Sustainability and
Project Controls Solutions


Using Options to Manage Dynamic Uncertainty in Acquisition Projects

Uncertainty in acquisition projects and environments can degrade performance. Traditional project planning and management tools and methods can effectively deal with uncertainties in relatively stable environments. But in more uncertain environments conditions can evolve beyond the assumptions used in pre-project planning and require major deviations from initial plans. Important uncertainties often cannot be identified and described adequately during pre-project planning to design optimal strategies. Therefore rigid project strategies prepared solely based on the most likely outcomes as perceived during pre-project planning can result in sub-optimal performance. In these cases acquisition planners must explicitly incorporate flexibility into project plans to keep effective strategies available until uncertainty resolves adequately to reveal the best choice. This paper makes the case that options provide a framework for designing, evaluating, and implementing flexible acquisition project strategies and therefore can improve project performance. A large complex defense project illustrates the potential and challenges of options and research needs to expand and improve their use to manage uncertainty.


Using Options to Manage Dynamic Uncertainty in Acquisition Projects 

David N. Ford and B. Kagan Ceylan 

(Original of this article was published in October 2002 Issue of the Journal of the Defense Acquisition University. CLICK HERE TO GO TO THIS PUBLISHED ARTICLE.)

Maximizing the value of acquisition projects in dynamic environments is difficult partially because project managers must manage a variety of environmental and internal uncertainties as well as more common project complexities. Miller and Lessard (2000) report that success for sixty large ($985 million average cost) engineering projects, including research and development projects, depended largely on how uncertainty was managed. Many large complex defense acquisition projects also include technology research and development in dynamic and unpredictable environments. These development efforts can pose significant risks for the entire project because their outcomes are often predecessors of major activities and failures or delays in these efforts propagate through the project. How can managers of large complex defense projects plan for critical uncertainties?

Development projects risk sub-optimal performance if uncertainty is not explicitly incorporated into project planning. Many acquisition strategies are based on a project's characteristics and environment during front end planning. If these characteristics and environments are relatively stable initial plans can absorb changes in the project or its environment, changes in acquisition strategies are not required, and traditional pre-project planning is adequate. However, when critical parameters are difficult or impossible to accurately predict, uncertainty must be managed strategically because changes that occur during project execution may render the best course of action, as determined during front end planning, sub-optimal (Gupta and Rosenhead, 1968). Ford, Lander, and Voyer (2002) refer to these uncertain project components and environmental impacts that only evolve adequately for strategy selection after pre-project planning as "dynamic uncertainties" and describe in more detail why they are difficult to manage.

Three characteristics of uncertainty in large complex defense projects make it particularly difficult to manage. First, one-of-a-kind research and development efforts provide few opportunities to develop routines that can be evaluated and thereby improved. Therefore, historical experience is not available to inform forecasts of uncertain features. In addition, an inadequate understanding of new technologies or their implementation is available during pre-project planning to make forecasts that are accurate enough for strategy selection. Second, long project durations (e.g. 10.7 years average in the previously sited Miller and Lessard study) allow environments to evolve far from pre-project conditions. These dynamic uncertainties can cause strategies that are optimal during pre-project planning to become obsolete in later stages of the project. Third, tight coupling among project components create complex dynamic systems. Understanding individual project components is inadequate for understanding the system (Sterman, 1994; Senge, 1990). This increases project uncertainty and the difficulties of forecasting and planning. How can defense acquisition project planners proactively prepare for dynamic uncertainty?

