Linking product and machine network structure using nested pattern analysis

Abstract The structure and variety of products that a company produces have a direct influence on the way a manufacturing system is designed. The implications of changes in the product variety or structure on the manufacturing system and resulting performance differences need to be considered by designers before introducing or changing products. In this paper, we suggest a measure derived from community ecology called “nestedness” to assess how changes in the product variety or structure can affect the operation-machine network of a manufacturing system in terms of its performance robustness. We define performance robustness as the manufacturing system's ability to keep a steady performance even in the face of disruptions such as product variety changes. We measure nestedness in an exemplary case study on a data set from a tool manufacturing job-shop and find the matrix of the network to be nested. The nested pattern means that there is a systematic relationship between operations and machines which results in performance robustness: if machines break-down, most other machines can be substituted. Similarly, if products are taken out from the portfolio, machines are still needed for the operations of other products. As such, our study is a first in examining the relationship between manufacturing system structure and performance robustness using interdisciplinary knowledge transfer with network science.