​​Linear-Quadratic Controls in Risk-Averse Decision Making cuts across control engineering (control feedback and decision optimization) and statistics (post-design performance analysis) with a common theme: reliability increase seen from the responsive angle of incorporating and engineering multi-level performance robustness beyond the long-run average performance into control feedback design and decision making and complex dynamic systems from the start. This monograph provides a complete description of statistical optimal control (also known as cost-cumulant control) theory. In control problems and topics, emphasis is primarily placed on major developments attained and explicit connections between mathematical statistics of performance appraisals and decision and control optimization. Chapter summaries shed light on the relevance of developed results, which makes this monograph suitable for graduate-level lectures in applied mathematics and electrical engineering with systems-theoretic concentration, elective study or a reference for interested readers, researchers, and graduate students who are interested in theoretical constructs and design principles for stochastic controlled systems.​

Optimization problems involving stochastic models occur in almost all areas of science and engineering, such as telecommunications, medicine, and finance. Their existence compels a need for rigorous ways of formulating, analyzing, and solving such problems. This book focuses on optimization problems involving uncertain parameters and covers the theoretical foundations and recent advances in areas where stochastic models are available. Readers will find coverage of the basic concepts of modeling these problems, including recourse actions and the nonanticipativity principle. The book also includes the theory of two-stage and multistage stochastic programming problems; the current state of the theory on chance (probabilistic) constraints, including the structure of the problems, optimality theory, and duality; and statistical inference in and risk-averse approaches to stochastic programming.

Providing readers with a detailed examination of resilient controls in risk-averse decision, this monograph is aimed toward researchers and graduate students in applied mathematics and electrical engineering with a systems-theoretic concentration. This work contains a timely and responsive evaluation of reforms on the use of asymmetry or skewness pertaining to the restrictive family of quadratic costs that have been appeared in various scholarly forums. Additionally, the book includes a discussion of the current and ongoing efforts in the usage of risk, dynamic game decision optimization and disturbance mitigation techniques with output feedback measurements tailored toward the worst-case scenarios. This work encompasses some of the current changes across uncertainty quantification, stochastic control communities, and the creative efforts that are being made to increase the understanding of resilient controls. Specific considerations are made in this book for the application of decision theory to resilient controls of the linear-quadratic class of stochastic dynamical systems. Each of these topics are examined explicitly in several chapters. This monograph also puts forward initiatives to reform both control decisions with risk consequences and correct-by-design paradigms for performance reliability associated with the class of stochastic linear dynamical systems with integral quadratic costs and subject to network delays, control and communication constraints.

Risk-averse stochastic programming provides means to incorporate a wide range of risk attitudes into decision making. Pioneered by the advances in financial optimization, several risk measures such as Value-at-Risk (VaR) and Conditional-Value-at-Risk (CVaR) are employed in risk-averse stochastic programming for a variety of application areas. In this work, we consider risk-averse modeling approaches for stochastic multicriteria and stochastic sequential decision-making problems. First, we propose a new multivariate definition for CVaR as a set of vectors. We analyze its properties and establish that the new definition remedies some potential drawbacks of the existing definitions for discrete random variables. Motivated by the computational challenges in the optimization of vector-valued multivariate definitions of CVaR, next, we study two-stage stochastic programming problems with multivariate risk constraints utilizing a scalarization scheme. We formulate this problem as a mixed-integer program (MIP) and devise two delayed cut generation algorithms. The effectiveness of the proposed modeling approach and solution methods are demonstrated on a pre-disaster relief network design problem. Finally, we study the Markov Decision Processes (MDPs) under cost and transition probability uncertainty with the objective of optimizing the VaR associated with the expected performance of an MDP model. Based on a sampling approach, we provide an MIP formulation and a branch-and-cut algorithm, and demonstrate our proposed methods on an inventory management problem for long-term humanitarian relief operations.

We discuss how Whittle's (Whittle, 1990) approach to risk-sensitive optimal control problems can be applied in economics and finance. We show how his analysis of the class of Linear Exponential Quadratic Gaussian problems can be extended to accommodate time-discounting, while preserving its simple and general recursive solutions. We apply Whittle's methodology investigating two specific problems in financial economics and monetary policy.

This book revises the well-known capacity control problem in revenue management from the perspective of a risk-averse decision-maker. Modelling an expected utility maximizing decision maker, the problem is formulated as a risk-sensitive Markov decision process. Special emphasis is put on the existence of structured optimal policies. Numerical examples illustrate the results.