Abstract-Uncertainty management is crucial for achieving high performance in enterprises that develop or operate complex engineering systems. This study focuses on flexibility as a means of managing uncertainties and builds upon real options analysis (ROA) that provides a foundation for quantifying the value of flexibility. ROA has found widespread applications ranging from strategic investments to product design. However, these applications are often isolated to specific domains. Furthermore, ROA is focused on valuation, rather than the identification of real options. In this paper, we introduce a framework for holistic consideration of real options in an enterprise context. First, to enable a holistic approach, we use a generalized enterprise architecture framework that considers eight views: strategy, policy, organization, process, product, service, knowledge, and information technology (IT). This expands upon the classical IT-centric view of enterprise architecture. Second, we characterize a real option as a mechanism and type. This characterization disambiguates among mechanisms that enable flexibility and types of flexibility to manage uncertainties. Third, we propose mapping of mechanisms and types to the enterprise architecture views. We leverage this mapping in an integrated real options framework and demonstrate its benefit over the traditional localized approach to ROA.

Abstract-Complex systems are subject to uncertainties that may lead to suboptimal performance or even catastrophic failure if unmanaged. Uncertainties may be managed through real options that provide a decision maker with the right, but not the obligation, to exercise actions in the future. While real options analysis has traditionally been used to quantify the value of such flexibility, this paper is motivated by the need for a structured approach to identify where real options are or can be embedded for uncertainty management. We introduce a logical model-based approach to identification of real option mechanisms and types, where the mechanism is the enabler of the option, while the type refers to the flexibility provided by the option. First, we extend the classical design structure matrix and the more general multipledomain matrix (MDM), commonly used in modeling and analyzing interdependencies in complex socio-technical systems, to the more expressive Logical-MDM that supports the representation of flexibility. Second, we show that, in addition to flexibility, two new properties, namely, optionability and realizability, are relevant to the identification of real options. We use the Logical-MDM to estimate flexibility, optionability, and realizability metrics. Finally, we introduce the Real Options Identification (ROI) method based on these metrics, where the identified options are valued using standard real options valuation methods to support decision making under uncertainty. The expressivity of the logic combined with the structure of the dependency model allows the effective representation and identification of mechanisms and types of real options across multiple domains and lifecycle phases of a system. We demonstrate this approach through a series of unmanned air vehicle scenarios.Index Terms-Complex systems, decision making under uncertainty, design structure matrix (DSM) model, flexibility, multiple-domain matrix (MDM), real options, unmanned air vehicles (UAVs).

Uncertainties can be managed through real options that provide a decision maker the right, but not the obligation, to exercise actions at a later time. In previous work [1] we introduced an integrated real options framework (IRF) that distinguishes among option mechanism and type. The mechanism is the enabler of the option, while the type refers to the flexibility provided by the option. The idea behind IRF is to use models of a system or enterprise as a coupled dependency structure matrix (C-DSM) in order to identify and value enablers and types of flexibility. In this paper, we first show how the distinction among mechanisms and types of options leads to the identification of some new "ilities", such as optionability, that are relevant to the options identification problem. Second, we show that the semantics of a traditional dependency model does not allow for the representation and estimation of flexibility and optionability. Therefore, we extend the C-DSM model to a logical C-DSM that is capable of representing logical relations among dependencies. Finally, we present metrics for estimating flexibility and optionability from the logical C-DSM model. We discuss the results of applying these metrics to identify mechanisms and options in purchasing a swarm of uninhabited air vehicles.