基于响应面和NSGA-II的AGV系统多目标优化配置

doi:10.16182/j.issn1004731x.joss.20-0908

摘要/Abstract

摘要：

AGV（automated guided vehicle）系统对于制造系统的生产柔性和生产效率具有重要作用，由于AGV系统具有许多的变量且有动态性、随机性特点，其优化配置比较复杂。提出了一种将系统仿真、数学解析和多目标优化相结合的方法，对AGV系统进行了优化配置；运用离散事件仿真模拟AGV系统运行，利用敏感性分析分离设计变量，采用析因试验和响应面方法拟合多目标优化数学模型，基于非支配解排序多目标遗传算法求解多目标优化解。通过AGV系统实例，证明了该方法的有效性，可为制造系统和物流仓储领域中AGV系统的优化配置提供一种有效的系统性分析方法。

关键词: AGV, 响应面, 离散事件仿真, 非支配解排序多目标遗传算法

Abstract:

Automated guided vehicle(AGV) system plays an important role in the production flexibility and efficiency in manufacturing systems. Due to the dynamic and stochastic characteristics of AGV system with many variables, its optimal configuration is relatively complex. A method combining system simulation, mathematical analysis and multi-objective optimization is proposed to optimize the configuration of AGV system. The discrete event simulation is used to simulate the operation of AGV system, the sensitivity analysis is used to separate design variables, the factorial experiments and response surface methods are used to build the fitting multi-objective optimization mathematical model, and the non-dominated sorting genetic algorithm-II(NSGA-II) is used to solve the multi-objective optimization problem. The effectiveness of the method is proved by an industrial case study, which provides an effective systematic analysis method for the optimal configuration of AGV system in manufacturing or logistics systems.

Key words: automated guided vehicle(AGV), response surface methodology, discrete event simulation, NSGA-II (non-dominated sorting genetic algorithm-II)

中图分类号:

TH166
TP391

付建林, 丁国富, 张剑, 江海凡, 郭沛佩. 基于响应面和NSGA-II的AGV系统多目标优化配置[J]. 系统仿真学报, 2022, 34(5): 994-1002.

Jianlin Fu, Guofu Ding, Jian Zhang, Haifan Jiang, Peipei Guo. Multi-Objective Optimization Configuration of AGV System Based on Response Surface and NSGA-II[J]. Journal of System Simulation, 2022, 34(5): 994-1002.

