Proven MBD Value Metrics

Since 2008, SOLIDWORKS has been a pioneer in supporting an industry practice and standard called Model Based Definition (MBD) and standards such as ASME Y14.41-2003 (updated in 2012) and Military-Standard-31000 (Revision A in 2013). The ultimate goal of MBD is to eliminate the costly and often error-prone drawing-based documentation still used to communicate design intent and manufacturing details. In retrospect, it is amazing that after almost 40 years since the introduction of Computer Aided Design we still rely on paper drawings to get things done. Part of MBD’s goal is to make all design details also readable by a robot, from the tolerances necessary to decide the correct manufacturing operations to the instructions to operate quality control robots (Coordinate Measuring Machines).

Figure 1: 3D PMI Definition on a Shaft

At a SOLIDWORKS WORLD 2016 MBD roundtable discussion, Glenn Kuenzler with GE Lighting (now Current powered by GE) asked for a side-by-side time and cost comparison between 2D drawing and MBD approaches. Glenn was not alone. We all understand MBD looks more intuitive and less ambiguous than 2D drawings, but so what? How does that translate into tangible benefits? The engineering community is looking for solid metrics to justify this paradigm shift and convince others in doubt. An independent and objective study by the National Institute of Standards and Technology (NIST) in the U.S. made solid progress in this study: Testing the Digital Thread in Support of Model-Based Manufacturing and Inspection. This blog will share a few exciting findings from NIST’s research. More details can be found in the report.

NIST worked with industrial partners such as Rockwell Collins to select three practical test case models (Figure 2, 3, and 4). Each test case was timed at three key steps in production:

Annotating design definitions;
Machining;
Inspection.

Figure 2: Test Case 1: A Hollowed-Rectangular Housing Requiring Milling Operations

Figure 3: Test Case 2: A Cylindrical Bearing Seal Requiring Milling and Turning

Operations

Figure 4: Test Case 3: A Connector Requiring Milling Operations

Here are several observations:

A model-based approach saved from 65% (Test case 3) to 80% (Test case 1) of hours spent in annotating, machining, and inspection. Table 1 shows the detailed numbers.

Beyond net hours, the model-based supplier delivered parts in approximately five weeks, but the drawing-based supplier took approximately eight months. The drawing-based supplier raised 12 questions related to interpreting the product definition from drawings, which led to work stoppages because the job had to be removed from the queue until clarifications were provided. In contrast, the model-based supplier asked no questions during its manufacture and inspection work.
Miscommunication of drawings led to quality issues such as this unintended through-hole in Figure 5 which scrapped the whole part. The MBD approach prevented it by taking the model as the authority. You may also notice the seal groove on the drawing-based part on the right doesn’t match the design in Figure 2, which was a minor issue, but does reflect another discrepancy in the drawing-based approach.

Figure 5: Comparison of a delivered test case 1 part showing the addition of an unintended through-hole in the drawing-based part

At the Test case 1 annotation step, MBD did take more hours than the drawing because the engineer had to learn to define intelligent (machine-readable) 3D PMI while being fully versed in 2D detailing already.
After this learning curve, on test case 2 and 3, MBD annotations were actually a bit faster than 2D drawings.

Now you may wonder why the MBD approach was so much faster. The NIST paper elaborates the reasons in detail, but in short, it was thanks to the machine-readable 3D dimensions and tolerances which were read by CAM and CMM to automate Numeric Code (NC) programming.

An MBD implementation blog explained machine-readable 3D dimensions and tolerances. Part 2 of this blog shared several integrated SOLIDWORKS partner products such as CAMWorks and CheckMate, which can read SOLDIWORKS MBD 3D annotations to automate machining and inspection.

As a summary, Glenn put this MBD paradigm shift into a resonating perspective: when we started to use computers, learning to type was challenging: it was slower to memorize each key location and type with all fingers than to look at the keyboard and “hunt and peck” with two index fingers. Now, of course, we are way faster with all fingers and it was worth the effort to learn it differently. The same applies to MBD here. At the initial annotation step, we may not see much time saving, but that’s just a preparation, or a prelude of a magnificent symphony in production: the real power of MBD kicks in at downstream automations such as CAM, CMM, Tolerance Analysis, Quoting, and Process Planning. To learn more about how SOLIDWORKS MBD can help you define machine-readable PMI, please visit its product page.