Functionalities and Implementation of Future Informational Assistance Systems for Manual Assembly

Abstract

The demand for customized products increases, leading to smaller product volumes and batch sizes, down to batch size one. The necessary flexibility and variety places high demands on assembly and increases the complexity. Therefore, the automation of manual assembly processes is often not cost-effective. To cope with these basic conditions, workers in the manual assembly should be supported cognitively by informational assistance systems. In addition to the typical product- and process-related aspects, adaptable human-centered functionalities must be considered, aiming to improve productivity, quality, workers’ health, and motivation. Thus, this paper examines the assistance functionalities that future assistance systems should provide for manual assembly processes and presents approaches for their implementation. Design Science Research is the framework for our research activities. The starting point is the analysis of existing assembly assistance systems and a determination of process optimization potentials. Through interviews with experts and the modeling of a manual assembly process, we determine the support dimensions and required functionalities for future assistance systems. Subsequently, the overall system architecture and the subsystems are designed and implemented. Intelligent image processing and deep learning algorithms are the basis for process progress recognition and analysis of the ergonomic situation. Gamification and augmented reality are further methods used. The processual changes resulting from the application of the presented novel assistance system are modeled in a case study, and the optimized aspects and implications for both workers and companies are discussed.