Estimate the demand for a product family with a discrete choice model

With the growing diversification in consumer life styles and needs, organizations are increasingly competing in coming up with product lines instead of focusing on a single product; in expectation to cover a broader range of market segments. However, putting too many products in a product line may b...

Full description

Saved in:
Bibliographic Details
Main Author: Lian Lim, Philips
Other Authors: School of Mechanical and Aerospace Engineering
Format: Final Year Project
Language:English
Published: 2014
Subjects:
Online Access:http://hdl.handle.net/10356/61533
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:With the growing diversification in consumer life styles and needs, organizations are increasingly competing in coming up with product lines instead of focusing on a single product; in expectation to cover a broader range of market segments. However, putting too many products in a product line may bring more damage than benefit to the company hence it is essential for manufacturers to know their optimal product line. One key aspect to rate the performance of a product line would be on how much market demand it covers. A critical problem faced with manufacturers is that demand forecasting in a high variety market is often inaccurate, sometimes misleading. Hence the objective of this research is to develop a demand estimation model based on discrete choice analysis that would help manufacturers in their business decisions. Using Mixed Logit due to its ability to model complex interactions among different choices, the report presents a mathematical model to predict how much demand one product line could capture. The proposed methodology was illustrated with a case study about smartphones in a university campus. Data collected based on surveys were used to derive the preference distribution of consumers in the market. The preference distribution was then used to estimate the choice probability of different smart phone models, which was further utilized to estimate the demand of each smart phone model. Finally, the product line demand estimated by the model was compared with the demand estimated by statistics from the surveys. The result indicates that the estimation was quite close to the demand estimated from survey statistics, which demonstrates the validity of the methodology.