In many companies, recurring problems arise due to incomplete or inaccurate master data, both from customers and within internal data collection processes. Such deficiencies reduce the reliability and efficiency of subsequent operations. A common example is the entry of incorrect information into the system, which often remains unverified or uncorrected. Over time, these errors can negatively affect various business processes, as inaccurate master data form the basis for many decisions and workflows.
The goal of this project is to enhance the quality of operational processes through the early, automated detection and correction of data issues using AI-based video analysis. This will be conducted within a feasibility study focusing on the separation of goods in the inbound area. For instance, warnings could be issued in cases of counting errors, image evidence could be generated for quantity deviations, and reference images could be stored for subsequent processes such as inventory or outbound control. By introducing AI-supported correction and verification mechanisms, data quality can be significantly improved, thereby increasing the efficiency and accuracy of logistics operations. The data for this study are provided by a logistics company from practical operations.