Efficient Algorithms for Dynamic Cellular Manufacturing Systems by Considering Blockchain-Enabled (Case Study: Stone Paper Factory)

Manufacturing systems face various challenges that significantly influence production planning. Optimal layout design, effective production planning, and the adoption of new technologies can enhance organizational decision-making and improve production systems. In this research, a comprehensive framework is proposed for designing dynamic cellular manufacturing systems that consider order rejection, inventory management, and the integration of blockchain technology. Machines have a deterioration effect, resulting in increased processing times due to reduced efficiency. The status of the machines is recorded on a blockchain platform, and when the processing time of orders exceeds the allowable limit, the machine requires repair. Repairs are managed by contractors, with three repair types available, each varying in cost, and time. The specific type of repair is determined using a smart contract implemented on the blockchain platform. A mathematical model is proposed with two objectives: maximizing profit and maximizing the number of accepted orders. Additionally, a metaheuristic algorithm, combining a genetic algorithm (GA), artificial neural network (ANN), and boxing match algorithm (BMA), named GBA, is developed to solve the problem. The performance of the GBA is compared with other algorithms, including Particle Swarm Optimization (PSO), GA, and BMA. Furthermore, the validity of the proposed GBA is evaluated using real-world data. The cost analysis reveals that approximately 75% of the total costs are attributed to tardiness, inventory, repair, and completion time. This research offers an effective solution to reduce tardiness and completion time costs by rejecting certain orders, decreasing inventory costs through an ANN-based method, and optimizing repair costs using smart contracts. The results indicate that order rejection increases profit by an average of 38.6%. The ANN reduces the purchase of raw materials by 12.19% and decreases raw material holding by 9.74% compared to real-world data. This is while the proposed neural network, compared to the case study, has a difference of less than 4.5% in predicting the required inventory. Additionally, the smart contract improves machine availability by an average of 3.73%. The study also highlights the significant impact of the second objective, as organizations may lose approximately 2.11% of their profits to accept 1% more orders. This trade-off enables organizations to attract more customers and remain competitive in the market. Additionally, the relocation of machines that can be transferred results in an average cost reduction of up to 4.59%. Production managers can improve the production system and reduce costs by relocating machines with minimal expense.