This content is not included in your SAE MOBILUS subscription, or you are not logged in.

Analyzing Customer Preference to Product Optional Features in Supporting Product Configuration

Journal Article
2017-01-0243
ISSN: 1946-3979, e-ISSN: 1946-3987
Published March 28, 2017 by SAE International in United States
Analyzing Customer Preference to Product Optional Features in Supporting Product Configuration
Sector:
Citation: Sha, Z., Saeger, V., Wang, M., Fu, Y. et al., "Analyzing Customer Preference to Product Optional Features in Supporting Product Configuration," SAE Int. J. Mater. Manf. 10(3):320-332, 2017, https://doi.org/10.4271/2017-01-0243.
Language: English

Abstract:

For achieving viable mass customization of products, product configuration is often performed that requires deep understanding on the impact of product features and feature combinations on customers’ purchasing behaviors. Existing literature has been traditionally focused on analyzing the impact of common customer demographics and engineering attributes with discrete choice modeling approaches. This paper aims to expand discrete choice modeling through the incorporation of optional product features, such as customers’ positive or negative comments and their satisfaction ratings of their purchased products, beyond those commonly used attributes. The paper utilizes vehicle as an example to highlight the range of optional features currently underutilized in existing models. First, data analysis techniques are used to identify areas of particular consumer interest in regards to vehicle selection. Consumer responses related to the reasons for/against the purchase of a specific vehicle model form the basis of extracting these areas of interest. Second, satisfaction rating variables are leveraged to characterize and quantify those identified areas of interest. Several methods of handling missing values of rating variables are developed to impute the data sets in supporting the choice modeling. Finally, multinomial logit models are constructed using engineering attributes, customer demographics as well as the rating variables to quantitatively assess customer preferences, evaluate the feasibility of different imputation methods, and predict customers’ choices. The results show that the inclusion of satisfaction ratings improves model fit and predictive capabilities.