Forecasting methods using fuzzy concepts

Tin-Chih Chen*, Mao Jiun J. Wang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

55 Scopus citations

Abstract

In this paper, fuzzy concepts are applied in forecasting product price and sales in the semiconductor industry which is often conceived as a highly dynamic environment. First, two fuzzy forecasting methods including fuzzy interpolation (FI) and fuzzy linear regression (FLR) are developed and discussed. Forecasts generated by these methods are fuzzy-valued. Next, the subjective beliefs about whether the industry is booming or slumping, and the speed at which this change in prosperity takes place during a given period are also considered. Two subjective functions are defined and used to adjust fuzzy forecasts. Practically, fuzzy forecasts are incorporated with fuzzy programming like fuzzy linear programming (FLP) or fuzzy nonlinear programming (FNP) for mid-term or long-term planning. Advantages over traditional methods are shown in our examples.

Original languageEnglish
Pages (from-to)339-352
Number of pages14
JournalFuzzy Sets and Systems
Volume105
Issue number3
DOIs
StatePublished - 1 Aug 1999

Keywords

  • Fuzzy forecasting
  • Fuzzy interpolation
  • Fuzzy linear programming
  • Fuzzy linear regression
  • Linguistic values
  • Semiconductor
  • Subjective functions

Fingerprint Dive into the research topics of 'Forecasting methods using fuzzy concepts'. Together they form a unique fingerprint.

Cite this