Optimization in science and engineering
zohre kiapasha; ali salmasnia
Abstract
Purpose: Cloud manufacturing is a service-oriented production model that centralizes production resources available in different geographical locations to respond to specific customer needs. One of the main issues in cloud manufacturing systems is the proper allocation of sub-tasks to enterprises and ...
Read More
Purpose: Cloud manufacturing is a service-oriented production model that centralizes production resources available in different geographical locations to respond to specific customer needs. One of the main issues in cloud manufacturing systems is the proper allocation of sub-tasks to enterprises and their optimal scheduling. Most studies in the literature assume that all tasks have only one type of structure, although it is possible to have tasks with different structures in one order set. Furthermore, existing scheduling models in the cloud manufacturing literature tend to assume that all tasks are available at time zero and that the logistics time/cost among enterprises is negligible. Therefore, in this study, an optimization model with three objective functions of task completion time, the cost imposed on the cloud manufacturing system, and the quality of the selected services is developed in which to get closer to the real world, three features are included in it: 1) the possibility of tasks with two structures, series and parallel, 2) different arrival times of tasks in the cloud manufacturing system, and 3) time/cost of logistics between different enterprises.
Methodology: First, six examples with different numbers of tasks and subtasks are designed with both sequential and parallel structures. In order to accurately solve the proposed model and achieve the global optimum, the CPLEX solver is used in the GAMS software.
Findings: In order to verify the importance of the characteristics of the developed model, two comparative studies are carried out. In the first comparative study, the presented model is compared with a similar model in which it is assumed that all tasks are available at time zero. The second comparative study examines the effect of considering logistics time/costs between enterprises when allocating subtasks to services. The results of the comparative studies show the misleading level of the cloud manufacturing manager in the face of the reduced models.
Originality/Value: The output of this research is to present a model for the simultaneous scheduling of tasks with sequential and parallel structures, taking into account the different task arrival times and logistics in the cloud manufacturing system.
Scheduling Modeling
Roja Ruhbakhsh; Esmaeil Mehdizadeh; Mohammad Amin Adibi
Abstract
Purpose: Lot streaming, which has much attention in recent years, is an effective technique to increase production efficiency in a production system by splitting a job into several smaller parts in a multi-stage production system. But important assumptions that exist in the real-world scheduling environment ...
Read More
Purpose: Lot streaming, which has much attention in recent years, is an effective technique to increase production efficiency in a production system by splitting a job into several smaller parts in a multi-stage production system. But important assumptions that exist in the real-world scheduling environment are always ignored. Hence, in this paper, these assumptions are discussed and the results are reviewed. In this paper, the aim is solving a multi objective mathematical model for solving hybrid flow shop scheduling problem with lot-streaming, setup time and transportation time.Methodology: At first, a multi objective mathematical programming model is presented for solving the problem. Then, by wighting method, the multi objective model convert to single objective model and GAMS software is used to solve the small size problems to show the performance of the mathematical mothel. Inspired by previous studies, two multi objective metaheuristic algorithms based on the genetic algorithm is used to solve the large-scale problems. To illustrate the performance of the proposed metaheuristic algorithms, the obtained results of the algorithms compared with GAMS outputs in single mode.Findings: To validate the proposed model, a sample is solved using GAMS software and compared with the genetic algorithm. The obtained results show the performance of the mathematical model. Then, two proposed algorithms are used to solve the large-scale problems. For this purpose, 30 instance problems are randomly generated and six indicators are used to compare the algorithms. After performing the experiments and comparing the algorithms with each other, the results show NRGA algorithm performs bether than NSGA-II.Originality/Value: In this paper, for solving a multi objective hybrid flow shop scheduling problem with lot-streamingm mathematical model with the aim of minimizing the makespan and total tardiness, the sequence-dependent setup time and the transportation time constraints between consecutive stages are considered. Since the problem is NP-hard, NSGA-II and NRGA algorithms were used to solve the proposed problem.