﻿ 基于遗传-禁忌算法的木材物流网络优化
 森林与环境学报  2017, Vol. 37 Issue (3): 309-314 PDF
http://dx.doi.org/10.13324/j.cnki.jfcf.2017.03.010
0

#### 文章信息

LIN Yuying, QIU Rongzu

Optimization of timber logistics network based on the genetic-tabu algorithm

Journal of Forest and Environment,2017, 37(3): 309-314.
http://dx.doi.org/10.13324/j.cnki.jfcf.2017.03.010

### 文章历史

Optimization of timber logistics network based on the genetic-tabu algorithm
LIN Yuying, QIU Rongzu
College of Transportation and Civil Engineering, Fujian Agriculture and Forestry University, Fuzhou, Fujian 350002, China
Abstract: In order to improve the logistics network of timber which could provide guarantee for the smooth running of timber logistics and reduce the cost of the whole area of timber logistics system, combining the production reality of timber logistics, the optimized mathematical model of logistics network of timber was established using the strategies and processes of genetic-tabu (GA-TS) algorithm. Setting the timber logistics activities of Yonglin group as an example to analyze the empirical analysis of the timber logistics network optimization. The model and algorithm were combined with computer technology. Solver procedure was written by C# and geographical information systems software platform (ArcGIS) was used to implement scheme visualization. Results showed that the model and the algorithm are feasibility and effectiveness, can improve the efficiency and precision of timber logistics network optimization decision, and reduce logistics cost.
Key words: genetic-tabu algorithm     timber logistics     logistics network optimization

1 木材物流网络优化模型建立

1.1 模型假设

1.2 模型建立

 \begin{align} & \text{min}F=\sum\limits_{h=0}^{H}{\sum\limits_{i=0}^{I}{\sum\limits_{d=0}^{K}{{{x}_{idh}}}}}{{C}_{idh}}+\sum\limits_{h=0}^{H}{\sum\limits_{i=0}^{I}{\sum\limits_{j=0}^{J}{{{x}_{ijh}}}}}{{C}_{ijh}}+\sum\limits_{h=0}^{H}{\sum\limits_{d=0}^{K}{\sum\limits_{j=0}^{J}{{{x}_{djh}}}}} \\ & {{C}_{djh}}+\left( \sum\limits_{d=0}^{K}{{{F}_{d}}}{{Z}_{d}}+\sum\limits_{h=0}^{H}{\sum\limits_{j=0}^{J}{\sum\limits_{d=0}^{K}{{{x}_{djh}}}}}{{q}_{dh}}-\sum\limits_{h=0}^{H}{\sum\limits_{d=0}^{K}{{{y}_{dh}}}}{{m}_{dh}} \right) \\ \end{align}

xijhδijhEh (i= 1，2，…，Ij= 1，2，…，Jh=1，2，…，H)

xijhδijhM (i= 1，2，…，Ij= 1，2，…，Jh=1，2，…，H)

 \begin{align} & \sum\limits_{k=1}^{K}{{{x}_{idh}}}+\sum\limits_{j=1}^{J}{{{x}_{ijh}}}\le {{S}_{ih}}\left( i=1,~2,~\cdots ,~I;h=1,~2,~\cdots ,~H \right) \\ & \sum\limits_{d=1}^{K}{{{x}_{djh}}}+\sum\limits_{i=1}^{I}{{{x}_{ijh}}}\le {{D}_{jh}}~(j~=\text{ }1,2,~\cdots ,J;~h=1,2,\cdots ,H) \\ & \sum\limits_{h=1}^{H}{{{x}_{djh}}}\sum\limits_{i=1}^{I}{{{x}_{idh}}}+{{y}_{dh}}\le {{Q}_{d}}{{Z}_{d}}~(d~=1,2,~\cdots ,K) \\ & \sum\limits_{i=1}^{I}{{{x}_{idh}}}\ge \sum\limits_{j=1}^{J}{{{x}_{djh}}}(d=1,2,~\cdots ,K;~h=1,2,~\cdots ,H) \\ \end{align}

xidh, xijh, xdjh≥0 ∀h, i, j, k

 ${{Z}_{d}}=\left\{ \begin{array}{*{35}{l}} 0 & 备选木材物流中心d未被选中 \\ 1 & 备选木材物流中心d被选中 \\ \end{array} \right.$
 ${{\delta }_{ijh}}=\left\{ \begin{array}{*{35}{l}} 0 & 供材点i与需材点j存在材种h的供需关系 \\ 1 & 供材点i与需材点j不存在材种h的供需关系 \\ \end{array} \right.$

2 基于遗传-禁忌算法的模型求解

 图 1 遗传-禁忌混合算法运算流程 Fig. 1 The operation flow of genetic-tabu algorithm
 \begin{align} & f\left( h_{i}^{t} \right)=1-\frac{u\left( h_{i}^{t} \right)}{u_{\text{max}}^{t}+u_{\text{minu}}^{t}} \\ & u\left( h_{i}^{t} \right)=F\left( h_{i}^{t} \right)+C(h_{i}^{t}) \\ \end{align}

3 实例分析 3.1 研究区概况与数据来源

 供材点Supply sites 供应量Supply quantity/m3 H1 H2 S1 24 000 5 920 S2 43 900 10 890 S3 20 045 4 165 S4 25 800 5 567 S5 38 050 8 465 S6 27 198 4 800 S7 28 899 6 066

 需材点Demand sites 需求量Demand quantity/m3 H1 H2 D1 0 45 788 D2 87 795 0 D3 67 830 0 D4 47 875 0

