A STRATEGIC, AREA-WIDE CONTINGENCY PLANNING-MODEL FOR OIL-SPILL CLEANUP OPERATIONS WITH APPLICATION DEMONSTRATED TO THE GALVESTON BAY AREA

This paper develops a general integer programming model for strategic, area-wide contingency planning of oil spill cleanup operations. Model inputs include the set of risk points and the likely spill scenarios and response requirements for each, the sites of existing storage loca tions and the inventory of components at each, and potential sites for new storage locations. The model prescribes a minimum total cost plan to either build new storage locations, expand existing ones, or both, to purchase new components and pre-position them, and a contingency p lan that determines which response systems should be composed to enabl e an effective time-phased response for each likely spill scenario. A family of heuristics based on linear programming (LP) is devised to re solve this strategic problem, providing an area-wide contingency plan. The heuristics are evaluated on a set of 10 test problems that involv e 1869 general integer variables and 3264 constraints. Computational t ests indicate that four heuristics are quite effective, prescribing so lutions for each of 10 test cases within 1.41% of optimum and within a few minutes runtime. This study focused on modeling the Galveston Bay Area, and the test problems represent application in that area. A sen sitivity analysis is demonstrated by assessing the impacts of componen t availability and the degradation of cleanup capability over time. Us e of the model as a decision support aid by responsible parties, contr actors, governmental organizations and others is described.