Optimitzation of aggregate planning and inventory in the sunflower supply chain based on situational analysis using soft system dynamics methodology

Abstract

Purpose: This study seeks to improve the international competitiveness of the Indonesian sunflower industry. The study includes several particular objectives to accomplish this goal: (1) To assess the present condition of the sunflower industry supply chain, and (2) To formulate an optimization model for production and inventory management.

Design/methodology/approach: The Soft System Dynamics Methodology (SSDM) was used for situational analysis, followed by optimization of production planning through demand forecasting using the Artificial Neural Network method and aggregate planning using the Heuristic method. Furthermore, inventory optimization was carried out using the working capital and storage space restriction model with three algorithms: Genetic Algorithm, Particle Swarm Optimization, and Simulated Annealing.

Findings: Based on the inventory optimization using the three algorithms, it was found that the Genetic Algorithm (GA) resulted in the lowest total inventory cost calculation, amounting to IDR2,943,675.

Research limitations/implications: This study investigates a sunflower industry located in Bandung Regency, Indonesia. The optimization models applied—namely Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Simulated Annealing (SA)—were specifically adapted to the available operational data. As a result, the findings may not be directly generalizable to other agro-industries operating under different conditions. Furthermore, limitations in resources, such as machinery and storage capacity, posed constraints on the system simulation for inventory planning.

Practical implications: This research offers practical methods and approaches for sunflower agro-industries to enhance operational efficiency, reduce production costs, and optimize storage space utilization. Recommendations, such as the moderate production scenario, provide actionable insights for fulfilling consumer demand optimally while avoiding lost sales.

Social implications: The study has potential socioeconomic benefits, particularly for farmers and local workers involved in the industry. Improved planning systems can contribute to economic stability for these communities while also ensuring high-quality sunflower-derived products become more widely available and affordable to the public.

Originality/value: The integration of diverse methodologies, including the Soft System Dynamics Methodology (SSDM), demand forecasting through Artificial Neural Networks (ANN), aggregate planning via heuristic methods, and inventory optimization using GA, PSO, and SA, distinguishes this study as a unique, comprehensive approach. This holistic framework adds significant value to the academic field of production and inventory planning, especially for small-scale industries like sunflower agro-industries.