terms of effort and hard money. Incorporating semantic product manufacturing information (PMI) and model-based processes is one way to ensure success.

Building an MBD with Semantic PMI

When people first consider transitioning to MBD-based development processes, they tend to center their thoughts on design documentation. Meaning, they think of an MBD as simply a way to replace the traditional 2D drawing used in the product design process.

This is often the first step for many organizations as they move toward becoming a model-based enterprise (MBE). But it doesn’t come close to fully realizing the vast potential of MBD.

Companies stand to gain a great deal of value when they leverage MBD for automation and standardization needs further down the production pipeline. You can save time and money by adding semantic PMI or annotations. They document geometry in a way that can be consumed and interpreted both by other functional department workers and a variety of different software applications. That, in turn, saves significant time and cost in the product development process. It also opens the door to the use of model-based practices for downstream automation.

Here, we will discuss the role that semantic PMI plays in MBD, describe how you can use it in model-based practices, and examine the many benefits and advantages it can provide to product development.

What is Semantic PMI?

In traditional design documentation, an engineer adds pertinent data to a 2D drawing. That information likely includes geometric dimensions, surface finishes, geometric dimensioning and tolerancing (GD&T), and notes, just to start. Collectively, this essential data is the PMI. When engineering organizations transition to the use of an MBD, they add all of this information to a 3D model instead of to the 2D drawings.

PMI can be attached to many types of different geometry, and it isn’t always the mapping you might expect. For example, a flatness geometric dimension can be associated with an edge on a 3D model, even if it should be associated with a surface. On a drawing, these kinds of disparities might not make much of a difference. However, when you are spinning or interrogating a 3D model, this kind of misapplied PMI is confusing and can lead to misinterpretation, requests for clarification, or unintentional errors.

But with semantic PMI, all the design information is associated with correct and appropriate references. Semantic PMI eliminates the confusion of the previous scenario.

Why is this important? It all comes down to how you use an MBD in the product development process.

How is Semantic PMI Used?

Once the engineering organization starts to produce and release an MBD as the design documentation deliverable, other functional departments rely on the resulting models to do their own work. The simplicity that an MBD brings to this process is one of their most significant benefits. Employees in those downstream departments can simply open, view, and then interrogate the MBD for the data they need instead of struggling to interpret all the annotations in the drawing to understand design intent.

Make no mistake: There is considerable value in making the switch to an MBD just to improve design documentation. MBD is less ambiguous than a drawing. It gives others the ability to interactively view and interrogate the deliverable to get what they need. That can reduce or eliminate downstream errors, scrap and rework, and change orders.

But you gain even more value when you adopt model-based processes and annotate an MBD with semantic PMI. Then, software applications can read the semantic PMI to understand which geometric references are associated with each dimension, tolerance, note, or other piece of data, and automatically produce a deliverable. This saves considerable time.

For example, a computer-aided manufacturing (CAM) application can read the surface finish information associated with the geometry at the bottom of a pocket. The CAM system can then automatically produce a series of numerically controlled (NC) paths with the appropriate speeds, feeds, and step-overs to machine out the pocket while achieving the desired surface finish.

Of course, this sort of automation does not eliminate the need for human effort in the product development process. Someone still needs to review those toolpaths to verify everything works as expected. But the automation enabled by semantic PMI relieves those workers of some onerous manual tasks.

What is the Value of Semantic PMI?

You can see how using semantic PMI can accelerate tasks across the product development process. And in an age of ever-shrinking development schedules, successful automation is a boon for the entire enterprise.

There is just as much value, however, in using semantic PMI for standardization. Consider this: Any time someone views the drawing and manually enters parameters, like feeds and speeds for a tool path, there will be some variation. But semantic PMI and model-based automation allow companies to optimize production by defining and employing best practices to be used on a broader scale.

Using semantic PMI also lessens the potential for human error. As much as we may try to avoid it, there will always be a higher risk of input mistakes when you depend upon human operators. But automating tasks reduces those errors, saving precious time and resources.

Moving Toward a Model-Based Enterprise

There is tremendous potential value in MBD initiatives. They are a key component of design documentation improvements, to start. But to maximize their return on investment, companies should harness semantic PMI and model-based processes to facilitate the consumption of design documentation and automate manual tasks. In this way, you can reduce errors and minimize production delays.