Supply and Demand Analysis by Comparison of Forecasting Method in Manufacturing Motorcycles Tires

As one of the motorcycle tire manufacturing industries in Indonesia, it is faced with uncertainty of capacity and resources, due to the mismatch of forecasts causing an increase in inventory to 2,207,537 tires, the highest number in the last five years. The purpose of this paper is to research to analyze differences in sales forecast, demand, supply, and production from January 2015 to February 2020, then measure the error rate of demand data using POM for Windows with the Naive Method (NM), Moving Average (MA), Weighted Moving Average (WMA), Exponential Smoothing (ES), Exponential Smoothing with Trends (ESWT), Regression/Trend Analysis (R/TA), and Multiplication Decomposition (MD Seasonal). The lowest error measurement results using the Multiplicative Decomposition (MD Seasonal) method have a Mean Absolute Deviation (MAD) of 303,577 and a Mean Absolute Percentage Error (MAPE) of 14.15%. Using the Multiplicative Decomposition (MD Seasonal) method, demand forecast had been obtained as a reference for capacity planning such as machine resources and manpower planning, so that there were reduced production from 80,000 pcs/day to 60,000 pcs/day and makes stock inventory decrease to 757,997 pcs.