Market demand estimation for new product development by using fuzzy modeling and discrete choice analysis

R. Aydin, Chun Kit Kwong, Ping Ji, H. M.C. Law

Research output: Journal article publicationJournal articleAcademic researchpeer-review

9 Citations (Scopus)


Market demand estimation is an important process to assess the financial feasibility of new product development (NPD) projects. The development of models for market demand estimation involves market potential estimation and choice modeling. Previous studies commonly used conjoint analysis to develop utility functions which were then used in discrete choice models to generate market share models. The jury of executive opinion method is commonly used in industries wherein a number of experts and/or consultants are always involved in the market potential estimation. However, a high degree of fuzziness always exists in the data obtained from conjoint surveys and the market potential estimation because of the subjective judgments of respondents and experts. However, ignorance of the fuzziness would lead to the over-estimation of market demands. This research aims to tackle the fuzziness associated with market potential estimation and survey data in the development of market demand models. In this paper, a new methodology of developing fuzzy market demand models for NPD is proposed to address the fuzziness by which market demands can be estimated for the worst, normal, and best scenarios. The proposed methodology involves fuzzy choice modeling based on fuzzy regression and discrete choice analysis, and fuzzy estimate generation of market potential. To evaluate the effectiveness of the proposed methodology, a case study of market demand estimation of a new tablet PC is conducted based on the proposed methodology. The results of the implementation are compared with those based on a popular multinominal logit (MNL) based demand model. From the comparison, it can be noted that the estimated market demand based on the MNL model is very close to that for the normal scenario based on the proposed fuzzy demand model. However, the MNL model cannot provide estimates for other scenarios.
Original languageEnglish
Pages (from-to)136-146
Number of pages11
Publication statusPublished - 22 Oct 2014


  • Conjoint analysis
  • Discrete choice analysis
  • Fuzzy market demand model
  • Fuzzy regression
  • New product development
  • Uncertainty

ASJC Scopus subject areas

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this