| Summary: | Product models selection as one of the key decision making processes in enterprise resource allocation and product development, is gaining popularity in many industries, such as electronics, aircraft, and automobiles etc. Due to the uncertain nature of the selection outcome and influence on the company, a risk and return trade-off needs to be carefully considered.
In this dissertation, the selection process is conducted from a top-down level, where the final evaluation is based on the product models portfolio’s performance. A Mean-Variance mathematical model is constructed via introducing the Modern Portfolio Theory in the financial field, along with the product model’s features weight allocation analysis via Analytic Hierarchy Process in the engineering field. A linkage is built between these two fields via adjusting the correlation matrix in portfolio construction from traditional time zone to frequency zone by introducing the similarity matrix in Similarity Analysis.
Aiming to find out the optimal portfolio that can maximize the return while minimizing the risk, the portfolio with maximum Sharpe ratio is found and relevant resource weight allocation is given via implementing and comparing two different methods: Monte Carlo simulation and Gradient Descent method.
As an explorative study, future perspectives are also discussed to attract more open and in-depth studies for more robust applications of product models selection from a portfolio optimization perspective.
|
|---|