Hybrid Deep Neural Network Scheduler for Job-Shop Problem Based on Convolution Two-Dimensional Transformation

Table 1

Description and formula of hybrid deep neural network input.

Feature	Description	Formula
	Position order [3]
	Ratio of machine index i to machine number m [23]
	Ratio of job index j to job number n [23]
	Remaining processing time of job j [3]
	Ratio of operation processing time to total processing time [23]
	Ratio of operation processing time to processing time of machine i [23]
	Ratio of operation processing time to processing time of job j [11]
	Ratio of processing time of machine i to total processing time [11]
	Ratio of job j’s processing time to total job processing time
	Ratio of job j’s processing time to processing time of machine i

References indicate that this feature has been used in the corresponding literature.