This open access book delves into methodologies for addressing prevalent statistical problems in supply chains--including prediction with multi-source data, density response prediction and ranking problems--leveraging model averaging techniques and numerical experiments. Its innovation lies in addressing uncertainty in model design and variable selection by integrating multiple viable candidate models instead of relying on a single one, while enhancing model performance through tailored model-averaging weight criteria. The intended readership of this book includes undergraduate students in universities, graduates, and academic researchers in the field of management science, data science and statistics. This book is suitable for those with a master's degree or above in management science or statistics.