Analysis of a maximum marginal return assignment algorithm

In the Strategic Defense Initiative, battle management and command, control, and communications, one of the problems that arises is the multiweapon, multitarget assignment problem. Weapon platforms are satellites in orbit containing kinetic kill vehicles (projectiles) which can be sent to targets (boosters/missiles). A given weapon platform's projectiles have a known probability of hitting any particular target. This assignment problem is formulated as a large-scale nonlinear integer programming problem. Efficient near-optimal algorithms are found. The most promising maximum marginal return algorithm is studied in depth, using analytic and simulation techniques.<>