There are significant quality and cost benefits when tolerance analyses (stack-ups) are detailed, complete, and most of all reliable. Software technology to perform three dimensional tolerance analyses has been around for over twenty years and offers significant improvements in obtaining meaningful results that can reduce cost of development, recurring costs (COGS), improve quality, and do it all faster. However, there is a road block in using 3D tolerance analysis and that is culture change.

Culture change remains the most significant issue when developing new lean and innovative processes within new product development organizations. But change we must in order to maintain a competitive edge. Pressure continues to bear down on designers and engineers to develop more robust design, get to market faster, and not just keep costs down but reduce costs. Well, we have been trying to do that for years and there are many programs put in place, involving significant investments, to achieve the benefits of better, faster, and cheaper.

Programs like the popular six sigma process and design excellence initiatives have really helped to change culture in lean development and production while putting more focus on voice of the customer. These initiatives have been working and have accomplished some significant improvements in product development. However, mention GD&T and the need for training and the scramble is on for reasons why the training is either unnecessary or otherwise keep the training to a minimum investment. Fundamental GD&T training is hardly enough to perform reliable detailed tolerance stack-ups. But, any discussion on advanced training in GD&T or stack-up analysis is often tabled for another time.

So, it is no wonder that three dimensional tolerance analysis still has not caught on. Organizations typically feel that whatever GD&T practices they have been using for years is working and the same is true with hand calculated 2D tolerance stacks involving GD&T. The questions are; "Is the application of GD&T and tolerance Stack-ups within the organization reducing costs and improving quality the way GD&T is intended to do?" Or, is inappropriate use of GD&T and unreliable tolerance stacks costing time, money, and quality? I believe the later of the two is the reality for many organizations.

Replacing 2D hand calculations with advanced 3D software is no different then the culture change that took place to accept 2D CAD and then 3D CAD over the manual drawing table. But there is a difference. People were comfortable with manual 2D drawings and uncomfortable with CAD. For 3D tolerance analysis, we have a situation where people are already uncomfortable with GD&T and more uncomfortable with stack- up analysis using GD&T. The idea of going 3D with tolerance analysis is just another world and makes people more uncomfortable. Well, that's right. 3D tolerance analysis is another world; a better one.

The discomfort is obvious in even a fundamental class on GD&T. With a group of experienced engineers who have approved dozens or hundreds of drawings that use GD&T, an instructor can ask a basic question and the class remain silent. Ask for a volunteer to do a relatively simple stack and again no hands. So, no wonder designers and engineers hide behind present day practices. After all, present day practices for tolerance analysis typically do not include a check and approval cycle. The irony is that using a good 3D tolerance analysis tool can actually make designers and engineers comfortable with not only the GD&T but with the results too; Results that are truly value added and help provide more robust design, reduced development costs, reduced part and inspections costs, reduced tooling changes, and reduce time to volume to mention just a few benefits.

There is another culture issue. When companies are pushing for more virtual prototyping with CAE in finite element, thermodynamic, fluid dynamic, and dynamic mechanism analyses, three dimensional variation system analysis (tolerance and assembly analysis) keeps getting the lower priority in the product data lifecycle. Why? Well in my college days, I took thermodynamics, fluid dynamics, strength of materials, and the rest of the typical mechanical engineering courses. But, tolerance analysis was not required. In fact, it was not even an elective. I was in engineering for three years before I even heard of GD&T. So, who normally does stack analysis in many companies? The designers are doing hand stacks because the engineering culture is unknowingly taught that tolerance analysis is somehow less significant then the other CAE analysis in use today. I mean who cares if the round peg does not fit into a square hole as long as it is strong and aerodynamic? My point is that when management listens, management listens to the requests of engineers, and since tolerance analysis is commonly not an engineer's responsibility to do (only that it gets done) engineers are focused on the other CAE tools. The culture issue is that designers are often not on an even level of an engineer in a corporate hierarchy.

I know this is a bold statement. But it is simple to test the perceived importance of tolerance analysis within a company. Ask these questions. Is GD&T used? Do persons defining requirements with GD&T have certifications such as the ASME GD&T Professional technical and/or senior level certifications or some other real world equivalent experience? Do associates typically attend a three day on site course within which they sleep, answer cell phones and e-mail, and stretch ten minute breaks into thirty minutes? Don't get me wrong. I know professionals, very good professionals that offer these courses. The irony is that an on site course is to reduce time and cost of training when it is the knowledge of GD&T that is supposed to reduce time and cost of producing higher quality product delivered to the customer. Who, if anyone, checks the tolerancing for correct application? This is not the same as signing off on a 2D drawing. That is a whole other issue. Who checks the tolerance stack-up and what are his or her credentials to do so? A personal favorite; "Is the tolerance analysis a requirement in any of the stage gates within a design excellence program and is the analysis put into a release cycle for an approval signature and design control?" And my all time favorite; "Are the black belts, in whatever process excellence program your company may have, required to take even one class, much less be certified, in GD&T?"

I know these questions can hit some sore spots; rock the boat. But now go to the people who assemble and test product on production lines. Go to the people who have to inspect components and ask them; "what is the single biggest issue when production assemblies either do not assemble or otherwise do not function correctly?" It almost always comes down to dimensional controls. In particular, production issues come down to not having "knowledge" of specific contributors and therefore impact studies on tolerance deviations are almost useless and home grown tolerance stack-ups are either unavailable or so unclear that the analysis can not be read (sometimes not even by the author of the analysis), or worse the analysis is plain wrong. We know there are huge costs associated with assembly line shut downs, recalls, and customer dissatisfaction, and we know that these costs can be associated in whole or in part to tolerances specifications that are simply not understood.

There are many benefits for implementing a 3D tolerance analysis tool like the UGS Teamcenter Visualization VSA product known as VisVSA. When used early on in the detailed design phase of new product development, it can help reduce the number of physical prototypes just as other CAE processes like FEA. Without valid tolerance analysis prototype parts are received and changes are physically made to features to get prototypes parts to assembly and/or function. Often, the dimensional issues result in prototype tool changes resulting in additional physical prototypes, time, and money just to get a design assembled and functioning for further verification and validation. With VisVSA 3D tolerance analysis, thousands of virtual engineering builds can be performed using Monte Carlo simulations and tolerance iterations can be performed quickly based on each set of results to optimize the design for function with optimal tolerances that allow for easier lower cost manufacturing with better capability to clear, concise, and valid requirements.