This dissertation research addresses the challenge of scaling up algorithms for sequential decision making under uncertainty. In my dissertation, I developed new approximation strategies for planning and learning in the presence of uncertainty while maintaining useful theoretical properties that allow larger problems to be tackled than is practical with exact methods. In particular, my research...

In this study, from the user’s perspective, a spare ordering policy for replacement of a non-repairable product/one-unit system, which with finite useful lifetime limit under a rebate warranty, is developed. The spare unit for replacement is available only by an order and the lead-time for delivering that spare is random. By introducing the gain due to rebate warranty, and the costs due to orde...

We propose a method of approximate dynamic programming for Markov decision processes (MDPs) using algebraic decision diagrams (ADDs). We produce near-optimal value functions and policies with much lower time and space requirements than exact dynamic programming. Our method reduces the sizes of the intermediate value functions generated during value iteration by replacing the values at the termi...

It is shown how to model a cooperative path planning system for multiple autonomous air vehicles within the framework of a stochastic (dynamic programming) decision process. The proposed approach allows the vehicles to cooperate and find near-optimal search paths over a given environment in the presence of uncertainty and constraints on movement and computational power.

In this work, we provide a more consistent alternative for performing value of information (VOI) analyses to address sequential decision problems in reservoir management and generate insights on the process decision-making. These are often solved modeled as stochastic dynamic programs, but once state space becomes large complex, traditional techniques, such policy iteration backward induction, ...

Within a decision-making hierarchy of production planning process, the majority of problems are dependent on the time component and strongly sensitive to endogenous and exogenous components. These problems can be related to an important class of stochastic optimal control. Difficulties in providing a closed-loop policy for them, lead to search for sub-optimal alternatives. In this paper, four d...

In this paper we develop optimal dispatching controllers for elevator systems during uppeak traffic. An uppeak traffic period arises when the bulk of the passenger traffic is moving from the first floor up into the building (e.g., the start of a business day in an office building). The cars deliver the passengers and then return empty to the first floor to pick up more passengers. We show that ...

limitation of water resources in the earth has turned it as a valuable economic commodity. accordingly, the agricultural sector of which according to studies has the largest wasted water should be managed. therefore, it is advisable to use the application of methods and programming models such as the decision support system along with the abilities of gis in order to management of the water res...

SCALABLE APPROXIMATE DYNAMIC PROGRAMMING MODELS WITH APPLICATIONS IN AIR TRANSPORTATION Poornima Balakrishna, PhD George Mason University, 2009 Dissertation Co-Director: Dr. Rajesh Ganesan Dissertation Co-Director: Dr. Lance Sherry The research objective of the dissertation is to develop methods to address the curse of dimensionality in the field of approximate dynamic programming, to enhance t...