Recent publications underscore the critical implications of adapting to dynamic environments forenhancing the performance of material and information flows. This study presents a systematicreview of literature that explores the technological capabilities of smart production logistics (SPL)when applying machine learning (ML) to enhance logistics capabilities in dynamic environments.This study applies inductive theory building and extends existing knowledge about SPL in threeways. First, it describes the role of ML in advancing the logistics capabilities of SPL across variousdimensions, such as time, quality, sustainability, and cost. Second, this study demonstrates the applicationof the component technologies of ML (i.e. scanning, storing, interpreting, executing, andlearning) to attain superior performance in SPL. Third, it outlines how manufacturing companiescan cultivate the technological capabilities of SPL to effectively apply ML. In particular, the studyintroduces a comprehensive framework that establishes the technological foundations of SPL, thusfacilitating the successful integration of ML, and the improvement of logistics capabilities. Finally,the study outlines practical implications for managers and staff responsible for the planning andexecution of tasks, including the movement of materials and information in factories.