Comparative Analysis of Simple Moving Average and Single Exponential Smoothing Methods for Forecasting Readymix Concrete Demand
Year: 2026
Type: article
Abstract: General Background Ready-mix concrete represents a vital material in construction projects, yet its rapid setting characteristic prevents long-term inventory storage and creates operational vulnerabilities. Specific Background Demand fluctuations driven by unpredictable project schedules and external factors such as weather cause severe challenges in production planning for batching plants like PT X Sidoarjo. Knowledge Gap Although short-term time-series forecasting is widely recognized, comparative evaluations combining error metrics and tracking signal analysis for daily ready-mix concrete production remain limited. Aims This study evaluates and compares the Simple Moving Average and Single Exponential Smoothing methods to identify the optimal forecasting model with minimal error for daily ready-mix concrete demand. Results Quantitative analysis of historical sales data from April 2026 demonstrates that the Single Exponential Smoothing model with a smoothing constant of 0.7 outperforms the Simple Moving Average model, achieving lower error rates with Mean Absolute Deviation of 40.606, Mean Squared Error of 2373.946, Mean Absolute Percentage Error of 44.99%, and superior tracking signal stability within safety control limits. Novelty This research incorporates daily tracking signal evaluation into short-term ready-mix concrete forecasting, proving that exponential smoothing mitigates cumulative bias during severe demand shifts. Implications Applying the selected Single Exponential Smoothing model yields a daily demand projection of 33.771 m3 for May 2026, establishing an operational safety baseline to prevent overproduction losses and optimize raw material inventory management. Key Findings Highlights Single exponential smoothing with a smoothing factor of 0.7 achieves superior forecast accuracy compared to simple moving average. Tracking signal testing confirms that exponential smoothing maintains cumulative bias stability within acceptable control limits. The projected daily demand of 33.771 cubic meters serves as a safe operational baseline for batching plant production scheduling. Keywords Readymix Concrete, Demand Forecasting, Single Exponential Smoothing, Time Series Analysis, Tracking Signal
Authors: Adinda Dwi Cahayani, Inggit Marodiyah
Institution: Universitas Muhammadiyah Sidoarjo
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W7221811079
