Analysis of Drug Forecasting with Single Moving Average and Single Exponential Smoothing Approach (Case Study in Jombang Regency 2017-2019)

Abstract

Drug forecasting is done by all health facilities including government's health facilities. It aims to consider the type of medication, the amount needed to realize a good drug management system. However, a of methods requires large selection using evaluation multiple measures of error in drug forecasting. The study aims to find out the method with the size of the least forecasting error between Single Moving Average and Single Exponential Smoothing. The data used is the annual primary data of the period 2017-2019. Samples amounted to 35 Public health centers. Forecasting calculation has been done using Single Moving Averages and Single Exponential Smoothing methods and testing errors using Mean Absolute Deviation and Mean Square Error methods. The study used 5 major medications with the most forecasting for 3 years. The drug with the most consecutive planning period 2017-2019 is paracetamol tablets 500 mg of 694,911 tablets, 126,713,379 tablets and 11,705,308 tablets. The conclusion of results showed a Single Moving Averages = 12,681 milions tablet with MAD = 0,594 milions and MSE = 666,841 millions. While, Single Exponential Smoothing = 7,949 milions with MAD = 4,557 milions and MSE = 372,884 millions. So, both of methos have small error measurement.