JOURNAL OF INDUSTRIAL ENGINEERING RESEARCH IN PRODUCTION SYSTEMS (IERPS)
Year:2018 | Volume:6 | Issue:12 |Papers Count:7

In this paper, three-machine scheduling problem in robotic production cells that producing different parts and assuming the tool change has been investigated. The main issue in this article is make decisions about how to move the parts by a robot between machines, the order of parts to robotic production cells as well as how to tool switch on the machine is to achieve the optimum cycle time. A new approach to calculate the number of tool switching is provided. Then a programming model for minimizing cycle time is presented in this problem and the results showed that tool changing and parts sequencing has a direct impact on robot movement policy and cycle time. Finally, numerical examples have been studied using Genetic algorithms. The results were compared with software GAMS. It has been proven that GAMS does not answer for more than eight parts in a reasonable time, but examples up to 80 different parts using Genetic Algorithm reach to answer in a reasonable time. The results showed that tool switching time and parts sequencing have direct impact on cycle time.

Process planning and scheduling are two key sub-functions in the manufacturing system. Traditionally, these two, were carried out in separate and sequential way with a single criterion optimization and regard to some hypothesizes. In real-world these hypothesizes such as resources and machines permanent availability and process planning inflexibility make the solution will become infeasible. In this paper to improve efficiency and adapt more to the real-world production, with four criteria, alternative operation sequences and dynamic feature such as machine breakdown and new order arrival used to optimize integrated process planning and scheduling (IPPS) problem. In solving problem process, cooperative game theory based on compromise method has been developed and a meta heuristic hybrid algorithm (GA, TS) are used. The approach has been tested and the result show that the developed approach is a proper method to solve a multi objective IPPS with supposed constraints.

Growths of technology and innovation have made continual changes in fashion business and costumer tastes. In situations like that, retailers and manufacturers select agility and flexibility as their main supply chain strategies. In this study, a bi-level model including the retailer and manufacturer who traditionally compete on the product quantity and price is proposed; then another model is developed by adding some characteristics of agility to first model. In the proposed model, in addition to the competition between the supply chain members, influences of the customers’ behavior on the decisions of supply chain members are considered. This study is aimed at proposing efficient solutions for determining the price and quantity of ordering and production, considering the situations of competition, customers and market toward maximization of the manufacturers’ and retailers’ profit. The proposed bi-level model is converted to a single-level one using the Karush-Kuhn-Tucker (KKT) and the results of the model are investigated and discussed by employing in a numerical example. Results show that the retailer and manufacturer, by making proper and precise decisions, can increase their sale price. Further, by improving their decisions, they can reduce the product clearance sale at the end of the sales season, which ends in the growth of profit for both of the supply chain members.

In this paper, we investigate location-allocation problem for multi-layer congestible facilities. In many real word location situations, a service center is not capable of serving all the simultaneous requests made for the service and as a result forming queues and congestion is inevitable. For this purpose, a multi-objective nonlinear integer programming model for queuing facility location problem with the same framework to the M/M/1 series queuing network is designed, in which facilities have several layers and customers should pass all the layers for service completion. The objective functions of the model are minimizing the sum of customers traveling times to facilities and waiting times in the system, and minimizing the maximum idle probability of the facilities. The proposed mathematical model is validated by sensitivity analysis, and the effect of the probable variations of the parameters on the Pareto solution is investigated. The results show that the model behaves correctly to the sensitive parameters of the problem. To evaluate the model, some numerical experiments are presented and solved with the Augmented ε-constraint technique of multi-objective optimization as well. The appropriate location among potential sites for appropriate number of facilities and allocation of customers to facilities of each layer are determined by Pareto optimal solutions found.

In this paper, integrated scheduling of production and distribution with vehicle routing problem is considered. A manufacturer with parallel production lines receives customer orders; after producing them, they are then delivered to the customers in batches by a fleet of vehicles. Unlike a direct delivery of products from the manufacturer to each customer, batch delivery reduces the transportation costs because of the maximum utilization of the vehicle capacities, but it may increase the holding and tardiness costs. The objective is to find an integrated schedule of production and distribution so as to minimize the setup, holding, distribution and tardiness costs. The problem is first formulated as a mixed integer linear programming model. In view of its NP-hardness, a procedure by incorporating dominance properties with imperialist competitive algorithm is then proposed to solve large-sized problem instances. To evaluate the performance of the proposed algorithm, several instances are generated and solved. Computational results demonstrate that the algorithm has a good performance for large problems.

Service Composition (SC) is an important problem in the Cloud Manufacturing (CM) paradigm in which, after receiving customers’ requests, a composition of cloud services is determined for accomplishment of their needs. Actually, a customer’ s need is decomposed to some distinct tasks such that a manufacturing resource or a group of them can perform each task. The ultimate goal of SC is optimal assignment of tasks to resources while specific objective function(s) and constraint(s) are considered. In this paper, as the main contribution, an Integer Programming (IP) model of service composition problem is presented which includes transportation between manufacturing resources scattered over the globe. Then, two different scenarios of SC problem are generated such that each center of Iran's provinces includes a resource which can perform some pre-determined tasks. In addition, distance of center of Iran's provinces as well as transportation time between them are estimated based on real data. In order to solve the mentioned scenarios, Branch and Bound (B&B) exact algorithm is used. Furthermore, before developing a metaheuristic algorithm for implementation in service composition problem, landscape analysis of the problem is completed. Based on the results of this analysis, the problem has a random uniform nature and its local optima are scattered over the search space. As a result, simple singlesolution based algorithms such as Local Search (LS) heuristic can be efficient in SC problem solving. Results of comparison between B&B and LS algorithms indicate superiority of LS in finding optimum or near-optimum solution with lower computational cost.

In this research, the scheduling of loading and distribution for high consumption oil products from dispersed warehouses is studied. The high consumption of oil products requires direct shipment of orders issued by different customers in different geographical locations. In this paper, an integrated nonlinear model is proposed to minimize the total cost of the supply chain including purchasing cost, loading, transportation, and late delivery costs. Because of the NP-hard nature of problem, a new hybrid solution approach based on simulation and meta-heuristic algorithms is proposed. Operations simulation ensure always feasible salutation and meatheuristic algorithms provide intelligent search within the feasible solutions. The efficiency of proposed solution approach is evaluated through several random test problems. Moreover, the performance of a latest versions of genetic algorithm (GA-MPC) and particles swarm optimization (SPSO 2011) as well as the traditional versions are evaluated within the proposed solution approach to solve the real-world scale problems. The results illustrate the better performance of genetic algorithm for large-scale problems.