Research on Digital Shipyard Design-procurement Coupling Based on Multi-agent Systems

doi:10.16182/j.issn1004731x.joss.26-0021

Abstract

Abstract:

To address the issues of low supply chain collaboration efficiency and inadequate dynamic scheduling capability caused by the disconnection between design and procurement in digital shipbuilding, a method integrating multi-agent simulation with the Q-learning reinforcement learning algorithm is proposed, and a four-tier dynamic supply chain model coupling design and procurement is developed, covering mines, steel mills, shipyards, and shipowners. Based on the depth of collaboration, steel mills are classified into three levels, including full collaboration, partial collaboration, and non-collaboration, with differentiated state spaces and reward functions designed to simulate intelligent pricing and game behaviors under information asymmetry. Simulation results show that shipyards need to intelligently adjust collaborative procurement strategies based on the dynamic trade-off between construction cycle and cost. This provides a quantifiable intelligent simulation method and decision-making basis for achieving precise collaboration, dynamic scheduling, and optimized decision-making throughout the entire process from design to procurement execution.

Key words: design-procurement coupling, shipbuilding supply chain management, multi-agent system, Q-learning, system simulation

CLC Number:

TH186

Peng Kaitai, Feng Dingyi, Zhao Wenbo, Chen Feier. Research on Digital Shipyard Design-procurement Coupling Based on Multi-agent Systems[J]. Journal of System Simulation, 2026, 38(10): 2857-2871.

Figures/Tables 17

Fig. 1

Fig. 2

Fig. 3

Table 1

Fig. 4

Fig. 5

Table 2

Candidate ranges of key algorithm parameters

参数	取值范围
学习率 $α$	{0.1, 0.3, 0.5, 0.7}
折扣因子 $γ$	{0.9, 0.93, 0.95, 0.97, 0.99}
探索率 $ε$	{0.1, 0.3, 0.5, 0.7, 0.9}
趋势权重 $χ$	{0.2, 0.4, 0.6, 0.8}
偏差容忍度 $υ$	{0.1, 0.2, 0.3, 0.4, 0.5}

