Tedarik zincirlerinde kamçı etkisinin kapsamlı empirik analizi

The Bullwhip Effect (BWE), where demand variability amplifies as orders move upstream, remains a key supply chain challenge. Despite extensive theoretical study, industry-level empirical analysis is limited. Many assumptions about BWE's impact on inventory, production, and labor remain untested across sectors. This dissertation fills this gap by analyzing industry data to identify BWE drivers, assess its effects during disruptions like COVID-19, and propose advanced forecasting techniques. The study first examines BWE's impact across industries using indicators like inventory, backorders, job losses, and capacity utilization. Findings show that industries experience BWE differently. While inventory smoothing is thought to reduce BWE, empirical evidence is inconsistent. Similarly, backordering does not always amplify fluctuations, whereas job loss rates correlate more strongly with BWE, underscoring the need for sector-specific strategies. Next, the study evaluates importation's role in BWE. While extended lead times are often blamed, results show that factors like information distortion and supply chain rigidity are more influential. Strengthening collaboration and real-time data sharing proves more effective in mitigating BWE than nearshoring or reshoring. Finally, the research explores advanced forecasting methods. Machine learning models such as Long Short-Term Memory (LSTM) networks and LightGBM significantly enhance forecast accuracy and supply chain stability. The findings highlight the crucial role of AI-driven forecasting and digitalization in managing modern supply chains.