TY - JOUR
T1 - Demand trend mining for predictive life cycle design
AU - Ma, Jungmok
AU - Kwak, Minjung
AU - Kim, Harrison M.
N1 - Funding Information:
The work presented in this paper is supported by the National Science Foundation under Award No. CMMI-0953021. Any opinions, findings and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the National Science Foundation.
PY - 2014/4/1
Y1 - 2014/4/1
N2 - The promise of product and design analytics has been widespread and more engineering designers are attempting to extract valuable knowledge from large-scale data. This paper proposes a new demand modeling technique, Demand Trend Mining (DTM), for Predictive Life Cycle Design. The first contribution of this work is the development of the DTM algorithm for predictability. In order to capture hidden and upcoming trends of product demand, the algorithm combines three different models: Decision tree for large-scale data, discrete choice analysis for demand modeling, and automatic time series forecasting for trend analysis. The DTM dynamically reveals design attribute pattern that affects demands. The second contribution is the new design framework, Predictive Life Cycle Design (PLCD), which connects the DTM and data-driven product design. This new optimization-based model enables a company to optimize its product design by considering the pre-life (manufacturing) and end-of-life (remanufacturing) stages of a product simultaneously. The DTM model interacts with the optimization-based model to maximize the total profit of a product. For illustration, the developed model is applied to an example of smart-phone design, assuming that used phones are taken back for remanufacturing after one year. The result shows that the PLCD framework with the DTM algorithm identifies a more profitable product design over a product life cycle when compared to traditional design approaches that focuses on the pre-life stage only.
AB - The promise of product and design analytics has been widespread and more engineering designers are attempting to extract valuable knowledge from large-scale data. This paper proposes a new demand modeling technique, Demand Trend Mining (DTM), for Predictive Life Cycle Design. The first contribution of this work is the development of the DTM algorithm for predictability. In order to capture hidden and upcoming trends of product demand, the algorithm combines three different models: Decision tree for large-scale data, discrete choice analysis for demand modeling, and automatic time series forecasting for trend analysis. The DTM dynamically reveals design attribute pattern that affects demands. The second contribution is the new design framework, Predictive Life Cycle Design (PLCD), which connects the DTM and data-driven product design. This new optimization-based model enables a company to optimize its product design by considering the pre-life (manufacturing) and end-of-life (remanufacturing) stages of a product simultaneously. The DTM model interacts with the optimization-based model to maximize the total profit of a product. For illustration, the developed model is applied to an example of smart-phone design, assuming that used phones are taken back for remanufacturing after one year. The result shows that the PLCD framework with the DTM algorithm identifies a more profitable product design over a product life cycle when compared to traditional design approaches that focuses on the pre-life stage only.
KW - Data-driven product design
KW - Decision tree
KW - Demand Trend Mining
KW - Product and design analytics
KW - Product life cycle design
UR - http://www.scopus.com/inward/record.url?scp=84897111094&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84897111094&partnerID=8YFLogxK
U2 - 10.1016/j.jclepro.2014.01.026
DO - 10.1016/j.jclepro.2014.01.026
M3 - Article
AN - SCOPUS:84897111094
SN - 0959-6526
VL - 68
SP - 189
EP - 199
JO - Journal of Cleaner Production
JF - Journal of Cleaner Production
ER -