0
Research Papers: Design Automation

Exploring Product Solution Differences Due to Choice Model Selection in the Presence of Noncompensatory Decisions With Conjunctive Screening Rules

[+] Author and Article Information
Jaekwan Shin

Department of Mechanical
and Aerospace Engineering,
North Carolina State University,
911 Oval Drive,
Raleigh, NC 27695
e-mail: jshin5@ncsu.edu

Scott Ferguson

Associate Professor
Department of Mechanical
and Aerospace Engineering,
North Carolina State University,
911 Oval Drive,
Raleigh, NC 27695
e-mail: scott_ferguson@ncsu.edu

1Corresponding author.

Contributed by the Design Automation Committee of ASME for publication in the JOURNAL OF MECHANICAL DESIGN. Manuscript received January 6, 2016; final manuscript received October 8, 2016; published online December 12, 2016. Assoc. Editor: Harrison M. Kim.

J. Mech. Des 139(2), 021402 (Dec 12, 2016) (14 pages) Paper No: MD-16-1005; doi: 10.1115/1.4035051 History: Received January 06, 2016; Revised October 08, 2016

Research in market-based product design has often used compensatory preference models that assume an additive part-worth rule. These additive models have a simple, usable form and their parameters can be estimated using existing software packages. However, marketing research literature has demonstrated that consumers sometimes use noncompensatory-derived heuristics to simplify their choice decisions. This paper explores the quality of optimal solution obtained to a product line design search when using a compensatory model in the presence of noncompensatory choices and a noncompensatory model with conjunctive screening rules. Motivation for this work comes from the challenges posed by Bayesian-based noncompensatory models: the need for screening rule assumptions, probabilistic representations of noncompensatory choices, and discontinuous choice probability functions. This paper demonstrates how respondents making noncompensatory choices with conjunctive rules can lead to compensatory model estimations with distinct respondent segmentation and relative, large absolute part-worth values. Results from a product design problem suggest that using a compensatory model can provide benefits of smaller design errors and reduced computational costs. Product design optimization problems using real choice data confirm that the compensatory model and the noncompensatory model with conjunctive rules provide comparable solutions that have similar likelihoods of not being screened out when using a consideration set verifier. While many different noncompensatory heuristic rules exist, the presented study is limited to conjunctive screening rules.

FIGURES IN THIS ARTICLE
<>
Copyright © 2017 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

Flowchart of the study used to compare compensatory models and a Bayesian-based noncompensatory model with conjunctive screening rules for synthetic choice data

Grahic Jump Location
Fig. 2

Conceptual procedure for consideration set verifier using hypothetical screening rules

Grahic Jump Location
Fig. 3

An example of a simulated noncompensatory choice using discrete choice data obtained from an actual survey

Grahic Jump Location
Fig. 4

Histogram of aggregate posteriors for transmission attribute obtained using the HB-ML model

Grahic Jump Location
Fig. 5

Conceptual diagram to show the absence of a strict threshold in compensatory modeling of noncompensatory choice

Grahic Jump Location
Fig. 6

Interval comparison between the max. and min. choice probabilities of each attribute

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In