 物流中心Potential logistics center 年均建设成本Annual constructioncost /(yuan·a-1) 候选点规模Scale of potential site /m3 单位流转费用Unit conversion cost/(yuan·m-3) H1 H2 K1 79 000 85 000 20 24 K2 105 500 131 000 23 20 K3 98 300 100 190 15 25 K4 85 299 801 000 25 20
3.2 优化过程与结果分析

 物流中心Logistics center 单位运输费用Per-unit transportation cost/(yuan·m-3·km-1) S1 S2 S3 S4 S5 S6 S7 D1 D2 D3 D4 K1 17 10 14 15 15 18 14 2 4 3 4 K2 15 19 18 16 14 14 18 4 6 7 4 K3 14 15 16 14 18 16 13 5 8 3 6 K4 18 11 17 12 13 13 15 5 3 4 7

 需材点Demand sites 单位运输费用Per-unit transportation cost/(yuan·m-3·km-1) S1 S2 S3 S4 S5 S6 S7 D1 23 21 24 22 29 15 18 D2 27 26 14 26 17 19 15 D3 15 27 23 15 27 21 22 D4 17 25 15 18 25 17 27

 节点Site 需材量Demand quantity/m3 供材量Supply quantity/m3 D2 D3 D4 K1 K4 S1 8 312 5 232 10 357 0 0 23 901 S2 23 571 0 0 10 604 9 723 43 898 S3 1 850 14 630 3 455 0 0 19 935 S4 11 360 0 3 081 10 132 0 24 573 S5 19 681 0 6 808 9 901 0 36 390 S6 8 130 0 0 11 099 7 100 26 329 S7 14 891 6 232 7 354 0 0 28 477 K1 0 41 736 0 0 0 41 736 K4 0 0 16 820 0 0 16 820 合计Total 87 795 68 730 47 875 41 736 16 823 262 059

 图 2 最优木材物流网络地形图 Fig. 2 The map of best timber logistics network
4 小结

 [1] 张兰怡, 林玉英, 邱荣祖. 基于BPNN的木材物流中心选址方法与实证分析[J]. 福建林学院学报, 2014, 34(1): 82–86. [2] FIGUEIRA G, AMORIM P, GUIMARÃES L, et al. A decision support system for the operational production planning and scheduling of an integrated pulp and paper mill[J]. Computers & Chemical Engineering, 2015, 77: 85–104. [3] 郝斯琪, 沈微. 基于SLP方法的木材物流中心布局:以绥芬河地区为例[J]. 森林工程, 2015, 31(1): 159–163. [4] 邱荣祖, 林雅惠, 钟聪儿. 基于ArcGIS的木材物流中心选址[J]. 林业科学, 2010, 46(6): 113–117. DOI:10.11707/j.1001-7488.20100618 [5] 刘娜翠, 邱荣祖. 基于遗传算法的木材运输方案优化技术[J]. 福建农林大学学报(自然科学版), 2010, 30(4): 380–384. [6] 张淑芬, 邢艳秋, 吴红波, 等. 基于GIS和RS技术的木材运输线路优化研究:以吉林省汪清林区为例[J]. 森林工程, 2011, 27(2): 48–51, 60. [7] ACUNA M, MIROWSKI L, GHAFFARIYAN M R, et al. Optimising transport efficiency and costs in Australian wood chipping operations[J]. Biomass and Bioenergy, 2012, 46: 291–300. DOI:10.1016/j.biombioe.2012.08.014 [8] 陈诚, 邱荣祖. 多周期木材物流网络优化研究[J]. 福建林学院学报, 2014, 34(3): 274–278. [9] 王忠伟, 宋雨屏. 动态环境下林产品物流网络模型研究[J]. 物流技术, 2012, 31(8): 240–242. [10] KONG J, RÖNNQVIST M. Coordination between strategic forest management and tactical logistic and production planning in the forestry supply chain[J]. International Transactions in Operational Research, 2014, 21(5): 703–735. DOI:10.1111/itor.12089 [11] LAITILA J, ROUTA J. Performance of a small and a medium sized professional chippers and the impact of storage time on Scots pine (Pinus sylvestris) stem wood chips characteristics[J]. Silva Fennica, 2015, 49(5): 1–19. [12] 陈诚, 邱荣祖. 基于混合整数规划模型的木材物流网络优化[J]. 中南林业科技大学学报, 2013, 33(1): 94–98. [13] 张全全, 谭文安. 基于遗传-禁忌算法的应急救援前摄性调度优化[J]. 计算机技术与发展, 2016, 26(6): 119–122. [14] 郭强, 朱若函, 张晓萌. 基于遗传禁忌算法优化的模糊神经网络垂直切换算法[J]. 计算机应用研究, 2016, 33(3): 840–842, 847. [15] 周骞, 韦凤连, 刘菊. 基于遗传禁忌算法的公交线路发车间隔优化[J]. 交通科学与工程, 2015, 31(2): 81–86. [16] 熊杰, 杨东升, 王允森. 遗传禁忌搜索算法在工业机器人结构参数辨识上的应用[J]. 组合机床与自动化加工技术, 2015(12): 4–7, 11. [17] 吴晓燕. 路标图像识别的禁忌搜索遗传算法研究[J]. 计算机工程与设计, 2014, 35(6): 2109–2113. [18] 林玉英, 林宇洪, 邱荣祖. 木材物流决策支持系统的研究现状与展望[J]. 物流工程与管理, 2012, 34(11): 79–82. DOI:10.3969/j.issn.1674-4993.2012.11.030 [19] 周新年, 赖阿红, 周成军, 等. 山地森林生态采运研究进展[J]. 森林与环境学报, 2015, 35(2): 185–192.