A design of experiment approach to tolerance allocation
This thesis presents a design of experiment approach to tolerance allocation. The purpose of this thesis is to develop a new approach to tolerance allocation problems. The new approach is based on the Taguchi method, and the objective is to determine : a) the stacking function, b) the significant component tolerances that affect the stackup function, and c) the levels of the tolerances that result in the highest quality products. A bench vice is used as a case study example to demonstrate the methodology. The methodology applies Taguchi's parameter design concept to determine the set of component feature tolerances that will result in the lowest cost product subject to quality constraints on the product's function. The parameter design concept employs inner and outer orthogonal arrays to identify the significant control factors that are least sensitive to noise. For tolerance allocation, all tolerances in the stackup function are the control variables (inner array), and each is set at its high (loose tolerance) or low (tight tolerance) level. The direction of the tolerances (+ or - ) represent the noise variables (outer array). For the bench vice case study, the gap between the end plate and the movable jaw was selected as the response variable. Twelve component features affect the gap, thus requiting an L16 inner and outer array. A simulation model was developed in Microsoft Excel version 4.0 to compute the gap between the two plates. The data were transformed according to Taguchi's smaller-the-better S/N ratio. The significant component features and tolerances were identified from a graphical analysis of the S/N ratio's and verified by ANOVA. The concept tested was found to be a valuable tool and is a novel technique for tolerance allocation. The main advantage of this methodology is that the functional stackup relationship need not to be known explicitly, as in statistical tolerancing or Monte Carlo simulation analysis. The major disadvantage is that the method requires an inordinate number of experiments.
School Location:USA - Ohio
Source Type:Master's Thesis
Keywords:tolerance allocation stacking function taguchi s parameter design
Date of Publication:01/01/1995