A simulation-based risk analysis technique to determine critical assets in a logistics plan

Identifying risks in military logistics planning is essential. Particularly, identifying risks that emanate due to unforeseen unavailability of key transport assets during a defence mission is vital to evaluate the robustness of contingencies built in to support the mission's logistics plan. An unforeseen unavailability of a transport may be caused by unscheduled maintenance or its assignment to another concurrent mission. Replacing unavailable specialized transport assets such as strategic lift ships and aircraft during an on-going mission generally mandates significant re-planning effort and time. Therefore, it is essential to examine and revise contingency arrangements before a logistics plan is put into action. One method to achieve this is by simulating the logistics plan with asset unavailabilities and observing the corresponding consequences. The consequence of an asset's unavailability as determined using simulation together with the likelihood of the unavailability occurring are then used to derive the risk the asset poses to a logistics plan. This paper describes initial research conducted to develop a risk analysis technique that determines the risks posed by transport assets in a logistics plan. We present a metric called failure threshold that serves as a measure of the consequence an asset's unavailability has on a plan. The failure threshold of an asset is a point in time within a plan such that unavailability of the asset before the threshold will cause the plan to fail. We also present a binary-search based algorithm that determines the failure threshold of an asset by simulating the effects for such unavailability at different points of time within the plan. The consequence, as determined by the failure threshold of an asset's unavailability is combined with a user-specified likelihood to derive the risk posed by the asset. The risk analysis technique presented in the paper is built as an extension to the multi-agent logistics distribution and movements planning system presented in (Marsh, et al. 2010). The risk analysis technique presented in the paper determines critical assets in a plan, where unavailability of critical transports may cause the plan to fail. The identification of critical assets allows planners to effectively incorporate suitable contingencies such as improving maintenance of critical assets or sourcing additional assets before the plan is put into action. In order to effectively plan contingencies, it is essential to have the ability to measure the effect the contingency arrangement has on a plan. To this end, we show that the failure threshold of an asset can be used to determine whether sourcing an additional asset with similar capability reduces the risks in the plan.