Analyzing losses from hazard exposure: a conservative probabilistic estimate using supply chain risk simulation

We present a supply chain risk analysis that is based on a Monte Carlo simulation of a generalized semi-Markov process (G.S.M.P.) model. Specifically, we seek to estimate the probability distribution of supply chain losses caused by disruptions. This distribution is computed conditional on conservative hypotheses which are the following: (1) no additional risk reduction measures are implemented beyond those already in place, (2) all the products whose production has been canceled are counted as losses at their market value. The simulation thus yields conditional probabilities of loss levels that firms may reasonably use in the evaluation of business interruption costs and insurance coverage limits. The model also enables the comparison of supply chain designs based on their resilience in recovering from risk events. The approach is novel for it connects stochastic modeling of risks from an insurance perspective with supply chain network design.