SECEIVED F€B 2 8 19% An algorithm for solving large nonlinear optimization problems with simple bounds is described. It is baaed on the gradient projection method and uses a limikd-memory BFGS matrix to appraximak the Hessian of the objective function. We show how to take advantage of the form of the limited-memory approximation t a implement the algorithm efficiently. The remits of numerical tests on a set of large problems are reported.

No abstract

No abstract

End-Of-Life (EOL) product recovery is proved to be an attractive way to achieve sustainable manufacturing while extending the producer's responsibility to closed-loop product service. However, it is still a challenge to provide flexible and smart recovery plans for industrial equipment at different periods of product service. In this paper, we investigate the smart recovery decision-making problem. We propose a system framework for the implementation of smart EOL management based on product condition monitoring. Different product-level EOL business strategies and component-level recovery options are suggested in this recovery decision support system. Then, multi-objective optimization models are formulated to identify the age-dependent recovery roadmap that best matches the product condition and meets the business goals. In order to achieve environmentally friendly recovery, both recovery profits and energy performances are optimized in our models. We conduct a case study of belt lifter used in the automobile assembly line. The Non-dominated Sorting Genetic Algorithm II is used to solve the proposed model. Numerical experiments validate our models and provide practical insights into flexible recovery business.