Forecasting and Raw Material Planning in Traditional Songkok Production Using ARIMA and Simple Exponential Smoothing

Abstract

This study investigates the applicability of time series forecasting models—Autoregressive Integrated Moving Average (ARIMA) and Simple Exponential Smoothing (SES)—for optimizing raw material planning in traditional songkok production. Utilizing monthly production data from a small-scale manufacturer in East Java, Indonesia (July 2020–August 2024), the ARIMA(1,1,1) model demonstrated superior forecasting performance, particularly under weak and irregular seasonality. Compared to SES, ARIMA yielded lower MAE, MSE, and MAPE values, enabling more precise production planning. The forecasts were translated into raw material requirements, resulting in improved inventory precision and operational efficiency, with monthly material usage gains ranging from 2.05% to 2.18%. These improvements are especially critical for micro-enterprises constrained by limited resources and seasonally driven demand cycles. While the univariate approach is a limitation, the findings provide a foundation for integrating contextual data in future multivariate models. The study offers practical insights for digital transformation in artisanal sectors and contributes to the broader discourse on data-driven production planning in culturally embedded industries.