Application of Taguchi design to retail service

Abstract

Purpose The purpose of this paper is to introduce the concept of robust design in retail service literature. Design/methodology/approach Using a Taguchi design comprising of inner L 8 (2 7 ) and outer 2 2 arrays, a discrete choice task was designed for 233 students respond to. Signal‐to‐noise ( S / N ) ratio was used to test design robustness. Findings Negative effects of uncontrollable design factors on service choice were minimized through the use of inner and outer Taguchi arrays. The single composite measure of consumer choice called S / N ratio accounted both for mean and variance of choice probabilities. Research limitations/implications Use of student sample was a major limitation. Also, the interaction between design factors was not tested as it required the use of more complicated designs. Practical implications This method can be used to improve design robustness by minimizing the impact of uncontrollable noise factors. Use of S / N ratio can help to select the design that simultaneously maximized the choice probabilities and minimized performance variation. Originality/value The paper makes important methodological and operational contributions to the retail service design literature. First, the concept of controllable and uncontrollable factors in choice‐based designs is introduced. Second, the use of S / N ratio as a single, composite measure of design robustness was incorporated. Operationally, this study highlights the impact of less‐studied concepts of wayfinding and customer incompatibility on satisfaction of retail customers.