基于Agent的体系过程A-GERT网络“刺激-反应”学习模型

doi:10.12305/j.issn.1001-506X.2022.08.19

摘要/Abstract

摘要：

针对图示评审技术(graphic evaluation and review technique, GERT)网络应用于体系活动决策过程中未考虑网络节点自身的学习能动性问题, 基于智能体(Agent)技术, 结合刺激-反应模型构建A-GERT网络模型。首先, 根据体系活动的逻辑, 搭建A-GERT网络框架, 并通过提出网络传递效用函数, 设计带有反馈机制的A-GERT网络。然后, 利用刺激-反应模型的自适应优势构建学习方程, 以GERT网络期望概率与期望时间等指标度量刺激强度, 进一步拓展刺激-反应模型。最后, 给出网络智能决策节点的学习步骤, 并以创新技术开发体系活动进行算例研究, 结果表明了所提方法的有效性与实用性。

关键词: 图示评审技术, 智能体技术, 复杂适应系统, 刺激-反应模型, 自适应学习

Abstract:

In view of the graphic evaluation and review technique (GERT) network applied in the decision-making process of system activities without considering the learning initiative of network nodes themselves, an A-GERT network model is developed based on Agent technology and stimulus-response model. Firstly, the framework of the A-GERT network is developed, and through the utility function of network transmission, the A-GERT network with feedback mechanism is designed. Then, we employ the adaptive advantages of the stimulus-response model to design the learning equation, and measure the stimulus intensity with the expected probability and expected time of the GERT network, so as to further expand the stimulus-response model. Finally, the learning steps of network intelligent decision nodes are given, and an example of innovation technology development system is given. The results show the effectiveness and practicability of the proposed method.

Key words: graphic evaluation and review technique (GERT), Agent technology, complex adaptive system, stimulus-response model, adaptive learning

中图分类号:

N945

方志耕, 夏悦馨, 张靖如, 熊仪, 陈静邑. 基于Agent的体系过程A-GERT网络“刺激-反应”学习模型[J]. 系统工程与电子技术, 2022, 44(8): 2540-2553.

Zhigeng FANG, Yuexin XIA, Jingru ZHANG, Yi XIONG, Jingyi CHEN. A stimulus-response learning model for Agent-based system process A-GERT network[J]. Systems Engineering and Electronics, 2022, 44(8): 2540-2553.

