Uniform parallel machine scheduling problems with a truncation sum-of-logarithm-processing-times-based learning effect

Abstract

In internet manufacturing, a manufacturing cloud usually deals with multiple requests at the same time. Similar problems may appear if a manufacturing cloud is equipped with some kind of intelligence and can learn to process a later job more quickly. This research considers the problem of scheduling jobs on uniform parallel machines to minimise total weighted tardiness (TWT) under a truncation sum-of-logarithm-processing-times-based learning effect. In the proposed learning model, the actual job processing time is a function that depends not only on the processing times of the jobs already processed but also on a control parameter. An iterated local search (ILS) has been proposed to solve this problem. Computational results of applying the ILS to some cases show that the proposed ILS outperforms other metaheuristics (simulated annealing and ant colony optimisation) in terms of TWT. The present research also analyses the impacts of truncation learning effects on objective values.