• The KIPS Transactions:PartB
• /
• v.14B no.2
• /
• pp.127-134
• /
• 2007

There are many researches about the intelligent stock trading systems with the help of the advance of the artificial intelligence such as machine learning techniques, Though the establishment of the reasonable trading policy plays an important role in the performance of the trading systems most researches focused on the improvement of the predictability. Also some previous works, which treated the trading policy, treated the simplified versions dependent on the predictors in less systematic ways. In this paper, we propose the integrated multiple simulation' as a method of optimizing trading performance of stock trading systems. The propose method is adopted in the NXShell a development environment for neural network based stock trading systems. Under the proposed integrated multiple simulation', we simulate the multiple tradings for all combinations of the neural network's outputs and the trading policy parameters, evaluate the learning performance according to the various metrics and establish the optimal policy for a given prediction module based on the resulting performance. In the experiment, we present the trading policy comparison results using the stock value data from the KOSPI and KOSDAQ.

• Korean System Dynamics Review
• /
• v.10 no.1
• /
• pp.109-129
• /
• 2009

This paper explores the ways by which the authority concerned with nuclear energy policy-making in Korea can strengthen its organisational sustainability from long-term perspective. In doing so, it applies the system dynamics approach to predict what would happen to the organisational sustainability of the nuclear energy authority in the future. In the process of analysis, it also draws causal loop map of components contained in the simulation model and constructs user-interface simulation model. It shows different predicted future values regarding organisational sustainability of the nuclear energy authority in Korea and puts forward some policy implications for practitioners and academics involved in nuclear energy policy.

• Convergence Security Journal
• /
• v.18 no.3
• /
• pp.3-9
• /
• 2018

The Deep Deterministic Policy Gradient (DDPG) algorithm is an algorithm that learns by using artificial neural network s and reinforcement learning. Among the studies related to reinforcement learning, which has been recently studied, the D DPG algorithm has an advantage of preventing the cases where the wrong actions are accumulated and affecting the learn ing because it is learned by the off-policy. In this study, we experimented to control the bicycle autonomously by applyin g the DDPG algorithm. Simulation was carried out by setting various environments and it was shown that the method us ed in the experiment works stably on the simulation.

• Korean System Dynamics Review
• /
• v.8 no.1
• /
• pp.173-186
• /
• 2007

This research consists of (1) building (or molding) the Dynamics Simulation Model on Return Policy mainly used in publications, phonographs, and computer industry; which are seen used in supply chain contract known as effective control mechanism under the varied supply chain situations, and (2) analyzing the effects that return rates and seller's contract parameters have in the outcome of the model and (3) observing how the effectiveness of Return Policy changes under such circumstances where the buyer's sales ability and the seller's risk inclination are taken into account. Thus, the main purpose of this research lies in analyzing what exactly are the effects (and or outcome) that sales ability and risk inclination have in Return Policy, and additionally by conducting comprehensive research on Supply Chain Policy Dynamics Simulation Model, we will try to prove that not only the Systems Dynamics approach is highly contributive in supply chain management but also that it will serve as a foothold in future research. As a result of the research, supply chain achievement level turned out to be high when Return Policy is enforced, and the achievement level was even higher when seller's sales ability was taken into consideration along with the Return Policy. On the other hand, the achievement level decreased when the seller had risk aversive tendencies.

• IE interfaces
• /
• v.21 no.4
• /
• pp.354-364
• /
• 2008

This paper deals with a design problem of simulation for MIME (multi indenture and multi echelon) with lateral transshipment. Especially, we consider lateral transshipments in case that (S-1, S) ordering policy is used in multi echelon repair system. Some rules for ordering spare parts in lateral transshipments between the lowest-level units are studied and are implemented by an activity diagram in object-oriented method. By numerical examples, we compare regular (S-1, S) ordering policy and (S-1, S) policy with lateral transshipment.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1999.05a
• /
• pp.145-153
• /
• 1999

We consider a Direct Input Output Manufacturing System(DIOMS) which has a munber of machine centers placed along a built-in Automated Storage/Retrieval System(AS/RS). The Storage/Retrieval (S/R) machine handles parts placed on pallets for the machine centers located at either one or both sides of the As/Rs. This report studies the operational aspect of DIOMS and determines the optimal operating policy by combining computer simulation and genetic algorithm. The operational problem includes: input sequencing control, dispatching rule of the S/R machine, machine center-based part type selection rule, and storage assignment policy. For each operating policy, several different policies are considered based on the known research results. In this report, using the computer simulation and genetic algorithm we suggest a method which gives the optimal configuration of operating policies within reasonable computation time.

• Korean System Dynamics Review
• /
• v.5 no.1
• /
• pp.35-47
• /
• 2004

Analyzers adopt a variety of techniques in analyzing public policy. Based on the results of exploration, they explain problems and suggest prescriptions such as policy alternatives. By doing so, however, they have different degree of confidence in their own results according to what techniques they adopt. This study explores whether system dynamics apporach gives analyzers more confidence on their own simulation results than other approaches do. This study designed an experiment. Three research teams are organized. Each of them applies different analytic techniques on the same policy issue, i.e., introduction of transportation tax. It flnds that researchers using S,D. simulation technique are most confident on what they predict, while researchers adopting survey techniques show moderate degree of confidence on their results. Researchers using regression are least confident.

• Proceedings of the Korea Society for Simulation Conference
• /
• 1998.10a
• /
• pp.36-40
• /
• 1998

Tool shop of the‘D’Motor Co.(DMC) fabricates dies for producing automotive press panels. In order to increase the production capacity of the tool shop, DMC established an FMS that consists of high-speed machines and a CMM at the tool shop in 1997. Due to the difference in machine capability among the existing machines and the new FMS it is needed to find an optimal way of allocating workloads among machines to maximize the production. In a way to solve this, we model the die fabrication process of the tool shop and analyze its performance by computer simulation. In this study, we at first identify the bottleneck processes of the die fabrication process under the current operation policy. Then, we derive some alternative operating policies applicable to the tool shop, and analyze the optimal operation policy by comparing the performance of the tool shop following each alternative policy.

• Korea Journal of Hospital Management
• /
• v.16 no.4
• /
• pp.161-182
• /
• 2011

In today's competitive dental markets, it is of paramount importance to improve service quality and at the same time to use scarce resource efficiently. In this study, we present appointment policies and scheduling rules for private dental clinics to reduce the waiting time of patients and to increase the revenue by utilizing resource more effectively. This study validates the proposed appointment policies and scheduling rules based on simulation models. We show that the bottleneck-based appointment policy is the most effective among appointment policies, followed by the multiple-block appointment one. The shortest processing time among scheduling rules contributes most to the performance of the appointment system.

• Proceedings of the Korea Society for Simulation Conference
• /
• 1999.04a
• /
• pp.105-109
• /
• 1999

Tool shop of the D Heavy Industries Co.(DHI) fabricates engines for a bus, truck, small ship. In order to increase the production capacity of engines, DHI will be established the new tool shop that consists of a block line, head line, assembly line, test line and AS/RS in 1999. In order to assure the production capacity designed of the new tool shop for producing engines and improve the production process of it, it is needed to find a bottleneck process and an optimal way of allocating workloads among machines and workers to maximize the production. In a way to solve this, we model the engine fabrication process of the tool shop and analyze its performance by computer simulation. In this study, we at first identify the bottleneck processes of the engine fabrication process under the designed operation policy. Then, we derive some alternative operating policies applicable to the new tool shop of an engine, and analyze the optimal operation policy by comparing the performance of the tool shop following each alternative policy.