A possibilistic approach to the modeling and resolution of uncertain closed-loop logistics

作者：Wang Hsiao Fan; Hsu Hsin Wei^*

来源：Fuzzy Optimization and Decision Making, 2012, 11(2): 177-208.

DOI：10.1007/s10700-012-9120-2

• Closed-loop logistics
• Fuzzy number
• Possibilistic mean
• Genetic algorithms
• Shortage and surplus

Closed-loop logistics planning is an important tactic for the achievement of sustainable development. However, the correlation among the demand, recovery, and landfilling makes the estimation of their rates uncertain and difficult. Although the fuzzy numbers can present such kinds of overlapping phenomena, the conventional method of defuzzification using level-cut methods could result in the loss of information. To retain complete information, the possibilistic approach is adopted to obtain the possibilistic mean and mean square imprecision index () of the shortage and surplus for uncertain factors. By applying the possibilistic approach, a multi-objective, closed-loop logistics model considering shortage and surplus is formulated. The two objectives are to reduce both the total cost and the root . Then, a non-dominated solution can be obtained to support decisions with lower perturbation and cost. Also, the information on prediction interval can be obtained from the possibilistic mean and root to support the decisions in the uncertain environment. This problem is non-deterministic polynomial-time hard, so a new algorithm based on the spanning tree-based genetic algorithm has been developed. Numerical experiments have shown that the proposed algorithm can yield comparatively efficient and accurate results.

• 出版日期2012-6
• 单位清华大学