图/表 23

图1

图2

图3

图4

图5

图6

图7

图8

图9

表1

图10

图11

图12

图13

图14

图15

图16

图17

图18

表2

图19

表3

表4

参考文献 30

1	PRITSKER A A B . GERT: graphical evaluation and review technique, Part I, fundamental[J]. Journal of Industrial Engineering, 1996,
2	刘红旗, 方志耕, 陶良彦. 复杂装备研制项目进度规划GERT网络"反问题"模型[J]. 系统工程与电子技术, 2015, 37 (12): 2758- 2763. doi: 10.3969/j.issn.1001-506X.2015.12.15
	LIU H Q , FANG Z G , TAO L Y . Complex equipment development project planning GERT network "inverse problem" model[J]. Systems Engineering and Electronics, 2015, 37 (12): 2758- 2763. doi: 10.3969/j.issn.1001-506X.2015.12.15
3	方志耕, 杨保华, 陆志鹏, 等. 基于Bayes推理的灾害演化GERT网络模型研究[J]. 中国管理科学, 2009, 17 (2): 102- 107. doi: 10.3321/j.issn:1003-207X.2009.02.016
	FANG Z G , YANG B H , LU Z P , et al. The GERT network model study of disaster evolution based on Bayes inference[J]. Chinese Journal of Management Science, 2009, 17 (2): 102- 107. doi: 10.3321/j.issn:1003-207X.2009.02.016
4	王欢, 方志耕, 邓飞, 等. 考虑质量价值水平的复杂产品供应链质量成本优化方法[J]. 控制与决策, 2019, 34 (9): 1973- 1980.
	WANG H , FANG Z G , DENG F , et al. Quality cost optimization method for complex product supply chain considering quality value[J]. Control and Decision, 2019, 34 (9): 1973- 1980.
5	方志耕, 邵瑞瑞, 王召, 等. 高轨卫星通信星座PS-GERT效能评估模型[J]. 系统工程与电子技术, 2020, 42 (10): 2356- 2365. doi: 10.3969/j.issn.1001-506X.2020.10.26
	FANG Z G , SHAO R R , WANG Z , et al. PS-GERT effectiveness evaluation model of GEO satellite communication constellation[J]. Systems Engineering and Electronics, 2020, 42 (10): 2356- 2365. doi: 10.3969/j.issn.1001-506X.2020.10.26
6	陈东平. 随机网络GERTS的关键线路法GERTS/CPM[J]. 水力发电, 1994, (1): 50- 52.
	CHEN D P . Critical line method GERTS/CPM for random network GERTS[J]. Water Power, 1994, (1): 50- 52.
7	杨保华, 方志耕, 刘思峰, 等. 基于GERTS网络的非常规突发事件情景推演共力耦合模型[J]. 系统工程理论与实践, 2012, 32 (5): 963- 970. doi: 10.3969/j.issn.1000-6788.2012.05.007
	YANG B H , FANG Z G , LIU S F , et al. Model of co-coupling in unconventional incidents based on GERTS network[J]. Systems Engineering-Theory & Practice, 2012, 32 (5): 963- 970. doi: 10.3969/j.issn.1000-6788.2012.05.007
8	郭本海, 陈玮, 吕东东. 基于GERT网络的战略性新兴产业技术瓶颈探测模型[J]. 系统管理学报, 2017, 26 (4): 728- 736.
	GUO B H , CHEN W , LYU D D . A GERT network model for detection of technology bottleneck in strategic emerging industry[J]. Journal of Systems & Management, 2017, 26 (4): 728- 736.
9	郭本海, 张笑腾, 甄美荣, 等. 基于多目标的企业创新资源优化配置GERT网络模型[J]. 科技管理研究, 2018, 38 (22): 161- 168. doi: 10.3969/j.issn.1000-7695.2018.22.025
	GUO B H , ZHANG X T , ZHEN M R , et al. A GERT network model about enterprise innovation optimal configuration of resources based on multi objectives[J]. Science and Technology Management Research, 2018, 38 (22): 161- 168. doi: 10.3969/j.issn.1000-7695.2018.22.025
10	郭本海, 彭莹, 薛会娟. 知识互溢视角下新能源汽车产业链功能演化GERT网络模型研究[J]. 科技进步与对策, 2021, 38 (2): 65- 74.
	GUO B H , PENG Y , XUE H J . The research of GERT network model of functional evolution of new energy automobile industry chain from the perspective of knowledge mutual spillover[J]. Science & Technology Progress and Policy, 2021, 38 (2): 65- 74.
11	张海涛, 李题印, 徐海玲, 等. 商务网络信息生态链价值流动的GERT网络模型研究[J]. 情报理论与实践, 2019, 42 (9): 35- 40.
	ZHANG H T , LI T Y , XU H L , et al. Research on GERT network model of value flow in business network information ecological chain[J]. Information Studies: Theory & Application, 2019, 42 (9): 35- 40.
12	陈妙云, 王雷, 丁治强. 基于DP-SAMQ行为树的智能体决策模型研究[J]. 计算机仿真, 2021, 38 (2): 301- 307. doi: 10.3969/j.issn.1006-9348.2021.02.065
	CHEN M Y , WANG L , DING Z Q . Research on Agent decision model based on multi-step Q-learning behavior tree[J]. Computer Simulation, 2021, 38 (2): 301- 307. doi: 10.3969/j.issn.1006-9348.2021.02.065
13	张杰, 王刚, 宋亚飞, 等. 基于自适应SGD-多智能体的防空资源部署优化[J]. 系统工程与电子技术, 2019, 41 (7): 1536- 1543.
	ZHANG J , WANG G , SONG Y F , et al. Optimization of air defense resource deployment based on adaptive SGD-multi-agent[J]. Systems Engineering and Electronics, 2019, 41 (7): 1536- 1543.
14	彭程. 多智能体情感决策学习方法及其在Flow智能交通中的应用[D]. 北京: 北京化工大学, 2020.
