Robust parameter design and economical multi-objective optimization on characterizing rubber for shoe soles

Abstract

Within Taguchi methods, robust parameter design is a widely used tool for quality improvement in processes and products. The loss function is another quality improvement technique with a focus on cost reduction. Traditional Taguchi methods focus on process improvement or optimization with a unique quality characteristic. Analytical approaches for optimizing processes with multiple quality characteristics have been presented in the literature. In this investigation, a case of analysis for two quality characteristics in rubber for shoe sole is presented. A methodology supported in robust parameter design in combined array is used in order to obtain optimal levels in the vulcanization process. Optimization techniques based on the loss function and the use of restricted nonlinear optimization with genetic algorithms are proposed.

A Study of the Impacts of Measurement errors on Robust Parameter Design for Multi-response

“…The data used for the study was from a case study of the vulcanization process of rubber shoe sole experiment provided in Mares-Castro [20] involving three control factors-pressure, time, mold temperature and one noise factor-the environmental temperature. Each of the input factors are set at low, medium and high levels respectively and coded as +1, 0, -1.…”

“…Nejlaoui et al [19] applied RPD and multi-response optimization technique in the analysis of a railway vehicle trajectory. Mares-Castro [20] also studied RPD and economical multi-objective optimization on characterizing rubber for shoe soles. However, the works do not consider the implication of measurement errors in the levels of control variables on RPD.…”