Table 2

Table 3

Algorithm parameter values

参数

最优取值

学习率

α

0.5

折扣因子

γ

0.95

探索率 $ε$

训练次数 $κ$

趋势权重 $χ$

偏差容忍度 $υ$

0.5

200

0.6

0.2

Table 3

Fig. 6

Fig. 7

Fig. 8

Table 4

Fig. 9

Fig. 10

Fig. 11

Fig. 12

Table 5

References 21

[1]	柴志苗, 李旭东, 龚迪. 船舶结构设计优化与物资采购策略协同研究[J]. 船舶物资与市场, 2025, 33(9): 58-60.
[2]	薛雷, 胡一博, 黄爱华, 等. 造船企业与供应商产品质量改进协作模型研究[J]. 中国造船, 2021, 62(1): 202-212.
	Xue Lei, Hu Yibo, Huang Aihua, et al. Research on Collaborative Model of Product Quality Improvement between Shipyard and Supplier[J]. Shipbuilding of China, 2021, 62(1): 202-212.
[3]	Chu Niansheng, Nie Xiang, Xu Jiang, et al. A Systematic Approach of Lean Supply Chain Management in Shipbuilding[J]. SN Applied Sciences, 2021, 3(5): 572.
[4]	Pant K, Palanisamy P. Industry 4.0 in the Perspective of Supply Chain Management: Evolution and Future Research Agenda[J]. Engineering Management Journal, 2025, 37(1): 52-70.
[5]	Harichandan Sidhartha, Kumar Kar Sanjay. Financing the Hydrogen Industry: Exploring Demand and Supply Chain Dynamics[J]. Environmental Science and Pollution Research, 2025, 32(22): 13281-13298.
[6]	Centobelli Piera, Cerchione Roberto, Maglietta Amedeo, et al. Sailing Through a Digital and Resilient Shipbuilding Supply Chain: An Empirical Investigation[J]. Journal of Business Research, 2023, 158: 113686.
[7]	Ramirez-Peña Magdalena, J Sánchez Sotano Alejandro, Pérez-Fernandez Víctor, et al. Achieving a Sustainable Shipbuilding Supply Chain Under I4.0 Perspective[J]. Journal of Cleaner Production, 2020, 244: 118789.
[8]	Cebral-Fernández Mar, Rouco-Couzo Marcos, Marta Quiroga Pazos, et al. Application of a Multi-level Simulation Model for Aggregate and Detailed Planning in Shipbuilding[C]//2017 Winter Simulation Conference (WSC). Piscataway: IEEE, 2017: 3864-3875.
[9]	冯昌斌. 简析造船企业的供应链管理路径[J]. 工程与管理科学, 2024, 6(10): 115-117.
	Feng Changbin. Summarize the Supply Chain Management Path of Shipbuilding Enterprises[J]. Engineering and Management Science, 2024, 6(10): 115-117.
[10]	吴拥军, 储年生, 李锟林. 造船精益供应链管理体系的建立[J]. 船舶与海洋工程, 2020, 36(5): 65-69.
	Wu Yongjun, Chu Niansheng, Li Kunlin. Establishment of Lean Management System for the Supply Chain of Shipbuilding[J]. Naval Architecture and Ocean Engineering, 2020, 36(5): 65-69.
[11]	DivassonJ A, Macarulla A M, Garcia J I, et al. Agent-based Modeling in Urban Human Mobility: A Systematic Review[J]. Cities, 2025, 158: 105697.
[12]	Wooldridge M, Jennings N R. Intelligent Agents: Theory and Practice[J]. The Knowledge Engineering Review, 1995, 10(2): 115-152.
[13]	Yang Yongbin, Wang Mengdie, Wang Jiyuan, et al. Multi-agent Deep Reinforcement Learning for Integrated Demand Forecasting and Inventory Optimization in Sensor-enabled Retail Supply Chains[J]. Sensors, 2025, 25(8): 2428.
[14]	Okubo Yui, Mitsuyuki Taiga. Ship Production Planning Using Shipbuilding System Modeling and Discrete Time Process Simulation[J]. Journal of Marine Science and Engineering, 2022, 10(2): 176.
[15]	Wiśnicki Bogusz, Dzhuguryan Tygran, Mielniczuk Sylwia, et al. A Decision Support Model for Lean Supply Chain Management in City Multifloor Manufacturing Clusters[J]. Sustainability, 2024, 16(20): 8801.
[16]	康为. 关于造船企业物资采购管理的探讨[J]. 船舶物资与市场, 2019(3): 50-51.
[17]	姜莱. 船舶物资采购成本风险评价与规避对策分析[J]. 造船技术, 2024, 52(2): 72-79.
	Jiang Lai. Cost Risk Evaluation and Avoidance Countermeasure Analysis of Ship Material Purchase[J]. Marine Technology, 2024, 52(2): 72-79.
[18]	Salsa Tara Aska, Praharsi Yugowati, Suhardjito Gaguk. Performance Analysis and Supplier Evaluation Using Analytical Hierarchy Process and Technique for Order Preference by Similarity to Ideal Solution at the Shipyard Company[C]//Proceedings of the International Conference on Applied Science and Technology on Social Science 2022 (iCAST-SS 2022). Dordrecht: Atlantis Press, 2022: 553-562.
[19]	Tokic Michel. Adaptive ε -greedy Exploration in Reinforcement Learning Based on Value Differences[C]//KI 2010: Advances in Artificial Intelligence. Berlin, Heidelberg: Springer Berlin Heidelberg, 2010: 203-210.
[20]	Sutton R S, Barto A G. Reinforcement Learning: An Introduction[M]. Cambridge: MIT Press, 2018.
[21]	赵玉蕾. 供应链中"牛鞭效应"的成因分析及弱化措施[J]. 时代经贸, 2019(30): 52-53.
	Zhao Yulei. Analysis of the Causes of the Bullwhip Effect in the Supply Chain and Its Weakening Measures[J]. Times of Economy & Trade, 2019(30): 52-53.

钢厂类型	钢板	型钢	高强度钢
A类	2 000	1 200	200
B类	1 000	700	—
C类	600	400	—

评估指标	A类钢厂	B类钢厂	C类钢厂
趋势匹配率/%	52.9	52.3	52.5
趋势匹配次数	396	391	393
平均价格偏差/(元/t)	220	277	328
平均价格偏差率/%	4.88	6.15	7.27

统计参数	14000TEU集装箱船			20000DWT散货船			8000DWT散货船
统计参数	模型实验	对比实验	提升/%	模型实验	对比实验	提升/%	模型实验	对比实验	提升/%
平均成本	1.72×10⁸	1.87×10⁸	8.02	1.34×10⁸	1.61×10⁸	16.77	0.59×10⁸	0.68×10⁸	13.24
标准差	0.19	0.26	26.92	0.17	0.29	41.38	0.05	0.07	28.57
极差	0.56	0.76	26.32	0.51	0.67	23.88	0.19	0.21	9.52
变异系数	0.11	0.14	21.43	0.13	0.18	27.78	0.08	0.10	20.00