Research On Service Performance Towards Intelligent Logistics Center

Intelligent logistics center is designed to provide users with high quality services.In this environment,we redesign the deployment of logistics center using edge architecture.Based on the designed logistics center,we construct a good recommendation system and a device tracing system to guarantee the cost-minimization of deployment while ensure the optimal service performance.The main contributions of this article are summarized as follows:(1)For the cost-minimization deployment issue of intelligent logistics center,an approach is introduced to redeploy an intelligent logistics center based on edge-cloud.Based on the deployment of the intelligent logistics center,a cost function model is constructed.Then a heuristic algorithm,the Monkey Algorithm(MA),simulates the climbing process of a group of monkeys to solve multiple costs in the process of deploying the logistics center to solve the problem.Specifically,the cost model cannot be solved within a certain polynomial,secondly,all decision variables related to the problem are binary.In addition,the algorithm is easy to fall into local optimum due to the diversity of monkey populations during the late stage of search,.In this paper,binary-decoding(Decoding)and genetic algorithm(GA)are introduced to construct a genetic-binary decoding monkey group(DecodingGenetic MA,DGMA)algorithm based on MA.This algorithm solves the problem of minimum deployment cost of intelligent logistics center under various constraints,and avoids the problem of the algorithm falling into local optimization.(2)Aiming at the problem that the logistics center's products need to be accurately recommended,this paper proposes a product recommendation system based on the Bidirectional Long Short-Term Memory(BLSTM),which extracts demand behavior characteristics from a large number of user demand commodity sequences,and then predicts the product that the user will need.Since the products are constantly changing in time series,the interactive model(Mogrifier,M)is introduced in LSTM,which improves the accuracy of prediction and improves the accuracy of recommendation.Specifically,on the first layer,a 1-D convolution neural network(CNN)is designed to reduce the computational costs.Then,MBLSTM affords the modelling of a richer space of interactions between inputs and LSTM to improve the accuracy of recommendation.Afterwards,a fully connected neural network(FCNN)is harnessed to learn and sample predicts from the MBLSTM.As a result,the recommendation system based on MBLSTM algorithm can improve the recommendation accuracy of time-varying products.(3)Aiming at solving the problem of popular commodities need to be traced in logistics centers,the paper proposes a commodity traceability system based on blockchain.Specifically,the blocks with more resources are used to preferentially configure the popular commodities,so as to first ensure the accuracy of the traceability of the popular commodities.Secondly,the fusion positioning algorithm is used to track the clustered popular commodities,then location information is generated into blocks which need to be authenticated with other blocks to ensure the security of the traceability process,finally the socket protocol improves the transmission efficiency of the traceability system.