In new product design, risk averse firms must consider downside risk in addition to expected profitability, since some designs are associated with greater market uncertainty than others. We propose an approach to robust optimal product design for profit maximization by introducing an α-profit metric to manage expected profitability vs. downside risk due to uncertainty in market share predictions. Our goal is to maximize profit at a firm-specified level of risk tolerance. Specifically, we find the design that maximizes the α-profit: the value that the firm has a (1 − α) chance of exceeding, given the distribution of possible outcomes. The parameter α ∈ (0,1) is set by the firm to reflect sensitivity to downside risk (or upside gain), and parametric study of α reveals the sensitivity of optimal design choices to firm risk preference. We account here only for uncertainty of choice model parameter estimates due to finite data sampling when the choice model is assumed to be correctly specified (no misspecification error). We apply the delta method to estimate the mapping from uncertainty in discrete choice model parameters to uncertainty of profit outcomes and identify the estimated α-profit as a closed-form function of decision variables for the multinomial logit model. An example demonstrates implementation of the method to find the optimal design characteristics of a dial-readout scale using conjoint data.

In new product design, risk averse firms must consider downside risk in addition to expected profitability, since some designs are associated with greater market uncertainty than others. We propose an approach to robust optimal product design for profit maximization by introducing an α-profit metric to manage expected profitability vs. downside risk due to uncertainty in market share predictions. Our goal is to maximize profit at a firm-specified level of risk tolerance. Specifically, we find the design that maximizes the α-profit: the value that the firm has a (1-α) chance of exceeding, given the distribution of possible outcomes. The parameter α [0,1] INTRODUCTIONOver the last three decades, a significant portion of the new product development (NPD) literature has been dedicated to the integration of engineering design and marketing processes for differentiated markets. Simple models to determine the most profitable characteristics of a single new product [1-2] have progressed to account for issues such as product-line design and preference heterogeneity [3][4][5][6][7], competitor reactions [8-10], cost structure [11][12], distribution channels [9,13-16], choice-set-dependent preferences [17], and coordination with constrained engineering design decisions [18][19][20][21][22][23][24][25][26].As Hsu and Wilcox [27] argue, the trend towards estimating marketing models at lower levels of aggregation that are more structural 1 in consumer behavior representation has led to models with many parameters and consequently greater uncertainty of those parameters. However, despite the advances in NPD methods, the research has not given much consideration to the intrinsic parameter uncertainty of the demand models. Demand uncertainty directly affects the risk of introducing a new product into the market, and firms evaluate potential projects not only in terms of expected return, but also in terms of risk.The purpose of this work is threefold. First, we define a robust a-profit metric and propose a general framework to incorporate demand uncertainty arising from choice model parameter estimation into the design decision process such that it accounts for varying levels of loss tolerance. Second, we apply the delta method to approximate the a-profit function in closed form for multinomial logit (MNL) demand models to be used efficiently in numerical optimization routines. Finally, we show how ignoring demand uncertainty can lead to suboptimal decisions for risk averse firms.We do not intend to consider all the various sources of demand model uncertainty [28], and several questions will remain open. In particular, we assume the discrete choice model is correctly specified and ignore uncertainty due to model misspecification, and we assume that the model parameters do not change over time or from the context in which the data were collected to the context in which predictions will be made. Nevertheless, the proposed methodology can be useful, and it serves as a first step in addressing design for profit maximization...

When design decisions are informed by consumer choice models, uncertainty in the choice model and its share predictions creates uncertainty for the designer. We take a first step in investigating the variation in and accuracy of market share predictions by characterizing fit and forecast accuracy of multinomial logit, mixed logit, and nested logit models over a variety of utility function specifications for the US light duty new vehicle market. Using revealed preference data for years 2004–2006, we estimate a multinomial logit model for each combination of a chosen set of utility function covariates found in the literature. We then use each of the models to predict vehicle shares for the 2007 market and examine several metrics to measure fit and predictive accuracy. We find that the best models selected using any of the proposed metrics outperform random guessing yet retain substantial error in fit and prediction for individual vehicle models. For example, with no information (random guessing) 30% of share predictions are within 0.2% absolute share error in a market with an average share of ∼0.4%, whereas for the best models 70% are within 0.2% (for the 2007 vehicle market this translates to an error of ∼33,000 units sold). Share predictions are sensitive to the presence of utility covariates but less sensitive to the form. Models that perform well on one metric tend to perform well on the other metrics as well. In particular, models selected for best fit have comparable forecast error to those with the best forecasts, and residual error in model fit is a major source of forecast error.