Robust Design and Variation Analysis

• Improve quality and/or reduce cost of a product
• Understand how a product will react under "real world" manufacturing and usage conditions
• Use as an aid in screening new product ideas when physical constraints are uncertain

Leading Edge's Robust Design and Variation Analysis experts can help your company in all of the above areas.  Whether you need robust parameter design, statistical tolerance analysis, or general Six Sigma projects Leading Edge can apply an array of powerful statistical tools to your engineering and product design problems.  If you are not sure where to begin, our consultants will assist in developing a strategy that will fit your companies size and vision.

Robust Design is the practice of making the response of a system insensitive or robust to uncontrollable variation: for both the system's target value and its variance. Traditional robust design methods relied solely on costly and time consuming experiments.  Leading Edge can simulate many of these experiments through tools like FEA, CFD, and Kinematic Simulation; thus reducing the cost and effort.  Robust Design techniques generally applied early in the design process when rough concepts are still being evaluated. Traditionally, when a new system is designed, nominal values are chosen as inputs in engineering calculations or experiments.  Consider the design of a new hand tool that must withstand being dropped throughout its life.  How high is the tool being dropped from? How hard is the floor that it is landing on? Which direction is it landing in? How resilient is the plastic that makes up the tool?  Generally, one single (and sometimes arbitrary) value is chosen.  We have the ability to simulate this drop thousands of times for our customers, with ranges of values for each of the questions. Tools and processes that we use for this include DOE (design of experiments), RSM (response surface methodology), Monte Carlo analysis, and other statistical tools.  We also draw from our Simulation and Analysis practice area to simulate experimentation.

Statistical Tolerance Stack-up Analysis is used to drive the design (both geometry and tolerances) to best fit the quality and cost requirements as well as the manufacturer’s capabilities.  This analysis can be performed various places throughout a product's lifecycle: from the time that one or multiple rough concepts have been developed to years after a product has been on the market.

Component and assembly tolerances are an often overlooked aspect of product design. If too little attention is paid to design tolerance, products may not perform well, assemble properly, or may experience a shortened lifecycle. If the design calls for tolerances that are too tight, products will meet requirements, but may be too costly to manufacture (if they can be manufactured at all).

The primary tool that we use for tolerance stack-up analysis is computer aided tolerance analysis software. The process begins with defining the items to be measured and their acceptable ranges. Once these details have been established, we build a 3D tolerance stack up model directly over existing design geometry.  When necessary, we can measure process capabilities of our customers’ suppliers.

Other Related Services include:

• Research and development of Cp and Cpk values (internal and suppliers)
• Research and development of Taguchi Quality Loss Function Values
• General Six Sigma Projects
• DFSS (Design for Six Sigma) or Robust Design strategy development