图/表 10

图1

图2

表1

图3

表2

图4

表3

表4

图5

表5

参考文献 23

[1]	付建林, 张恒志, 张剑, 等. 自动导引车调度优化研究综述[J]. 系统仿真学报, 2020, 32(9): 1664-1675.
	Fu Jianlin, Zhang Hengzhi, Zhang Jian, et al. Review on AGV Scheduling Optimization[J]. Journal of System Simulation, 2020, 32(9): 1664-1675.
[2]	Johnson M E, Brandeau M L. An Analytic Model for Design of a Multivehicle Automated Guided Vehicle System[J]. Management Science (S0025-1909), 1993, 39(12): 1477-1489.
[3]	Rajotia S, Shanker K, Batra J L. Determination of Optimal AGV Fleet Size for an FMS[J]. International Journal of Production Research (S0020-7543), 1998, 36(5): 1177-1198.
[4]	Arifin R, Egbelu P J. Determination of Vehicle Requirements in Automated Guided Vehicle Systems: a Statistical Approach[J]. Production Planning & Control (S0953-7287), 2000, 11(3): 258-270.
[5]	黄一钧. 车身车间AGV物料搬运系统小车数量配置规划[J]. 工业工程与管理, 2015, 20(4): 156-162.
	Huang Yijun. Planning on the Number of Vehicle Requirement for Body Shop AGV Material Handling System[J]. Industrial Engineering and Management, 2015, 20(4): 156-162.
[6]	Chawla V K. Automatic Guided Vehicles Fleet Size Optimization for Flexible Manufacturing System by Grey Wolf Optimization Algorithm[J]. Management Science Letters (S2417–2424), 2018, 8(2): 79-90.
[7]	宋绍京, 羊铭雨, 孙磊, 等. AGV数量规划及电池充电策略研究[J]. 长春理工大学学报(自然科学版), 2019, 42(6): 73-77.
	Song Shaojing, Yang Mingyu, Sun Lei, et al. Research on AGV Quantity Plan and Battery Charging Strategy[J]. Journal of Changchun University of Science and Technology (Natural Science Edition), 2019, 42(6): 73-77.
[8]	Gobal S L, Kasilingam R G. A Simulation Model for Estimating Vehicle Requirements in Automated Guided Vehicle Systems[J]. Computers & Industrial Engineering (S0360-8352), 1991, 21(1-4): 623-627.
[9]	陶翼飞, 陈君若, 刘美红, 等. 柔性制造环境下AGV车辆规模仿真优化研究[J]. 武汉理工大学学报(交通科学与工程版), 2012, 36(5): 1044-1048.
	Tao Yifei, Chen Junruo, Liu Meihong, et al. Study of AGV Fleet Size in Flexible Manufacturing Environment Based on Simulation Optimization[J]. Journal of Wuhan University of Technology (Transportation Science & Engineering), 2012, 36(5): 1044-1048.
[10]	张远春, 范秀敏, 驹田邦久. 基于仿真优化的多种类型AGV数量配置优化方法[J]. 中国机械工程, 2011, 22(14): 1680-1685.
	Zhang Yuanchun, Fan Xiumin, Kunihisa Komada. Multi-Types AGVs Quantity Configuration Optimization Based on Simulation Optimization[J]. China Mechanical Engineering, 2011, 22(14): 1680-1685.
[11]	Um I, Cheon H, Lee H C. The Simulation Design and Analysis of a Flexible Manufacturing System with Automated Guided Vehicle System[J]. Journal of Manufacturing Systems (S0278-6125), 2009, 28(4): 115-122.
[12]	Yifei T, Junruo C, Meihong L, et al. An Estimate and Simulation Approach to Determining the Automated Guided Vehicle Fleet Size in FMS[C]//2010 3rd International Conference on Computer Science and Information Technology. Chengdu: IEEE, 2010: 432-435.
[13]	Mousavi M, Yap H J, Musa S N, et al. A Fuzzy Hybrid GA-PSO Algorithm for Multi-objective AGV Scheduling in FMS[J]. International Journal of Simulation Modelling (S1726-4529), 2017, 16(1): 58-71.
[14]	姜衡, 管贻生, 邱志成, 等. 基于响应面法的立式加工中心动静态多目标优化[J]. 机械工程学报, 2011, 47(11): 125-133.
	Jiang Heng, Guan Yisheng, Qiu Zhicheng, et al. Dynamic and Static Multi-objective Optimization of a Vertical Machining Center Based on Response Surface Method[J]. Journal of Mechanical Engineering, 2011, 47(11): 125-133.
[15]	Box G, Wilson K. On the Experimental Attainment of Optimum Conditions[J]. Journal of the Royal Statistical Society (S1369-7412), 1951, 13(1): 1-45.
[16]	李沛峰, 张彬乾, 陈迎春. 基于响应面和遗传算法的翼型优化设计方法研究[J]. 西北工业大学学报, 2012, 30(3): 395-401.
	Li Peifeng, Zhang Binqian, Chen Yingchun. An Effective Transonic Airfoil Optimization Method Using Response Surface Model (RSM)[J]. Journal of Northwestern Polytechnical University, 2012, 30(3): 395-401.
[17]	田硕, 尚建勤. 基于均匀设计的喷丸指标响应面模型建立及应用[J]. 塑性工程学报, 2019, 26(4): 260-267.
	Tian Shuo, Shang Jianqin. Establishment and Application of Response Surface Model for Shot Peening Index Based on Uniform Design[J]. Journal of Plasticity Engineering, 2019, 26(4): 260-267.
[18]	张春宜, 宋鲁凯, 费成巍, 等. 柔性机构动态可靠性分析的先进极值响应面方法[J]. 机械工程学报, 2017, 53(7): 47-54.
	Zhang Chunyi, Song Lukai, Fei Chengwei, et al. Advanced Extremum Response Surface Method for Dynamic Reliability Analysis on Flexible Mechanism[J]. Journal of Mechanical Engineering, 2017, 53(7): 47-54.
[19]	Yang T, Tseng L. Solving a Multi-Objective Simulation Model Using a Hybrid Response Surface Method and Lexicographical Goal Programming Approach—A Case Study on Integrated Circuit Ink-marking Machines[J]. Journal of the Operational Research Society (S0160-5682), 2002, 53(2): 211-221.
[20]	Zhang H, Jiang Z B, Guo C T. Simulation-based Optimization of Dispatching Rules for Semiconductor Wafer Fabrication System Scheduling by the Response Surface Methodology[J]. International Journal of Advanced Manufacturing Technology (S0268-3768), 2009, 41(1/2): 110-121.
[21]	Sajadi S M, Esfahani M M S, Sorensen K. Production Control in a Failure-Prone Manufacturing Network Using Discrete Event Simulation and Automated Response Surface Methodology[J]. International Journal of Advanced Manufacturing Technology (S0268-3768), 2011, 53: 35-46.
[22]	Kalyanmoy Deb. Multi-Objective Optimization Using Evolutionary Algorithms[M]. Chichester, England: John Wiley & Sons, 2001.
[23]	Deb K, Pratap A, Agarwal S, et al. A Fast and Elitist Multi-Objective Genetic Algorithm: NSGA-II[J]. IEEE Transactions on Evolutionary Computation (S1089-778X), 2002, 6(2): 182-197.

