| In the trend of high demand in personalization,organization of customized production is transformed into workshop consisted of flexible manufacturing cells with robots from classical cluster-oriented process layouts.Material handling system with Automatic Guided Vehicle(AGV)has highly flexibility,scalability and maneuverability.Flow shop with conventional guide-path AGV system has become a typical structure for discrete customized smart workshop.Therefore,the way to combine flow shop production organization with AVG system in order to balance the logistical intensity in space and timeline and improve efficiency has become a difficult problem in the field of smart workshop layout optimization.Aiming at improving efficiency in effective transportation,this paper takes this type of workshop layout as research background to find a conventional guide-path network designed method;based on the proposed method,the integrated design of guided path and shop layout is investigated by considering the AGV guide-path network as an important part of shop floor area;After that,the equipment layout method for flexible manufacturing cell is studied to optimize the performance measures of flexible manufacturing cell.This study broadens the application of the queueing network model theory and digraph connectivity augmentation,and provide feasible theory for the layout design of flow shop with stochastic demands as well as decision support for the intelligent upgrade for discrete customized enterprisesIn this section,a k-strong connectivity guide-patch network design method is proposed for the conventional AVG guide-path network to ensure the transportation efficiency.Based on constraints of the k-strong connectivity and the routing length of guide-path network,a mixed integer programming model is established for the guide-path network.Frank-strong connectivity augment algorithm is improved to determine the minimized routing length.On the basis of this,initial solution and neighborhood operation are designed based on maintaining the orientation and connectivity of digraph,and improved variable neighborhood search(VNS)algorithm is proposed to find the optimal k-strong connectivity guide-path design.Existing benchmark cases are used to testify the effectiveness of the proposed refined VNS,2-strong connectivity benchmark is used to further explore the features of the guide-path network with 2-strong connectivity.2-strong connectivity is compared with strong connective guide-path network benchmark in different aspects,such as transportation distances,guide-path workload distribution,maximum guide-path workloads as well as topological structure index.This section provides a new way for the conventional AGV guide-path design.In this section,an integrating facility layout and guide-path network design method is proposed for the workshop facility layout problem by the considering the influence of conventional AGV guide-path network on shop layout.An AGV guide-path network design rule is proposed for the circumstances when shop floor layout is not given,and the mixed integer programming model of integrated design is established.Transitive constraint graph is designed to describe the relative position between cells,and layout structure graph is constructed by rectangular duality theory of graphs;guide-path network is designed on the basis of this to meet the connectivity demand,and a feasible integrated layout design is achieved without overlapped area between cells and guide-path.Then a heuristic method is proposed based on the transitive constraint graph to find an optimal solution for the integrated layout design.Last but not least,existing benchmark case is used to testify the effectiveness of transitive constraint graph,seven cases of optimal integrated layout design are given to analyze the influence of guide-path network on the corresponding shop floor layout area.This provides a theoretical support for the estimation of layout area in layout design stage.In this section,a flexible layout design method based on finite-buffer open queueing network is proposed with the target to minimize the system performance index in flexible manufacturing cell.Uncertainties,finite buffer storage and transportation capacity of robot is investigated on the system performance index.As the system performance index cannot be described in a closed form of facility location,finite open queueing network model is established for flexible manufacturing system.Considering that queueing network does not have a product form solution when buffer storage is finite,a generalized extension method is used to calculate the system performances approximately.Adaptive large neighborhood search algorithm is used to prevent the search process from falling into local optimum,and optimal work cell layout design is obtained.Three scales of cases are designed to find out the influence of transportation path network and process plan network on the output rate and production cycle. |
PDF Full Text Request