توزیع مدیریت زنجیره تامین با استفاده از بهینه سازی کلونی مورچه ها

Successful supply chain management requires a cooperative integration between all the partners in the network. At the operational level, the partners individual behavior should be optimal and therefore their activities have to be planned using sophisticated optimization tools. However, these tools should take into account the planning of the remaining partners, through the exchange of information, in order to allow some kind of cooperation between the elements of the chain. This paper introduces a new supply chain management technique, based on modeling a generic supply chain with suppliers, logistics and distributers, as a distributed optimization problem. The different operational activities are solved by the optimization meta-heuristic called ant colony optimization, which allows the exchange of information between different optimization problems by means of a pheromone matrix. The simulation results show that the new methodology is more efficient than a simple decentralized methodology for different instances of a supply chain.

Supply chain (SC) systems are nowadays entering the age of adaptive and intelligent supply chains, a new generation of networks that features collaboration and visibility features across the different partners to deal with the system dynamics, such as supplier failures or demand uncertainty [6] and [21]. Supply chains systems are a set of separate and independent economic entities more interested in their local objectives than in the global system performance. Therefore, centralized management approaches, where a single partner such as the logistic center optimizes the global performance, are becoming less realistic and being replaced by decentralized management approaches, where each member optimizes its own performance, albeit knowing that collaboration with other partners can improve the individual and global performance. In any case, the key issue is to align the members objectives and coordinate their decisions to optimize the supply chain performance, but this is particularly more difficult to attain with a decentralized management approach [13]. At the operational level, supply chain management (SCM) is now seeking to determine the stock levels at the logistic centers depending on the demand, or the size and frequency of batches produced at the suppliers to feed the producers in time, or even the delivery planning that minimizes the transportation costs and environmental impacts [8]. The coherence between the different decision making centers in the chain can be easily accomplished by a multi-agent framework [17]. These systems are based on explicit communication between specialized agents assigned to structural elements of the chain (e.g. supplying or logistic agents) about their tasks and using an interaction protocol with a specific message structure, conversation rules, action and reaction behaviors [1]. The research in this field has tackled mainly the interaction between the agents and the optimization issues are usually solved through some simple dispatching rules. However, these methods are usually not sufficient to deal with the complexity of the real-world problems and the agents need to use more powerful optimization techniques [14]. Moreover, to take full advantage of the supply chain framework, the communication protocols should support the possibility of exchanging information during the optimization process, in order to allow agents to react to failures or other type of dynamic disturbances. However, it is necessary to take into account all sort of problems regarding level of disclosure and the asymmetry between the supply chain members that can generate opportunistic behaviors. This paper introduces a multi-agent supply chain management methodology based on the description of the supply chain as a set of different distributed optimization problems and using the ant colony optimization (ACO) meta-heuristic [10] to achieve cooperation between different multiple partners. While optimizing, the ACO algorithm builds a pheromone matrix, which is an indirect record of the optimization steps. This matrix can be accessed at all times during the optimization process and contains no private information of any kind. If each system in a supply-chain is optimized by its own ant colony-agent, the pheromone matrices can be used to exchange information between the different systems as a multi-agent system, introducing in this way a coordination mechanism into the supply chain. The paper proceeds as follows: Section 2 describes the SC model that is used in this paper. Section 3 presents a short literature survey on decentralized supply chain management and models the management problem as a distributed optimization problem. Section 4 shows how the ant colony optimization can be used to solve this problem as a multi-agent system. The simulation results are presented in Sections 6 and 7 concludes the paper and presents the guidelines for future work.

This paper proposed a new management technique for operational activities of a generic supply chain, with suppliers, logistic and distribution partners. The methodology consists of modeling each of the partners as a combinatorial benchmark optimization problem and optimizing each problem using the ant colony optimization algorithm. This algorithm uses a pheromone matrix to keep an information record during the optimization procedure. With this matrix, it is possible to exchange information between the different optimization processes running in parallel and achieve a cooperation mechanism. The exchange of information with a particular partner may bias the solutions of the remaining partners towards a different but still optimal solution, that suits better the solution of the particular partner. The simulation results showed that this strategy was able to improve the global supply chain performance. The logistic and the distribution systems improved their performance, without compromising the performance of the suppliers. As future research work, we intend to evaluate the impact of the proposed methodology under different coordination mechanisms, such as contracts with penalty clauses or when the suppliers do not allow a decrease in their individual goal. This aspect can dealt with using fuzzy objectives, for instance [18]. We also aim the generalization of this methodology to different types of optimization algorithms, especially to other meta-heuristics.