Framework for Evaluating Cognitive Support in Mixed Model Assembly Systems

Abstract

The assembly process is often the final stage of the production process, which implies that the products have a lot of accumulated value hence making errors is expensive at this point of the product life-cycle. Empirical studies show that 90% of the final assembly tasks are still performed by humans and in most cases solely relying on own experience. Because of the product complexity and the operational complexity, caused by the mixed model structure, the mental workload imposed on the operators is typically high. Consequently, the probability of errors is higher and delays may occur. Cognitive automation ensures that the mental workload imposed on the operator will be reduced.This paper will focus on a method for evaluating the performance of a manual mixed model assembly system equipped with a cognitive information system. For this purpose a framework is designed. The framework is categorized in three inner sections: product, workstation and operator data. These sections will define the mixed model assembly system and create a performance score. Further, a technical description of a cognitive information system is given. The cognitive information system will support the operator, both intrinsic and extrinsic, by offering him process related information. The cognitive information systems have three important aspects: what, how and when. Also, configuration opportunities are added to the information system in which the operator is able to configure his own system.