The impact of supply network characteristics on reliability

Purpose - As globalisation makes supply networks more complex, the risk of material disruptions increases. Many factors have been considered as affecting the reliability of supply networks. However, no empirical research has been carried out to assess and evaluate the impact of each of these factors on the reliability of supply networks. This paper aims to address this issue. Design/methodology/approach - A gap in the literature was identified around the evaluation of the impact of supply network design characteristics on reliability. This gap is addressed by performing a full factorial experimental design considering all the factors described in the literature, and then analysing (by using analysis of variance and linear regression models), thousands of theoretical and extreme structures of supply networks, thus allowing the analysis of the influence of each factor on the overall network resilience. Findings - Results show that network density, node criticality and complexity are significant factors in reducing the reliability of supply networks. In particular, node complexity (i.e. the total number of nodes in the network) was found to have the strongest negative effect on network reliability, while the strongest positive factor was sources criticality (i.e. the level of redundancy of suppliers). Practical implications - The identification of these factors and their relative impacts on network reliability can serve as a guide for the design of more reliable networks, and to know which are the most important to consider when designed distribution networks. Originality/value - The paper identifies, from the literature, key factors affecting supply network reliability and evaluates their relative impact. Given the number of factors identified, an extensive Monte Carlo simulation is used for the first time, by considering simple and very complex networks, to allow the testing of the role of each factor in supply network reliability.