Study on Single-Machine Group Scheduling with Due-Window Assignment and Position-Dependent Weights

This article considers a single-machine group scheduling problem with due-window assignment, where the jobs are classified into groups and the jobs in the same group must be processed in succession. The goal is to minimize the weighted sum of lateness and due-window assignment cost, where the weights depend on the position in which a job is scheduled (i.e., position-dependent weights). For the common, slack, and different due-window assignment methods, we prove that the problem can be solved polynomially, i.e., in O N log N time, where N is the number of jobs.

Group Scheduling with Learning Effect and Random Processing Time

In this paper, we establish a stochastic grouping scheduling model. In the model, there is no installation time between the jobs in the same group, but each group has an installation time before processing. There are learning effects between groups and within groups, and the completion time of jobs is a random variable. We take the long expected schedule and the expected total completion time as the objective function, and the noninterruptible static priority strategy is obtained. At the same time, heuristic algorithms and examples are given.

Actor-critic learning-based energy optimization for UAV access and backhaul networks

In unmanned aerial vehicle (UAV)-assisted networks, UAV acts as an aerial base station which acquires the requested data via backhaul link and then serves ground users (GUs) through an access network. In this paper, we investigate an energy minimization problem with a limited power supply for both backhaul and access links. The difficulties for solving such a non-convex and combinatorial problem lie at the high computational complexity/time. In solution development, we consider the approaches from both actor-critic deep reinforcement learning (AC-DRL) and optimization perspectives. First, two offline non-learning algorithms, i.e., an optimal and a heuristic algorithms, based on piecewise linear approximation and relaxation are developed as benchmarks. Second, toward real-time decision-making, we improve the conventional AC-DRL and propose two learning schemes: AC-based user group scheduling and backhaul power allocation (ACGP), and joint AC-based user group scheduling and optimization-based backhaul power allocation (ACGOP). Numerical results show that the computation time of both ACGP and ACGOP is reduced tenfold to hundredfold compared to the offline approaches, and ACGOP is better than ACGP in energy savings. The results also verify the superiority of proposed learning solutions in terms of guaranteeing the feasibility and minimizing the system energy compared to the conventional AC-DRL.