The impact of production mix variations and models varieties on the parts-feeding policy selection in a JIT assembly system

One of the significant challenges in operating a mix-model assembly system is the feeding of parts to the productive units. In order to avoid production loss, assembly systems require uninterrupted availability of components to feed workstations. On the other hand, the feeding of assembly components has to be performed in a way that minimises the related costs. In the past, the feeding system most widely used was so-called 'line storage' in which the components were stored along the assembly stations in large quantities and were periodically refilled by the central warehouse. Following just-in-time principles, nowadays, assembly system feeding is undertaken by supermarkets, as in decentralised storage areas close to the assembly lines. From such kinds of warehousing, a growing number of manufacturers are adopting two other feeding strategies: the kanban system, which continuously refills the assembly stations through the pull kanban system, or the kitting system, in which kits of components are prepared and delivered following the product through the assembly stations. This paper aims to quantitatively analyse and compare these two recent feeding strategies, considering the production mix variation and the assembled models variety influence. Moreover, kanban-kitting feeding policy and the related optimization issues are considered as hybrid. The findings from an industrial case study and a simulation analysis are also reported. Finally, a decision-making tool that defines a series of 'convenience areas' for the different feeding policies is provided.