Title

Agent-based framework for the design, simulation, and assessment of agile organizations

Date of Completion

January 2008

Keywords

Engineering, Electronics and Electrical|Engineering, System Science|Operations Research

Degree

Ph.D.

Abstract

Agility is arguably one of the most important characteristics of successful organizations coping with the ever increasing demands of today's world. This thesis provides three essential contributions to the design, simulation, and assessment of agile organizations operating in dynamic and uncertain environments. Firstly, a new Markov Decision Process (MDP)-based method for managing goal attainment, the process of planning and controlling actions related to the achievement of a set of desired goals in the presence of resource and time constraints, is proposed. Specifically, the problem is addressed as one of optimally selecting a sequence of actions to transform the system and/or its environment from an initial state to a desired state. The study begins with a method to explicitly map an action-goal-attainment (AGA) graph to an MDP graph, and develop a dynamic programming (DP) recursion to solve the MDP problem. For larger problems having exponential complexity with respect to the number of goals, guided search algorithms: AO*, AO*C, and greedy search techniques, whose search power rests on the efficiency of their heuristic evaluation functions (HEF), are proposed. The contribution in this part stems from the introduction of a new problem-specific HEF to aid the search process. The study demonstrates reductions in the computational costs of the proposed techniques through performance comparison with standard DP techniques. The first study concludes with a method to address situations in which alternative strategies (e.g., second best) are required. The new extended AO* (EAO*) algorithm identifies alternative control sequences for attaining the organizational goals.^ Secondly, an agent-based approach for modeling the coordination and task execution in agile organizations is developed. A normative methodology and computational framework to assess the effectiveness and efficiency of command and control (C2) organizations is presented, where the process is based on quantitative representations of the organization and mission, and utilizes normative models of team and individual decision making. The assessment methodology has been applied to evaluate the benefits of the Sensing and Patrolling Enablers Yielding Effective Security (SPEYES) system—a ground-based decentralized C3I system comprised of emerging and existing sensing, SA/C2, and Shaping technologies. To facilitate the assessment analysis, the models are implemented using a computational agent framework for the Distributed Dynamic Decision making (DDD) virtual simulation platform.^ The third contribution is a distributed simulation paradigm and decision support system for exploring shared awareness and collaboration to gauge organizational performance and agility. The design and application of a decision support system (DSS) based on the third-generation distributed dynamic decision-making (DDD-III) simulator and contingency theory to increase the organizational cognitive capacity and to facilitate the processes of adaptation are discussed. The role of the DSS is to provide mission-monitoring and re-planning information to human decision-makers (DMs) in order to manage the explicit and tacit knowledge, codify them into simple and meaningful formats, and facilitate rapid transfer of knowledge among the DMs. An overview of contingency notions that incorporate three relevant components affecting organizational performance, viz., (1) environment, (2) organizational structure, and (3) strategy is presented along with a demonstration that triggers for organizational adaptation require an integrated multi-dimensional concept of congruence ("fit between the organization and mission environment") incorporating structure-environment, strategy-environment, and strategy-structure matches. Due to the elusive nature of knowledge necessary for efficient and effective organizational adaptation, the knowledge management strategies for efficiently codifying and rapidly transferring the knowledge among DMs is central to achieving superior organizational performance. The present DSS is a suitable test-bed to investigate these processes, and thus provides a means for organizations to gain competitive advantage. ^