Consideration behavior and design decision making

Abstract

Over the past decade, design engineering has developed a systematic framework to coordinate with consumer behavior models. Traditional consumer models applied in the past has mainly focused on the preference of compensatory trade-offs in the choice decisions. Recent marketing research has become interested in developing consumer models that are ”representative” in that they reflect realistic human decision processes. One important example is consideration: the process of quickly screening out many available alternatives using non-compensatory rules before trading off the value of different feature combinations. This research investigates the impact of modeling consideration behavior to design engineering, aiming at constructing consideration models that can inform strategic decisions. The study includes several features absent in existing research: quantifying the mis-specifications of the underlying choice process, tailoring survey instruments for particular models, and exploring the models? strategic value on product profitability and design decisions. First, numerical methods are explored to address the discontinuity in the profit-oriented optimization problem introduced by the consideration models. Methods based on complementarity constraints, smoothing functions, and genetic algorithms are implemented and evaluated with a vehicle design case study. Second, a simulation experiment based on synthetic market data compares consideration models and a variety of conventional compensatory choice models in model estimation and design optimization. The simulation finds that even when estimated compensatory models provide relatively good predictive accuracy, they can lead to sub-optimal design decisions when the population uses considerations; convergence of compensatory models to non-compensatory behavior is likely to require unrealistic amounts of data; and modeling heterogeneity in non-compensatory screening is more valuable than modeling heterogeneity in compensatory trade-offs. The synthetic experiment framework then further extends the comparison to include the survey design process guided by the different assumptions behind consideration models and traditional choice models. A product line design case study reveals