	PENG C. Multi-AGERT affective decision learning method with its application in Flow intelligent transportation[D]. Beijing: Beijing University of Chemical Technology, 2020.
15	WOOLDRIDGE M , JENNINGS N R . Intelligent agents: theory and practice[J]. Knowledge Engineering Review, 1995, 10 (2): 115- 152. doi: 10.1017/S0269888900008122
16	蒋子涵. 基于多智能体系统的共因失效系统故障预测GERT网络模型[D]. 南京: 南京航空航天大学, 2019.
	JIANG Z H. The GERT network for fault prediction of common cause failure systems based on multi-Agent system[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2019.
17	UCHIYAMA M . Formation of high-speed motion pattern of a mechanical arm by trial[J]. Transactions of the Society of Instrument and Control Engineers, 1978, 14 (6): 706- 712. doi: 10.9746/sicetr1965.14.706
18	皮道映, 孙优贤. 非线性时变系统开闭环P型迭代学习控制的收敛性[J]. 自动化学报, 1999, (3): 67- 70.
	PI D Y , SUN Y X . The convergence of iterative learning control with open-closed-loop P-type scheme for nonlinear time-varying systems[J]. Acta Automatica Sinica, 1999, (3): 67- 70.
19	孙明轩, 黄宝健, 张学智. 非线性系统的PD型迭代学习控制[J]. 自动化学报, 1998, (5): 137- 140.
	SUN M X , HUANG B J , ZHANG X Z . PD-type iterative learning control for a class of nonlinear systems[J]. Acta Automatica Sinica, 1998, (5): 137- 140.
20	CHI R H , HOU Z S , XU J X . Adaptive ILC for a class of discrete-time systems with iteration-varying trajectory and random initial condition[J]. Automatica, 2008, 44 (8): 2207- 2213. doi: 10.1016/j.automatica.2007.12.004
21	LIN N , CHI R H , HUANG B , et al. Multi-lagged-input iterative dynamic linearization based data-driven adaptive iterative learning control[J]. Journal of the Franklin Institute-Engineering and Applied Mathematics, 2019, 356 (1): 457- 473. doi: 10.1016/j.jfranklin.2018.10.010
22	CHEN J X , LI J M , ZHAO W J . T-S fuzzy model-based adaptive repetitive consensus control for multi-agent systems with imprecise communication topology structure[J]. International Journal of Systems Science, 2019, 50 (8): 1568- 1579. doi: 10.1080/00207721.2019.1617367
23	CHEN J X , LI J M . Fuzzy adaptive iterative learning coordination control of second-order multi-agent systems with imprecise communication topology structure[J]. International Journal of Systems Science, 2018, 49 (3): 546- 556. doi: 10.1080/00207721.2017.1412533
24	邝雄, 李忠杰. 基于复杂适应系统模拟的农村基础设施投融资博弈分析[J]. 系统科学与数学, 2021, 41 (3): 768- 787.
	KUANG X , LI Z J . The game analysis on investment and financing of rural infrastructure based on complex adaptive system simulation[J]. Journal of Systems Science and Mathematical Sciences, 2021, 41 (3): 768- 787.
25	蒙玉玲, 董晓宏, 刘润. 云创新助推科技型中小企业构建持续性学习机制[J]. 经济与管理, 2020, 34 (4): 82- 87.
	MENG Y L , DONG X H , LIU R . Cloud innovation helps technological SMEs build a continuous learning mechanism[J]. Economy and Management, 2020, 34 (4): 82- 87.
26	慕静, 马丽. 复杂适应系统理论视角下科技服务业创新模式研究[J]. 科技管理研究, 2019, 39 (9): 158- 162. doi: 10.3969/j.issn.1000-7695.2019.09.023
	MU J , MA L . Research on innovation mode of science and technology service industry from the perspective of complex adaptive system theory[J]. Science and Technology Management Research, 2019, 39 (9): 158- 162. doi: 10.3969/j.issn.1000-7695.2019.09.023
27	龙跃. 基于刺激—反应模型的产业技术创新联盟知识创新研究[J]. 当代经济管理, 2018, 40 (3): 19- 24.
	LONG Y . Study on the knowledge innovation of industrial technology innovation alliance based on stimulus-response model[J]. Contemporary Economic Management, 2018, 40 (3): 19- 24.
28	冯允成. 随机网络及其应用[M]. 北京: 北京航空学院出版社, 1987.
	FENG Y C . Random network and its application[M]. Beijing: Beihang University Press, 1987.
29	奥斯卡·兰格. 经济控制论导论[M]. 北京: 中国社会科学出版社, 1981.
	OSKAR L . Introduction to economic cybernetics[M]. Beijing: China Social Sciences Press, 1981.
30	周东浩, 韩文报. DiffRank: 一种新型社会网络信息传播检测算法[J]. 计算机学报, 2014, 37 (4): 884- 893.
	ZHOU D H , HAN W B . DiffRank: a novel algorithm for information diffusion detection in social networks[J]. Chinese Journal of Computers, 2014, 37 (4): 884- 893.

