Maximize Measurement Data Through Closed-Loop Dimensional Engineering

Keys to designing a strategy

Published: Monday, June 20, 2011 - 05:30

Across the life cycle of delivering a product to market, engineers face many obstacles. They often find themselves spending a lot of time reworking or repairing parts that fail inspection but may fit and function properly when assembled; or, they might pass inspection but do not fit with other parts and assemblies. Engineers also find themselves disagreeing with vendors concerning whether parts are made dimensionally correct or not, based on different understandings about design intent in regards to form, fit, and function. They must deal with team members who are intimidated when they see GD&T symbols on drawings, and grapple with high manufacturing costs due to tight tolerances on noncritical features.

Many of these problems are the result of an unclear definition of product and process requirements and a nonoptimal use of measurement data. The need has never been greater for a well-defined dimensional engineering process that enables the collection and analysis of relevant, meaningful variation measurements at every stage, from design through production.

Value of simulation-based tolerance analysis

Tolerance analysis is a powerful tool used to simulate manufacturing and assembly processes and predict amounts and causes of variation. It reduces the negative impact of variation on product dimensional quality, cost, and time to market

Tolerance analysis allows the user to visualize the variation for any component or assembly, consisting of any material, through virtual simulation. Literally thousands of assemblies can be built in a virtual environment and analyzed for dimensional integrity in just minutes. Engineers can look closely at variations, particularly in critical areas of clearance, interference, flushness, angle, and location.

Closed-loop process

Many leading OEMs benefit from tolerance analysis. However, to realize its full potential, it’s necessary to incorporate quality inspection data back into the tolerance simulation model and compare a simulation with reality. A suite of software solutions offered by Dimensional Control Systems Inc. (DCS) pulls together tolerance analysis software (3DCS) with quality management software (GDM-Web), enabling manufacturers to make timely and complete use of a vast amount of inspection data.

A comprehensive dimensional engineering process allows fit, finish, and function to be analyzed across the entire design-build life cycle. Multiple solutions can be tried through visual simulation, and then validated through automated, as-built reporting. This “closed-loop” system drives product quality all the way from design engineering to production.

The system looks like the closed-loop process in figure 1.

Figure 1: Closed-loop process 

A “closed-loop” quality management system enables engineers to correlate the theoretical tolerance analysis results produced during simulation to the actual as-built results determined at downstream stages of the quality process. Based on correlations, the as-designed simulation parameters can be validated, or they can be adjusted to more closely align to production process capabilities.

Additionally, a closed-loop system enables best practices to be validated, captured, and reused on future programs, for both engineering simulation and manufacturing. Leading OEMs are beginning to use closed-loop systems to improve their program efficiency as they continually face tighter and tighter development cycles and budgets. The information they gather through this system helps them discover ways to reuse existing and proven manufacturing process elements and modularized tooling on new and redesign programs—saving time and cost.

Real-life customer examples

Organizations that use inspection data through a closed-loop dimensional engineering process are able to reap benefits in cost savings and quality.

Chrysler has realized significant benefits over the years by making thorough use of its measurement data beginning in the early stages of its overall dimensional engineering process. “By applying tolerance analysis in the early design stages of our programs, and then comparing simulations with reality through a closed-loop process, we have been able to identify potential build issues early,” says Chrysler tolerance analysis manager Greg Medler. “This has enabled us to reduce physical prototypes, minimize costly gauge and tooling changes, and avoid quality issues during production.”

The EDAG Group is a major auto and aerospace supplier that maximizes its use of measurement data across its products’ entire life cycles. Management believes this is one of the most effective ways to reduce product life-cycle costs (PLCC). These savings result primarily from a reduction in scrapped parts; improved manufacturing uptime; reduced warranty defects; fewer reworked parts; and fewer reworked tooling, gauges, and drawings. EDAG tracked costs across many of its programs and noted that its PLCC could easily increase up to 10 times if the company did not maximize the use of inspection data.

Tail section of the Euro-fighter

With a simulation-based analysis, unlimited dimensional fit conditions can be identified quickly—a vast improvement over a model-based approach.

Figure 2: Model-based variation analysis 

The results from this simulation show that the output plots of the variation in the panel gaps look good. On the output plots of the fin tip variation (center), we see some issues at the upper and low limits of our spec. Based on these results, we can drill down to gather more detailed data.

Figure 3: Panel-gap variation and fin-tip variation 

Taking a deeper look at the panel gap variation at the left, measurements show a lot of unused tolerance. Holding this process capability closer than it needs to be is a waste of money. So, we’ve just identified an opportunity to save cost. We can optimize the use of data by applying tighter tolerance somewhere else in the assembly—a more critical area or one with particular challenges.

Robustness and timing

The Defense Advanced Research Projects Agency (DARPA) considers U.S. manufacturing capability a matter of national security, and “robust design” functionality is a key element of this capability. Tolerance analysis enables users to optimize designs and processes for dimensional integrity and robustness. Manufacturers can improve the robustness of designs and processes by maximizing part tolerances while maintaining dimensional requirements for final assembly.

Figure 4: Problem mitigation is least cost at this stage

Timing is a critical element to design robustness. Tolerance analysis allows engineers to evaluate design and assembly concepts up front, where problems can be identified early.

Maximizing use of data

In the real world, measurement data will often remain underutilized due to the time-consuming effort to “crunch” the numbers and root-cause the build problem. By pulling real inspection data or inspection data from legacy programs into a tolerance simulation model early in the product development cycle, engineers can make a more accurate assessment of the true source of variation and corrective actions can be simulated before manufacturing. Users will solve problems in a virtual world before they appear in the real world.?

Today’s manufacturers can truly “close the loop” on their overall engineering process, saving time and money.

About The Author

Donald Jasurda

Donald Jasurda has more than 30 years of engineering process improvement experience in the automotive, aerospace, medical device, and machinery industries where he worked on leading-edge projects ranging from mechanical artificial heart valves to composite commercial aircrafts, wind power generation and most recently, electric vehicles. Jasurda is the vice president of sales at Dimensional Control Systems Inc. (DCS), where he leads the sales and process transformation teams. DCS is a global provider of dimensional engineering consulting services and 3-D tolerance analysis and quality assurance software solutions that fully support the entire product life cycle.