Comparison of Forecasting the Number of Outpatients Visitors Based on Naïve Method and Exponential Smoothing

Senin, 26 Oktober 2020 - 03:23
Dokumen Lampiran
NoDeskripsiFilenameFilesize 
1Comparison of Forecasting the Number of Outpatients Visitors Based on Naïve Method and Exponential Smoothing45.-irfan-dwiguna-sumitra-s.kom-m.kom-p.h-d.pdf435 KB
Comparison of Forecasting the Number of Outpatients Visitors Based on Naïve Method and Exponential Smoothing
 
 
Ketua : IRFAN DWIGUNA SUMITRA S.Kom, M.Kom, P.h D
Anggota : I Basri K*
Departemen Pascasarjana, Universitas Komputer Indonesia, Indonesia
*E-mail: IlhamBasriK@gmail.com
 
Abstract. This paper aimed to predict outpatient visits from both general and BPJS category based on the Naïve and the Exponential Smoothing method, these two methods are compared to obtain the best method in predicting outpatient visits at XYZ hospitals. The data that were processed in this paper is based solely on the annual data collection over a period of 5 years from 2014-2018 which consists of 2 categories, namely general visit data and BPJS visit data. The annual general category data has the following data amounts: 2014 = 11028, 2015 = 12950, 2016 = 17587, 2017 = 21951, and 2018 = 19049. As for the BPJS visit data, the data as follows 2014 = 16869, 2015 = 14059, 2016 = 14217, 2017 = 13019, and 2018 = 9641. Results of predictions number of outpatient visits from Naïve method on the visit data which is categorized as general has MSE 5191788 and MAPE values of 16.335%, the categorized BPJS has a value of MSE 13165490 and MAPE 19.081%. While the Exponential Smoothing method on the visit data that is categorized as general has MSE 7790587 and MAPE values of 21.808%, BPJS category only having MSE 13165490 and MAPE values of 20.718%. Of the two methods, the method considered better is the Naïve method which has a percentage of MAPE values <20%. It can be concluded that the Naïve method is more suitable for forecasting the number of visits from the general category and BPJS compared to the Exponential Smoothing method because the forecast value and the std err are smaller.