volume 7 Turkeys issues December 2004
Fuzzy Modeling for Coordinating Logistics in Emergencies
Wei Yi & Linet Özdamar
Abstract
This paper describes a dynamic and fuzzy logistics coordination model used for conducting disaster response activities such as evacuation of affected people, transportation of wounded people to hospitals and of commodities from warehouses to aid distribution centers. Post disaster logistics is usually carried out in uncertain environments and information obtained from affected areas might be impeded by infrastructure damage and the loss of those on official duty. Furthermore, in many situations it is not possible to access affected districts and damage assessment is carried out from airborne vehicles on a vague scale.
Given the uncertainty in the number of people affected and wounded, and in the needs of people who have to stay in the region until they receive official help in finding shelter, the logistics problem is quite difficult to solve. In fact, vehicle routing and supplies coordination problems have their inherent difficulties even under certainty, because they are discrete problems classified as NP. In addition, discrete problems are known to be quite sensitive to changes in parameters.
We represent uncertainty by using fuzzy parameters related to demand, supply, injured people and hospital service rates. We then de-fuzzify these parameters in an efficient routing and transportation model. During the initial response periods, the model produces logistics plans based on fuzzy parameter intervals that are calculated by using regional disaster risk grades. The model is re-run in each planning period to handle new information that is communicated from affected areas. Parameter intervals are automatically re-adjusted according to new information and as the degree of uncertainty is reduced with time, parameters tend to have smaller intervals. We illustrate the implementation of the model on an earthquake scenario.
Key Words: logistics coordination in disaster response activities, dynamic routing and transportation, fuzzy modeling
volume 7 Turkeys issues December 2004