Optimizing an FMCG Supply Chain Using Discrete Event Simulation : A Proactive Approach for Managing Demand Volatility and Production Output Fluctuations to Improve Service Levels

Sammanfattning

The fast-moving consumer goods (FMCG) industry, particularly in the food sector faces persistent supply chain challenges. Some of these include high demand volatility, short product shelf life, production constraints, and service level expectations. These challenges make it difficult to maintain optimal inventory levels while achieving high service levels and minimizing waste. Traditional planning approaches that rely only on historical data and static models often fail to capture real-time variability and disruptions. At Company A, key performance indicators (KPIs) such as delivery accuracy (DA) and waste are used to evaluate performance. These KPIs are often used retrospectively and do not enable proactive decisionmaking. This thesis proposes a proactive approach using discrete event simulation (DES) to model Company A’s outbound supply chain and determine optimal safety stock (SS) levels per SKU across multiple terminals. The DES model developed in ExtendSim, simulates key supply chain processes including production scheduling, transportation, inter-terminal stock transfers, customer order fulfilment, and scrapping. Real historical operational data was used for modelling and validation. The DES model developed demonstrated a high degree of accuracy and reliability when compared to historical performance data. To identify optimal SS levels, a multi-objective optimization based on an evolutionary algorithm was integrated into the simulation. The optimizer was designed to identify inventory strategies to maximize DA while minimizing holding costs. The results indicated that optimized SS settings significantly improved DA and reduced excess inventory, even under fluctuating demand and variable production output. Scenario analysis was performed using ExtendSim’s scenario management block to evaluate the impact of forecast accuracy and SS levels on DA through a full-factorial design of experiment. Additionally, a wide range of “what-if” scenarios were simulated to evaluate the effect of changes in production frequency, transport schedules, minimum order quantity, production disruptions, and unexpected demand peaks. These scenarios enabled supply planners to understand how various operational changes influences DA, thereby improving resilience and contingency planning. The outcomes of this study include optimized inventory strategies, improved DA, and enhanced supply chain resilience. The DES model offers Company A with a robust decisionsupport tool to evaluate strategies, test operational changes, and respond proactively to supply chain variability. It facilitates the much needed shift from reactive to proactive planning, supporting continuous improvement and more resilient operations in a complex FMCG environment.