• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.10a
• /
• pp.450-453
• /
• 1997

There are many studies about optimization using genetic algorithm and desirability function. It's very important to find the optimal value of something like response surface or regression model. In this study I ind~cate the problem using the old type desirability function, and suggest the new type desirabhty functton that can fix the problem better, and simulate the model. Then I'll suggest the form of desirability function to find the optimum value of response surfaces which are made by mean and standard deviation using genetic algorithm and new type desirability function.

• Korean Management Science Review
• /
• v.31 no.3
• /
• pp.61-76
• /
• 2014

There are three methods for calculating the optimal level for spare part inventories in a MIME (Multi Indenture and Multi Echelon) system : marginal analysis, Lagrangian relaxation method, and genetic algorithm. However, their solutions are sub-optimal solutions because the MIME system is neither convex nor separable by items. To be more specific, SRUs (Shop Replaceable Units) are required to fix a defected LRU (Line Replaceable Unit) because one LRU consists of several SRUs. Therefore, the level of both SRU and LRU cannot be calculated independently. Based on the limitations of three existing methods, we proposes a improved algorithm applying marginal analysis on determining LRU stock level and genetic algorithm on determining SRU stock level. It can draw optimal combinations on LRUs through separating SRUs. More, genetic algorithm enables to extend the solution search space of a SRU which is restricted in marginal analysis applying greedy algorithm. In the numerical analysis, we compare the performance of three existing methods and the proposed algorithm. The research model guarantees better results than the existing analytical methods. More, the performance variation of the proposed method is relatively low, which means one execution is enough to get the better result.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2005.10a
• /
• pp.321-333
• /
• 2005

With the limited frequency spectrum and an increasing demand for cellular communication services, the problem of channel assignment becomes increasingly important. However, finding a conflict free channel assignment with the minimum channel span is NP hard. As demand for services has expanded in the cellular segment, sever innovations have been made in order to increase the utilization of bandwidth. The innovations are cellular concept, dynamic channel assignment and hierarchical network design. Hierarchical network design holds the public eye because of increasing demand and quality of service to mobile users. We consider the frequency assignment problem and the base station placement simultaneously. Our model takes the candidate locations emanating from this process and the cost of assigning a frequency, operating and maintaining equipment as an input. In addition, we know the avenue and demand as an assumption. We propose the network about the profit maximization. This study can apply to GSM(Global System for Mobile Communication) which has 70% portion in the world. Hierarchical network design using GA(Genetic Algorithm) is the first three-tier (Macro, Micro, Pico) model, We increase the reality through applying to EMC (Electromagnetic Compatibility Constraints). Computational experiments on 72 problem instances which have 15${\sim}$40 candidate locations demonstrate the computational viability of our procedure. The result of experiments increases the reality and covers more than 90% of the demand.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.7
• /
• pp.28-34
• /
• 2006

In general engineering problems, the purpose of an optimization is to get optimal design variables. It is the same problem to fix the total amount of the design variables and to judge the optimal mixing proportions of the design variables. That is to say, we can recompose the engineering problems in the concepts of the mixture amount experiments. The goal of mixture amount method is to get the response surfaces of varying both the mixing proportion of component and the total amount of the mixture. The solution of the aircraft fuselage optimization problem is obtained by the mixture amount method and genetic algorithm. In this study, it is shown that the mixture amount method can be utilized for the aircraft structural optimization problem. Also, this method in this study can be applied for the optimization problems over 12 design variables which is impossible for D-optimal design.

• Journal of the military operations research society of Korea
• /
• v.34 no.2
• /
• pp.63-77
• /
• 2008

With the limited frequency spectrum and an increasing demand for cellular communication services, the problem of channel assignment becomes increasingly important. However, finding a conflict free channel assignment with the minimum channel span is NP hard. The innovations are cellular concept, dynamic channel assignment and hierarchical network design. We consider the frequency assignment problem and the base station placement simultaneously. Our model takes the candidate locations emanating from this process and the cost of assigning a frequency, operating and maintaining equipment as an input. Hierarchical network design using genetic algorithm is the first three-tier (Macro, Micro, Pico) model. We increase the reality through applying to Electromagnetic Compatibility Constraints. Computational experiments on 72 problem instances which have $15{\sim}40$ candidate locations demonstrate the computational viability of our procedure. The result of experiments increases the reality and covers 90% of the demand.