A type-2 fuzzy intelligent agent based on sparse kernel machines for reducing bullwhip effect in supply chain

In this paper, a new type-2 fuzzy intelligent agent system (T2F-IAS) for reducing bullwhip effect in a supply chain is proposed. This system uses a special kind of sparse kernel machines, called support vector regression, for forecasting future demands of each echelon in a supply chain. The T2F-IAS...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2015 Annual Conference of the North American Fuzzy Information Processing Society (NAFIPS) held jointly with 2015 5th World Conference on Soft Computing (WConSC) s. 1 - 7
Hlavní autoři: Gamasaee, R., Fazel Zarandi, M. H., Turksen, I. B.
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.08.2015
Témata:
Bullwhip effect
Demand forecasting
Expert systems
Forecasting
Intelligent agent system
Intelligent agents
Kernel
Sparse kernel machines
Steel
Sulfur
Supply chains
Type-2 fuzzy
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In this paper, a new type-2 fuzzy intelligent agent system (T2F-IAS) for reducing bullwhip effect in a supply chain is proposed. This system uses a special kind of sparse kernel machines, called support vector regression, for forecasting future demands of each echelon in a supply chain. The T2F-IAS includes a data collector agent and a rule generator agent. A type-2 fuzzy c-regression clustering model is employed in the rule generator agent for generating the most proper rules. This agent uses an interval type-2 fuzzy (IT2F) hybrid expert system for demand forecasting. Moreover, adaptive network based fuzzy inference system (ANFIS) is applied to learn parameters used in the agents. Thereafter, the results of the proposed T2F-IAS are compared with type-1 fuzzy intelligent agent system (T1F-IAS) and a method in literature for validating the proposed method. The results indicate that bullwhip effect and forecasting error are remarkably reduced by using the proposed T2F-IAS.
DOI:10.1109/NAFIPS-WConSC.2015.7284207