<![CDATA[Any problems related to bad credits or problem loans in Indonesia are not constant, there can be any decrease or increase in each month. So, it can cause on uncertain provision of fund budget for underwriting payment of credit claims by credit underwriting institutions. Therefore, it is necessary for a system that can predict on value of underwriting payment on bad credit claims as a consideration to determine nominal value to be provided in the following months by the credit underwriting institutions. In this research, the prediction is conducted using Fuzzy Time Series method, because the data used are prepared in a consecutive time from month to month. To create better prediction, it is optimized using Particle Swarm Optimization (PSO) algorithm, because the PSO algorithm has high decentralization with simple implementation so that it can solve any optimization problems in an efficient manner. The error level is calculated using Root Mean Squared Error (RMSE). Based on the testing, the best solution has an average cost value by Rp. 159215 with its program operation time by 13,2 second. The solution is created with maximum iteration by 250, the population by 100, length of particle dimension by 250, value of cognitive coefficient variable (c1) is equal with 1 and the social coefficient variable (c2) is equal with 1.5, as well as inertia weight value (w) is equal with 0,6. So that it can be concluded that this research can be applied for prediction on value of underwriting payment on bad credit.]]>
Copyrights © 2018