节点i-j	初始概率p_ij	时间T_ij	传递函数W_ij
1-2	1	2	W₁₂=e^2s
2-1	0.5	3	W₂₁=0.5e^3s
2-3	0.5	2	W₂₃=0.5e^2s
3-4	0.4	4	W₃₄=0.4e^4s
3-5	0.6	6	W₃₅=0.6e^6s
4-4	0.7	8	W₄₄=0.7e^8s
4-6	0.3	6	W₄₆=0.3e^6s
5-7	1	10	W₅₇=e^10s
6-8	0.5	8	W₆₈=0.5e^8s
6-9	0.5	8	W₆₉=0.5e^8s
7-5	0.2	10	W₇₅=0.2e^10s
7-8	0.6	10	W₇₈=0.6e^10s
7-9	0.2	10	W₇₉=0.2e^10s

学习方式	节点6	节点4	节点7	节点3	节点2
不学习	0	0	0	0	0
a	127	98	136	89	84
b	127	100	152	87	84
c	101	92	135	82	67
d	101	92	143	80	65

学习方式	[p₂₃, p₃₄, p₃₅, p₄₆, p₆₈, p₇₈]	期望概率p_E₁₈	期望时间T_E₁₈
不学习	[0.5, 0.4, 0.6, 0.3, 0.5, 0.6]	0.65	41.743 5
a	[0.856 6, 0.091, 0.909, 0.967 5, 0.832 5, 0.952 3]	0.953 3	30.477 6
c	[0.857 1, 0.090 9, 0.909 1, 0.967 7, 0.833 3, 0.952 4]	0.957	30.542 9
b	[0.856 6, 0.909, 0.091, 0.967 5, 0.832 5, 0.952 3]	0.844 6	23.938 6
d	[0.857 1, 0.909 1, 0.090 9, 0.967 7, 0.833 3, 0.952 4]	0.845 3	23.933 8

方法	[p₂₃, p₃₄, p₃₅, p₄₆, p₆₈, p₇₈]	p_E₁₈	T_E₁₈
文献[11]方法	[0.55, 0.3, 0.7, 0.45, 0.541 7, 0.8]	0.863	38.484 2
本文方法	[0.857 1, 0.090 9, 0.909 1, 0.967 7, 0.833 3, 0.952 4]	0.957	30.542 9

[1]	陈都, 孟秀云. 基于自适应郊狼算法的无人机离线航迹规划[J]. 系统工程与电子技术, 2022, 44(2): 603-611.
[2]	方志耕, 邵瑞瑞, 王召, 刘思峰, 游伟青, 高素. 高轨卫星通信星座PS-GERT效能评估模型[J]. 系统工程与电子技术, 2020, 42(10): 2356-2365.
[3]	陶良彦, 刘思峰, 方志耕, 陈顶. GERT网络的矩阵式表达及求解模型[J]. 系统工程与电子技术, 2017, 39(6): 1292-1297.
[4]	刘红旗1,2, 方志耕1,2, 陶良彦1,2. 复杂装备研制项目进度规划GERT网络“反问题”模型[J]. 系统工程与电子技术, 2015, 37(12): 2758-2763.
[5]	徐哲，贾子君. 基于仿真的武器装备研制系统性能风险评估[J]. Journal of Systems Engineering and Electronics, 2011, 33(4): 793-798.