An advanced fuzzy collaborative intelligence approach for fitting the uncertain unit cost learning process

Abstract

Estimating the unit cost of each product precisely and accurately is a prerequisite to determining the profitability of a manufacturer, which is usually addressed by fitting the underlying learning process. However, existing methods for this purpose often deal with a logarithmic or log-sigmoid value, rather than the original value, of the unit cost. To resolve this problem, in this study, a new fuzzy collaborative intelligence (FCI) approach is proposed by considering the original value of the unit cost directly. The effectiveness of the new FCI approach is validated with a real dynamic random access memory (DRAM) case. The experimental results showed that the new FCI approach outperformed two existing methods in improving the fitting accuracy in terms of MAE and MAPE and also in reducing the average range of the fitted unit costs.