A bi-objective optimization mathematical model integrated with bulk arrival Markovian queueing system and machine learning in publishing industries

In the publishing industry, it is vital to comprehend customers' needs and take their pleasure into account. Thus, publishing houses (PHs) must effectively address the necessity to increase profitability as well as minimize the waiting time of the printing requests in the queue. Although some researchers have studied publishing industries' management, there is still a lack of research modeling PHs as a practical queueing system that reflects their performance precisely. Moreover, most models have not considered the machine learning technique for predicting the circulation of books integrated with a queueing model and mathematical programming. Furthermore, no study has modeled the PH as a M[X]/M/1 queueing system to practically improve the system. To fill these gaps, this paper develops a novel bi-objective mathematical model with a bulk arrival queueing system and a machine learning technique to minimize the total cost and maximize the resource utilization rate. Also, different machine learning classifier algorithms are implemented to estimate the circulation for each publishing product. Owing to the model's bi-objective character, an enhanced LP-metrics method is presented. We have also examined a real-world case study to validate our approach and demonstrate its applicability to real-world issues. A 25% increase in profitability served as proof of the model's effectiveness.