Scheduling jobs on parallel machines applying neural network and heuristic rules

The problem of scheduling jobs on identical parallel machines is investigated. The jobs are assumed to have sequence dependent setup times independent of the machine. Each job has a processing time, a due date, and a weight for penalizing tardiness. The objective of scheduling is to fine a sequence of the jobs which minimizes the sum of weighted tardiness. An extension of the ATCS (Apparent Tardiness Cost with Setups) rule which utilizes some look-ahead parameters for calculating the priority index of each job is proposed. An additional factor for measuring the problem characteristics is introduced and a neural network is utilized to get more accurate values of the look-ahead parameters.