Journal of Manufacturing Systems 31 (2012) 240–252 Contents lists available at SciVerse ScienceDirect Journal of Manufacturing Systems journa l h o me pag e: www.elsevier.com/locate/jmansys Technical Paper Intelligent evaluation of supplier bids using a hybrid technique in distributed supply chains Javad Soroor a,∗ , Mohammad J. Tarokh a , Farid Khoshalhan a , Sara Sajjadi b a Industrial Engineering Department, K.N. Toosi University of Technology, Tehran, Iran b Payam Noor University, Shiraz, Iran a r t i c l e i n f o Article history: Received 3 June 2011 Received in revised form 2 September 2011 Accepted 4 September 2011 Available online 21 October 2011 Keywords: Decentralized supply chain Distributed coordination Supplier evaluation and selection Agent-based systems Fuzzy hybrid method Ubiquitous control Product supply chain a b s t r a c t The main idea of this research is to devise the smart module to pick the best supplier bid(s) automat- ically. The hybrid model is composed of three useful tools: fuzzy logic, AHP, and QFD. The approach has been carefully implemented and verified via a real-world case study in a medium-to-large industry manufacturing vehicle tires and other rubber products. A collection of 12 assessment criteria classified into two categories have been considered. Eight factors are derived from customer suggestions and the other four are design specifications required to manufacture the product. The main outcomes are: a hybrid autonomous model to evaluate supplier bids without direct human intervention; devising a hybrid three- module method and overcoming complexity of computations in resulting algorithm by means of agents; outlining the best criteria to assess suppliers; evaluating the suppliers based on voice of customer during all stages of the process; and discussing analysis, design, and implementation issues of the evaluation agent. The paper includes implications for development of an integrated total system for supply chain coordination. The most important advantages of this work over earlier researches on supplier selection are: implementation of an autonomous assessment mechanism using intelligent agents for the first time, making the best out of three widely applied methodologies all at once, evaluation process mainly based on features of customer order, coordination of supply job based on a bidding system, and portal-mediated operation and control. © 2011 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved. 1. Introduction An “autonomous supply chain coordination” system works based on self-controlled interaction and negotiation of software agents, also known as softbots [1]. The detailed description of such multi-agent system (MAS) has been presented in previ- ous literature [2–10]. Multi-agents systems can be viewed as an improvement of artificial intelligence, in order to realize autonomous computational systems [11]. Autonomy is the main characteristic describing an agent, being the ability to carry out a task and reach its objectives without human, or any other, assistance [12]. Principally, each agent has got properties such as autonomy, rationality, sociability, mobility, and reactivity. Besides, an agent may have built-in reasoning method in order to propose intelligent solutions and to evolve by experiences. Thus, intelligent agents have attributes and methods, similar to the object-oriented programming, and moreover, they have inherent attitudes, desires, and intentions, linked with their environment and provide them of the states that determine their behavior [13]. ∗ Corresponding author at: POB 71555-313, Shiraz, Iran. Tel.: +98 917 7207966. E-mail address: Soroor@sutech.ac.ir (J. Soroor). One of the main tasks in a real-time system for agent-based coordination of the supply chain is to carefully evaluate and select the best supplier(s) in a timely manner. This role is assigned to the “Supplier Evaluation Agent”. This agent works based on its inherent knowledge and performs the designated jobs in coop- eration with other coordination agents in the supply network [1,3]. It makes use of the information provided by the supplier agents and in case alternatives are sought, may rank the suppli- ers. This study has narrowed its focus to single-product decentral- ized supply chains. It’s assumed that there are three tiers in the supply network: manufacturer, supplier, and vendor. The customer may access the vendor’s web site/electronic gateway to select prod- uct features and place an order. The vendor agents then interact with the multi-agent supply chain coordination system and make inquiry about possible suppliers of the specified product. Here, a bidding system performs the job. When the bids are ready, it’s time to evaluate and select the best supplier based on predefined and bid-oriented criteria. The main subject-matter of this research is to develop the best supplier assessment methodology to be applied as the decision- making procedure by the supplier evaluation agent. This hybrid model applies fuzzy logic to AHP (analytical hierarchical process) and QFD (quality function deployment) to provide an intelligent 0278-6125/$ – see front matter © 2011 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jmsy.2011.09.002