Uncertain Construction And Optimization Of Reverse Logistics System

Compared with traditional forward logistics, reverse logistics aiming to reuse and recycle effectively is a new logistics control mode with huge economic, social and environmental benefits. Uncertainty is the most obvious characteristics which increases the difficulty of design and optimization for system. Sponsored by the Chinese National Programs for High Technology Research and Development, based on in-depth analysis of uncertainty and forecast of waste products recovery level, this dissertation focused on take-back mode, network construction, performance optimization and information platform of reverse logistics. It provided a solid theoretical basis and guidance method for construction of reverse logistics system.In chapter one, background and significance of research paper were stated. Concept of reverse logistics was described. Research state at home and abroad of related fields were summarized. Then main research structure and method were presented.In chapter two, uncertainty of the system was analyzed and uncertain derived mechanical model of reverse logistics was described. Construction and optimization methodology architecture was researched based on fuzzy theory. Then main factors affecting the number of recycling waste products and recycling time were analyzed based on product life cycle. A fuzzy neural network prediction model of recovery level was presented.In chapter three, characteristic of common reverse logistics model was compared. Focusing on three types of take-back modes in market with leader of manufacturer, distributor or third-party, corresponding bilevel programming models for each mode under fuzzy environment were set up and then solved in Stackelberg game framework. Taking investment efficiency and risk preference into account, the optimal retail prices, product return rates, wholesale price and profits of channel members of each mode were compared quantatively. Rationality of manufacturer decision was judged from perspective of whole supply chain. In order to ensure relationship in outsourcing mode stable, assignment mechanism of recovery benefit under the contract was designed.Chapter four described reverse logistics network structure and facilities function. According to liasion with forward logistics network, three kinds of typical network organization structure were analyzed. Taking minimum total operation cost as optimization objective, a common model of network structure was developed. Regarding quantity and demand of returned products and facilities capacity as fuzzy parameters, the proposed model was solved by converting fuzzy chance constraints to their respective crisp equivalents. A new adaptive particle swarm optimization algorithm with dynamically changing inertia weight was brought forward to solve the problem. Then environmental adaptability of network was discussed, and performance robustness index was designed to evaluate network robustness.Chapter five examined the impact of environmental issues on long-term behaviour of reverse logistics system. The environmental issues examined were environmental protection strategy, remanufacturing capacity expansion strategy and market behavior. Behaviour of system was analyzed through a dynamic simulation model based on the principles of the system dynamics methodology. The dynamic model provided a useful tool, which can be used to evaluate the effect of environmental issues on long-term decision making in collection and remanufacturing activities.In chapter six, based on forword theoretical studies, demand analysis of reverse logistics information platform was carried on. Structural and functional model of platform was built up.In chapter seven, main conclusions of this dissertation were summarized and further research issues were put forward.