工件类型	工序	机床	工序时长/s	工件类型	工序	机床	工序时长/s
1	O₁₁	M₂	234	4	O₄₁	M₇	277
	O₁₂	M₆	211		O₄₂	M₅	397
	O₁₃	M₈	123		O₄₃	M₁₂	255
	O₁₄	M₁₂	78		O₄₄	M₈	256
	O₁₅	M₇	264		O₄₅	M₉	216
	O₁₆	M₅	231	5	O₅₁	M₁	356
2	O₂₁	M₃	156		O₅₂	M₇	157
	O₂₂	M₄	67		O₅₃	M₉	192
	O₂₃	M₁₁	273		O₅₄	M₄	234
	O₂₄	M₁₀	223		O₅₅	M₁₂	225
	O₂₅	M₉	243		O₅₆	M₁₀	225
3	O₃₁	M₁	78	6	O₆₁	M₃	452
	O₃₂	M₄	254		O₆₂	M₂	254
	O₃₃	M₆	542		O₆₃	M₁₀	195
	O₃₄	M₈	399		O₆₄	M₁₁	236
	O₃₅	M₁₁	244		O₆₅	M₁	123
	…	…	…		O₆₆	M₄	235

设计参数	初始取值	最小增减量	取值范围	可变/固定
AGV数量(个)	3	1	[1,5]	可变
AGV速度/(m/s)	2	0.5	[1,3]	可变
AGV加速度/(m/s²)	1.5	0.5	[0.5,2.5]	固定
AGV装载时间/s	10	2.5	[5,15]	可变
AGV卸载时间/s	10	2.5	[5,15]	可变
出入口缓冲容量(个)	8	2	[4,12]	固定

