Intelligent manufacturing technologies have been pursued by the industries to establish an autonomous indoor manufacturing environment. It means that tasks, which are comprised in the desired manufacturing activities, shall be performed with exceptional human interventions. This entails the employment of automated resources (i.e. machines) and agents (i.e. robots) on the shop floor. Such an implementation requires a planning system which controls the actions of the agents and their interactions with the resources to accomplish a given set of tasks. A scheduling system which plans the task executions by scheduling the available unmanned aerial vehicles and automated guided vehicles is investigated in this study. The primary objective of the study is to optimize the schedule in a cost-efficient manner. This includes the minimization of makespan and total battery consumption; the priority is given to the schedule with the better makespan. A metaheuristicbased methodology called differential evolution-fused particle swarm optimization is proposed, whose performance is benchmarked with several data sets. Each data set possesses different weights upon characteristics such as geographical scale, number of predecessors, and number of tasks. Differential evolution-fused particle swarm optimization is compared against differential evolution and particle swarm optimization throughout the conducted numerical simulations. It is shown that differential evolution-fused particle swarm optimization is effective to tackle the addressed problem, in terms of objective values and computation time.
In this paper a greedy randomized heuristic is used to solve a service technician routing and scheduling problem with time windows. Time window constraints occur in many sectors such as telecommunications, maintenance, call centres, warehouses and healthcare, and is a way of service providers differentiating themselves to maintain customer satisfaction and ultimately retain market share. The greedy randomized heuristic is coupled with a simulated annealing with restart metaheuristic and tested on 36 problem instances. The greedy randomized heuristic is compared against a parallel adaptive large neighbourhood search heuristic, presenting new best known results in 18% of the datasets.
Scheduling personnel to complete tasks is a complex combinatorial optimisation problem. In large organisations, finding quality solutions is of paramount importance due to the costs associated with staffing. In this paper we have generated and solved a set of novel large scale technician and task scheduling problems. The datasets include complexities such as priority levels, precedence constraints, skill requirements, teaming and outsourcing. The problems are considerably larger than those featured previously in the literature and are more representative of industrial scale problems, with up to 2500 jobs. We present our data generator and apply two heuristics, the intelligent decision heuristic and greedy heuristic, to provide a comparative analysis.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.