Considering the business system's complexity with a network approach

Business management involves collecting information, goods, and funds as they move from supplier to manufacturer to wholesaler to retailer to consumer. Such business comprises interconnected parts that can be fundamentally complex and dynamic. A disturbance in one subnet of the system may thus have an opposed impact on another subnets, thus disturbing the business. Disruptions can have expensive and extensive results. This research aims to improve an increased Bayesian network method to consider business disruptions. The goal is to develop strategies that can diminish the opposed impacts of the disruptions and improve overall system reliability. Two influence agents are specified: the Bayesian and junction lack influence agents. An industrial model is used to demonstrate the proposed application, making the business more reliable. Moreover, two network learning methodologies are reviewed to update the probabilities in a model. The neural network seems to be a more favorable updating tool.