This paper addresses the batch scheduling problem of unrelated parallel machines attempting to minimize the total weighted tardiness. Identical or similar jobs are typically processed in batches to decrease setup and/or processing times. Local dispatching rules such as the earliest weighted due date, the shortest weighted processing time, and the earliest weighted due date with a process utilization spread are tailored to the batch scheduling requirements. Based on the features of batch scheduling, a two-level batch scheduling framework is suggested. Existing heuristics, which show excellent performance in terms of total weighted tardiness for the single machine scheduling, such as the modified earliest due date rule and the modified cost over time rule, are extended for the problem. The simulated annealing algorithm as a meta-heuristic is also presented to obtain near optimal solutions. The proposed heuristics are compared through computational experiments with data from the dicing process of a compound semiconductor manufacturing facility

As K-wave has been strengthened via recent K-contents, K-wave items such as cosmetics and electronic devices have also gained attention globally. For small-sized export sellers who purchased the items and exported them to different countries, it is significant to discover which K-wave items are trending in specific countries. To do so, we proposed an ensemble recommender system by producing the dense vector, which is generated by a variant of Bidirectional Encoder Representations from Transformers (BERT), and balancing the vector with a sparse vector in order to ensure the efficient execution speed and recommendation accuracy. Based on the data we have collected specifically for potential K-items, our experiment showed that the proposed model outperforms the various baselines, which are used for content-based filtering.