Multiobjective missile rescheduling problem


Tezin Türü: Doktora

Tezin Yürütüldüğü Kurum: Orta Doğu Teknik Üniversitesi, Mühendislik Fakültesi, Endüstri Mühendisliği Bölümü, Türkiye

Tezin Onay Tarihi: 2017

Öğrenci: AHMET SİLAV

Danışman: ESRA KARASAKAL

Özet:

In this thesis, we address dynamic missile allocation problem for a naval task group (TG). We consider rescheduling of surface to air missiles (SAMs) in response to disturbances during the engagement process where a set of SAMs have already been scheduled to a set of attacking anti-ship missiles (ASMs). To produce an updated schedule, we propose mathematical models that consider efficiency of air defense and stability of the schedule. In the first part of thesis, we present a new biobjective mathematical model that maximizes the probability of no-leaker and minimizes total deviation from the existing schedule. We analyze the computational complexity of the problem and develop exact and heuristic solution procedures. In the second part of thesis, we develop a semi-autonomous decision making framework to update the engagement allocation plan due to rapid decision making requirement of a dynamic air defense scenario. The approach is a based on an artificial neural network (ANN) method that includes an adaptive learning algorithm to structure prior articulated preferences of decision maker (DM). Assuming that the DM's preferences are consistent with a quasi-concave utility function, ANN chooses one of the non-dominated solutions in each rescheduling time point and updates the existing schedule. In the third part of thesis, we consider a different stability criterion that minimizes total number of tracking changeover for SAM systems. We formulate the biobjective model and generate solutions by new exact and heuristic methods.