Parameter optimization of a control policy for unreliable manufacturing systems

Author(s):  
D. Song ◽  
F. Tu ◽  
S.X.C. Lou
2011 ◽  
Vol 317-319 ◽  
pp. 552-555
Author(s):  
Yi Sheng Huang ◽  
Ter Chan Row

Petri nets are employed to model flexible manufacturing systems (FMSs). However, the system deadlocked are possible happened. The conventional deadlock prevention policies are always to forbid the system entering the deadlock by using the control places. To obtain a live system, some dead markings must be sacrificed in the traditional policies. Therefore, the original reachability states of the original model can not be conserved. However, this paper is able to maintain all the reachability states of the original net and guaranty the control system live. Under our control policy, all number of reachability states of the original net will be preserved. Finally, two examples are performed that can reach the maximal permissiveness for FMSs using Petri net models (PNMs).


Author(s):  
J. Patrick Spicer ◽  
Hector J. Carlo

Scalable reconfigurable manufacturing systems (scalable-RMS) consist of standardized modular equipment that can be quickly added or removed to adjust the production capacity. Each modular machine, referred to as a scalable reconfigurable machine tool (scalable-RMT), is composed of identical modules that can be added to, or removed from the machine depending on its required throughput. In previous work, conceptual scalable-RMTs have been described. Additional scalable-RMTs are presented in this paper to highlight the applicability of this concept in manufacturing. As an extension to existing scalable-RMS literature, this paper incorporates multiple products in the system configuration design. Specifically, this paper proposes an integer programming based iterative algorithm for finding the minimum cost configuration of a multi-product system. It is shown that the proposed algorithm converges to the optimal solution under the majority of practical conditions. Then, a mathematical formulation to minimize the system investment and operational costs in a multi-product scalable-RMS is presented. A numerical example compares the solution obtained using the traditional approach of determining the system design and then the inventory control policy versus the proposed simultaneous approach. It is concluded that the simultaneous approach yields significant improvement over the traditional (decoupled) approach.


2011 ◽  
Vol 284-286 ◽  
pp. 1498-1501
Author(s):  
Yi Sheng Huang ◽  
Ter Chan Row

Deadlock prevention, deadlock detection and deadlock avoidance strategies are used to solve the deadlock problems of flexible manufacturing systems (FMSs). The conventional prevention policies were always attempt to prevent the system entering the deadlocked situation by using a few control places. On can know that one prohibits the deadlocked markings, some dead markings will be sacrificed. Therefore, the reachability states will become less than the initial net. However, our goal is to preserve all the reachability states of the initial net. Under our control policy, the deadlocks or deadlock zone will be channelized to live markings such that all the dead markings in reachability states will be conserved. Finally, an example is performed and can obtain the maximal permissiveness of a Petri net model. The other examples are all getting the same result. To our knowledge, this is the first work that employs the channelized method to prevent the deadlock problem for FMSs.


2021 ◽  
Author(s):  
Rawia Kaddachi ◽  
Ali Gharbi ◽  
Jean-Pierre Kenné

Abstract The problem of integrated production and maintenance control of unreliable manufacturing systems evolving in a stochastic and dynamic environment is studied in this paper. The considered system is subject to degradation and the produced products are perishable with random shelf-lives. The literature of operations management for perishable products reports a correlation between the shelf-life of the product and the machine degradation. In fact, the latter results in shelf-lives reduction. Ignoring this correlation effect may result in inaccurate values of the shelf-lives and inefficient control policies. The objective of this paper is to develop a joint production and maintenance control policy that minimizes the total cost composed of backlog, inventory holding, disposal and maintenance costs. The proposed parametrized joint control policy combines a multi hedging point policy and an age-based preventive maintenance policy. The optimization of the parameters of the proposed joint control policy is obtained using a simulation-based optimization approach and sensitivity analyses are provided to confirm its robustness. The obtained results show that the correlation between machine degradation and shelf-life reduction has a major influence on the control parameters and that preventive maintenance interventions can lead to increasing the shelf-life of products and minimizing the total cost. The proposed joint control policy is then compared to three other polices for a wide range of system and cost data. The obtained results show that the proposed joint control policy outperforms the other polices in terms of total incurred costs.


Processes ◽  
2020 ◽  
Vol 8 (8) ◽  
pp. 952
Author(s):  
Jia You ◽  
Ming Li ◽  
Kai Guo ◽  
Hao Li

The optimization of production cost has always been a key issue in manufacturing systems; for the single product type manufacturing systems, lots of research studies have proved the validity of the hedging point control policy in production cost control. However, due to the complexity of the multiple machines and multiple product types manufacturing systems with uncertain fault, it is difficult to achieve a good control effect only by using the hedging point control policy. To optimize the total production cost under constantly changing demands, an integrated control policy that combines the prioritized hedging point (PHP) control policy with the production capacity planning during production is proposed, and the decision variables are obtained by a particle swarm optimization (PSO) algorithm. The simulation experiments show the effectiveness of the proposed integrated control policy in production cost control for the multiple machines and multiple product types manufacturing system.


Sign in / Sign up

Export Citation Format

Share Document