Criteria used for demand forecasting are both qualitative and quantitative.
Fuzzy-logic-based systems are well suited when working with qualitative
criteria.
. Hybrid methods such as neuro-fuzzy techniques have more realistic results in the
forecasting area. Individual fuzzy systems are not suggested for demand
forecasting. ANFIS combines the reasoning capability of the fuzzy logic with
learning capability of the NN system.
. According to literature research and conversations with apparel manufacturers’
specialists, there is not any common analytic method for demand forecasting in
apparel industry and to our knowledge, there is not adequate number of study in
literature to forecast the demand with ANFIS for apparel manufacturers.
Thomassey et al. (2002, 2005) and Thomassey (2010) presented fuzzy systems for
fashion sales forecasting; but they focused on fashion distributors or fashion
retailers and their forecasting horizon is about one week (for short-term) or one
season (for mean-term).
. In the apparel industry, purchasing decisions can be easily affected by the
political or financial volatility of the environment. This volatility also increases
the complexity of the demand forecasting system. It is very difficult to capture
this volatility by using statistical methods. For this reason, the ANFIS technique
can be well suited approach for such a dynamic environment.
. The proposed ANFIS method can deal with the complexity of the decision
making process and does not require the formulation of the decision making
process.