No	x₁	x₂	x₃	x₄	y₁	y₂	y₃	y₄
1	2	1.5	7.5	7.5	8 773	0.799 512	0.445 52	3 500
2	4	1.5	7.5	7.5	5 120	0.765 543	0.812 203	6 380
3	2	2.5	7.5	7.5	1 824	0.705 768	0.648 24	5 092
4	4	2.5	7.5	7.5	5 886	0.523 901	0.811 137	6 374
5	2	1.5	12.5	7.5	1 047	0.748 572	0.417 823	3 281
6	4	1.5	12.5	7.5	6 451	0.738 016	0.813 193	6 388
7	2	2.5	12.5	7.5	2 118	0.641 999	0.589 434	4 630
8	4	2.5	12.5	7.5	8 985	0.514 027	0.811 232	6 373
9	2	1.5	7.5	12.5	1 028	0.749 232	0.417 263	3 276
10	4	1.5	7.5	12.5	6 265	0.738 452	0.812 872	6 386
11	2	2.5	7.5	12.5	1 998	0.642 495	0.590 785	4 641
12	4	2.5	7.5	12.5	7 667	0.514 716	0.811 775	6 379
13	2	1.5	12.5	12.5	1 103	0.704 424	0.392 879	3 084
14	4	1.5	12.5	12.5	6 601	0.697 34	0.778 196	6 113
15	2	2.5	12.5	12.5	2 063	0.589 703	0.541 42	4 253
16	4	2.5	12.5	12.5	8 923	0.504 503	0.811 642	6 379
17	2	2	10	10	1 525	0.692 047	0.510 053	4 007
18	4	2	10	10	7 048	0.613 785	0.812 635	6 386
19	3	1.5	10	10	2 944	0.746 891	0.622 647	4 890
20	3	2.5	10	10	5 649	0.616 453	0.812 066	6 380
21	3	2	7.5	10	4 370	0.715 551	0.794 983	6 245
22	3	2	12.5	10	4 795	0.660 532	0.734 758	5 771
23	3	2	10	7.5	4 345	0.715 068	0.795 499	6 249
24	3	2	10	12.5	4 751	0.661 539	0.734 395	5 769
25	3	2	10	10	4 456	0.687 192	0.763 956	6 002
26	3	2	10	10	4 545	0.687 244	0.763 495	5 998
27	3	2	10	10	4 424	0.687 147	0.764 593	6 007
28	3	2	10	10	4 475	0.687 277	0.763 663	5 998
29	3	2	10	10	4 445	0.686 992	0.764 774	6 007
30	3	2	10	10	4 444	0.687 087	0.764 416	6 004

度量值	f(1)	f(2)	f(3)	f(4)
R-Squared	0.997 506	0.991 576	0.997 775	0.998 859
Adj R-Squared	0.995 178	0.983 714	0.995 699	0.997 795
Pred-Squared	0.984 839	0.947 048	0.988 658	0.994 566
Adeq Precision	68.845 79	44.510 09	71.908 1	102.106 8
F	428.5	126.12	480.53	938.34
p-value	<0.000 1	<0.000 1	<0.000 1	<0.000 1

序号	数量(个)	速度/?(m/s)	装载时间/s	卸载时间/s	拥堵次数	AGV?利用率	机床利用率	产量(件)
1	4	1.72	5.02	5.05	4 737	0.760	0.940	7 428
2	3	1.84	5.10	5.16	3 243	0.820	0.907	7 172
3	4	1.55	5.03	5.00	4 345	0.799	0.908	7 148
4	4	1.50	5.00	5.00	4 206	0.812	0.896	7 037
5	4	1.44	5.02	5.09	4 070	0.823	0.878	6 880
6	3	1.73	5.32	5.20	3 045	0.835	0.864	6 763
7	4	1.32	5.04	5.02	3 733	0.851	0.842	6 575
8	4	1.28	5.06	5.02	3 627	0.860	0.829	6 463
9	3	1.58	5.04	5.12	2 725	0.865	0.824	6 380
10	2	2.91	5.00	5.00	1 733	0.722	0.792	6 325