Kharazmi University International Journal of Supply and Operations Management 2383-1359 13 2 2026 05 01 Reinforcement Learning for Blockchain-Enabled Supply Chain Network Design 316 332 11110 10.22034/ijsom.2026.111012.3553 EN Sara Hasheminezhad Faculty of Industrial Engineering, K. N. Toosi University of Technology, Tehran, Iran Mahsa Moayed Faculty of Industrial Engineering, K. N. Toosi University of Technology, Tehran, Iran Ardavan Babaei Faculty of Industrial Engineering, K. N. Toosi University of Technology, Tehran, Iran Erfan Babaee Tirkolaee Department of Industrial Engineering, Istinye University, Istanbul, Turkey and Department of Mechanics and Mathematics, Western Caspian University, Baku, Azerbaijan 0000-0003-1664-9210 Journal Article 2026 02 12 Objective: The objective of this study is to examine the optimal design of blockchain-enabled supply chain networks using reinforcement learning (RL). The goal is to develop an integrated network framework considering simultaneously supply chain connectivity, blockchain node activation and cross-layer integration decisions, to minimize the total network cost under operational constraints.Methods: The problem is formulated as a combinatorial optimization problem and modeled as a two-layer network of supply chain entities and blockchain nodes. A Q-learning framework is used to explore feasible network configurations in the presence of stochastic costs. The model includes various cost factors, connectivity requirements, constraints for blockchain activation, and penalty mechanisms to ensure feasible solutions.Results: The results of experiments show that the proposed RL-based method is able to find out the cost-effective network designs in accordance with network connectivity constraints. The convergence behavior is shown to be stable throughout different runs, and the method converges within an earlier stage of training. The sensitivity analysis indicates that the increment of minimum number of active blockchain nodes will increase the cost of the designed network, highlighting the trade-off between blockchain deployment requirements and economic efficiency. The resulting network structure turns out to be feasible, connected, and interpretable.Conclusion: Based on the findings, the research proves the feasibility of applying reinforcement learning to blockchain-enabled supply chain network design problems. The proposed framework provides an approach that can be utilized evaluating blockchain integration strategies and balancing implementation costs with operational requirements. The research further creates a base for further research involving larger-scale networks, dynamic environments, and multi-objective optimization settings. http://www.ijsom.com/article_11110_805d94d14046509f82c7bdb1819c64af.pdf