Research On Negotiation Strategy Of Collaborative Procurement With Cluster Supply Chain Based On Multi-Agent

Due to factors such as cooperative culture,operating pattern and product preferences,enterprises in the cluster supply chain(CSC)are likely to have conflicts of interest with others during the multi-resource collaborative procurement.It is a question how to solve the conflicts quickly,to reduce the influence of CSC effectively and to reach a reasonable agreement as far as possible.It is gradually becoming a practical challenge for collaborative procurement with CSC.First of all,it takes the purchaser and supplier in CSC as research objects.After analyzing the characteristics of multilateral transact and the negotiation process,this paper explains a construction of a collaborative procurement negotiation model based on multi-agent system(MAS)with a target constraint function.Under the premise of bilateral multi-issue negotiation,adaptive negotiation strategy becomes an important factor of multi-agent conflict resolution.In order to better promote the agent's self-adaptive negotiation ability,this paper proposes an adaptive negotiation strategy based on dynamic selective ensemble learning.The adjustment margin of adaptive learning is appropriately adjusted to achieve conflict resolution between buyers and suppliers via repeated negotiations.Procedures are as follows.First,K-means search is adopted to generate sample sets.Second,the SVM is used to learn the concession amplitude in each evaluation sample,then eliminate the poor performance of sub-learning model with RMSE,calculate the combined weight and the final dynamic selective SVM model is established.Third,the utility function is applied to decide whether to terminate the negotiation.At last it selectes the most appropriate partner on the basis of target constraint function.Finally,the multi-agent adaptive negotiation strategy based on selective ensemble learning is simulated with MATLAB.And it confirms the feasibility of the negotiation strategy.Compared with the traditional negotiation strategy of SVM,the effectiveness of adaptive negotiation strategy is verified from the aspects of prediction error rate,prediction accuracy and negotiation utility.Negotiation is a crucial way to resolve the procurement conflicts of CSC.Aimed at ameliorating the situation of underdeveloped self-adaptive negotiation,this paper constructs a negotiation model for collaborative procurement with CSC based on MAS and combines with a negotiation strategy built on selective ensemble learning.It provides a novel perspective for the analysis of intelligent supply chain management.And it achieves win-win cooperation between purchasing enterprise and supplier in CSC.