Decision Support System Based on Grey Systems and Markov Chain MCGM (1,1) Applied to Improve Forecast for Demand Uncertainty

Decision Support System Based on Grey Systems and Markov Chain MCGM (1,1) Applied to Improve Forecast for Demand Uncertainty

연구 분야: Databases

학회: Mexican International Conference on Artificial Intelligence

Real-world information often comes with a degree of vagueness and imprecision, leading to potential unreliability. To address such uncertainties, various models have been developed, including grey systems, fuzzy logic systems, and rough sets. Each of these models offers different ways to handle the inherent uncertainty in data. This paper introduces a novel forecasting model that synergistically combines Markov Chains with the GM (1,1) model. The integration of these two methodologies aims to enhance forecasting accuracy beyond what either model can achieve individually. The proposed hybrid approach utilizes a Markov Chain to model the transition probabilities between states and the GM (1,1) model to forecast future demand based on accumulated data. By merging these two techniques, the model produces an interval forecast that is more robust and reliable compared to traditional forecasting methods like ARIMA. The interval forecast derived from this combined approach provides a range of potential future values, offering a clearer picture of the uncertainty and variability in demand predictions. The application of this new forecasting model was demonstrated in the context of warehousing services (3PL). Comparative analysis with previous forecasting methods has shown that the hybrid model outperforms its predecessors. The results confirm the effectiveness of the new scheme, highlighting its superiority in providing accurate and reliable forecasts. This paper contributes to the field by presenting a practical solution for handling uncertainty in demand forecasting and offering a validated method that surpasses traditional approaches.