考虑负效应的城市废弃物中转站多目标选址模型及算法

Multi-Objective Model and Algorithm for Municipal Trash Transferring Depots Location Problems by Considering Undesirable Effects

现有研究中废弃物处理设施的负效应测度没有考虑风向、风力等因素的影响,不能很好地反映实际情况。本文提出了一种基于扩散距离的负效应测度方法;以带中转设施的城市废弃物三层物流网络为研究对象,建立以年均总成本和所有居民日人均负效应最小化为目标的中转站选址问题的混合整数规划模型;根据问题特征,设计一个模拟退火算法与贪心算法相结合的混合算法(HSA)同时求解设施选址、中转站容量及废弃物产生点的指派三个决策。算例结果显示,基于扩散距离的负效应测度方法能较好地反映实际情况;HSA算法在求解同类问题上是可行的和有效的。

Municipal waste disposal facilities are typical semi-desirable facilities. Their location decision often needs to apply multi- objective decision methods and consider systemic cost and the undesirable effect. Unfortunately, the existing methods of measuring undesirable effect of trash disposal facilities ignore some obvious factors, such as wind directions, speed, etc. Therefore, they cannot exactly describe practical situations. Moreover, most current study focuses on two-layer municipal trash logistics network, which only covers trash generation points and final disposal points. However, the municipal trash disposal networks of most large and medium-sized cities are three-layer network, in which the transferring depots are one kind of essential facilities. This article first analyzes the characteristics of municipal trash transferring depots in order to understand the undesirable effect, such as local wind directions, wind speed and facility capacity, etc. The concept of ＂diffusion distances＂ is used to amend the traditional Euclidean distance and to state the real effect by considering all the above-mentioned factors. Subsequently, a piecewise function is proposed to measure the undesirable effect based on ＂diffusion~ distances＂. We further study the discrete three-level municipal trash logistics network with transferring depots. A multi-objective mixed integer programming model is introduced to optimize the trash transferring depots location problem in the network and minimize both of the annual total cost and the average daily undesirable effect. In order to＂ solve the model, a two-stage hybrid simulated annealing algorithm （HSA） is devised according to problem features. Facilities location and the capacity of each facility are obtained in the first stage. The assignment of generation points among facilities selected in the previous stage is obtained in the second stage through a well- designed greedy algorithm. The multi-objective simulation of annealing algorithm is fundamental and takes the whole situation into account. The misalignment mechanism based on lp-norm number deals with the multi-objective fitness. All the solutions obtained in different runs are sorted by the strategies proposed by NSGA-II in order to provide decision-makers multiple non-dominate solutions. The analysis results of an example show that the undesirable effect of measurement method based on ＂diffusion distances＂ can state the true situation relatively and is feasible and effective for this kind of municipal waste disposal facilities. HAS is competitive in both algorithm convergence and computing speed with regard to solving the analogous problems.