A Study Of Decisions In A Remanufacturing System Based On Hybrid Intelligent Algorithm

With the increasing productivity and living standard, the type and quantity of a variety of products rapidly grow and product renewal circle gradually accelerates. Therefore, a large number of reverse logistics activities, such as product acquisition, recycling of waste, and repairing and remanufacturing, have appeared in recent years.Reverse logistics is a complex process, involving many decision-makers. The fact of decision-makers maximizing their own interests may result in dropping the overall interests of reverse supply chain. With regard to the general reverse recycling network optimization problem containing five-tier decision-makers, firstly an equilibrium model is established by using super-network theory and equilibrium theory. Then the equivalent variational inequality formulation of the equilibrium conditions that each decision-maker and the whole reverse supply chain simultaneously achieve is derived by using variational inequality theory. Finally the Corrected Projection Method is applied for solving proposed model. Some numerical examples are presented to illustrate the feasibility of our model and to analyze the effects of parameters (risk attitude and sensitivity) on equilibrium solutions.Remanufacturing is an effectively and extensively applied recycling method. However, due to differences from sources and arrival time, the condition of collected used products may highly volatile, which makes the remanufacturing system facing more risks. In order to reduce the uncertainty, the optimal pricing decision in a centralized remanufacturing system with random fuzzy quality of used products is established. Then, according to different management goals, the expected profit maximization model, (α,β)-profit maximization model, and chance maximization model are respectively developed. Random fuzzy simulation, neural network and genetic algorithm are integrated to produce a hybrid intelligent algorithm for obtaining the optimal acquisition price and sales price. Numerical examples are given to demonstrate the feasibility of our model and algorithm, and provide some practical suggestions. To improve the cooperative relations between the remanufacturer and the retailer in a decentralized remanufacturing system with random quality level of used products, a profit coordination model is proposed by adopting both whole-price-ordering and option contract, and is compared with the optimal decisions under the news-vendor model. The results of numerical examples are presented to illustrate the validity of option contract on profit coordination. The effects of the uncertainties of quality level and demand on decisions are also studied through numerical examples.