w1346947355/FJSP-RD

FJSP-RD

FJSP-RD is a supervised fine-tuning dataset developed for LLM-ES, a sequential decision-making paradigm for Flexible Job Shop Scheduling (FJSP). The dataset is designed to support research on large language models for FJSP solving. Each sample represents a scheduling problem together with structured task requirements and a complete scheduling decision trajectory, enabling models to learn FJSP solving through sequential scheduling decisions. FJSP-RD contains 30,000 English-language samples and is primarily intended for supervised fine-tuning of large language models in academic research.