Designing acquisition strategies that can be used to successfully manage dynamic uncertainties is an important but difficult part of project planning. Uncertainties that cannot be identified or forecasted can only be managed reactively with adaptive systems and managers (DeMeyer, Loch, and Pich 2002). But many dynamic uncertainties can be characterized adequately to be managed proactively by anticipating multiple scenarios to project goals and using flexible strategies to dynamically choose the best strategy based on how uncertainty resolves. An option is a right without an obligation to take specific future actions depending on how uncertain conditions evolve (Amram and Kulatilaka, 1999). Options can provide a framework for using flexible strategies to describe, design, evaluate, and implement strategies directed at dynamic uncertainties. Both options theory and decision analysis provide formal mathematical methods for valuing options that have been designed. However, less research exists on option design, assessment, and implementation processes for practicing planners and managers. The research to date on option design has mainly focused on modularity in the field of product development (Baldwin and Clark, 2000). In product development, modularity creates options because one can adapt to changing conditions in the environment by mixing and matching modules without changing the entire product. Therefore, modularity provides significant competitive advantage for modular products, particularly in environments in which product life cycles are increasingly shorter. However, tools and methods to create and measure options in acquisition projects are yet to be seen. A lack of structured methods and tools that can guide project planners in building flexible project plans to manage dynamic complexity remains a barrier to improved acquisition project management. Here options are described and evaluated as a tool for managing dynamic uncertainty from this managerial perspective. Based on this evaluation we hypothesize that the lack of such an operational options process theory constrains the description, evaluation, and advancement of options to improve acquisition.

To specify, clarify, and support our hypothesis descriptions of strategic approaches to managing uncertainty are followed by a description of options from a managerial process approach. One use of options in a large complex defense project is described to identify differences between options practice and available theories, and to identify research needs to improve acquisition planning and management.

Traditional Planning Tools for Managing Uncertainty

Strategic management, pre-project planning, and risk management provide planning tools and methods that can be used to address uncertainty. The relevant portions of these theories and their use by defense agencies are briefly described and evaluated to establish the available models of options to manage dynamic project uncertainty in large complex defense projects. 

Strategic management explicitly addresses the management of uncertainty with flexible plans and strategic adaptation after initial strategy selection (Mintzberg, 1978; McGrath and MacMillan, 2000). Although strategic management focuses on ongoing enterprises, uncertainty must also be explicitly incorporated into project strategies to maximize performance (Miller and Lessard, 2000). Strategic planning integrates environmental opportunities and threats with internal strengths and weaknesses into potentially flexible strategic plans that are the basis for specific projects (Mintzberg, 1995). According to Mintzberg (1978), strategic planning traditionally depicts a highly ordered, neatly integrated process explicated on schedules. However, in dynamic environments strategic management should also take into account possible changes in the environment beyond the plans developed based on the initial environment. This requires strategic adaptation, which updates strategies continuously, thereby remaining flexible (Porter, 1980). Processes have been developed for ongoing enterprises at relatively aggregate levels, but these concepts have not been developed into implementable processes for managing projects. 

Rigid strategic planning methods, such as those using solely critical path method coupled with risk analysis, have proven inadequate for the high complexity and dynamics of large public acquisition projects (Hughes, 1998). As a result several federal agencies (e.g. DoD and NASA) abandoned rigid approaches as early as the 1960s in favor of more strategic and adaptive approaches (Sayles and Chandler, 1971; DoD, 2001; DoE, 2000; DoN, 2001). For example, Department of Defense Regulation 5000.2-R (DoD, 1996) requires a phased decision making process with exit criteria reviews at each phase and the parallel development of multiple concepts. These provide options to abandon portions of projects as a means of strategic adaptation if the objectives can no longer be justified in the light of unfolding events.

Pre-project planning includes a project strategy selection process (Mintzberg, 1978) that is widely used by industry (CII 1995, CH2MHill 1996) and defense agencies (DoD 1996; DoE 2000; DoN 2001). Pre-project planning compares alternative technologies, sites, etc. to identify the best feasible project strategy within project constraints. This method is effective in some contexts, but assumes that planners are fully informed about the project and that the project environment is relatively stable, or at least predictable (Mintzberg, 1978). Therefore, pre-project planning does not provide the flexibility required to successfully manage the dynamic complexities inherent in many defense projects.

Risk management tools and methods are also used widely used by industry (CII, 1989; Chapman and Ward, 1997; Koller, 1999) and defense agencies (DoD, 2001; DoE, 2000; DoN, 2001) to manage project uncertainties by identifying critical risks, marshalling resources to absorb the consequences of uncertainty resolution that threaten project performance, and developing contingency plans. These methods react to uncertainties after they resolve in undesired ways. In marshalling resources risk management aggregates risks to assess their impacts and thereby estimate slack resource requirements. While useful for analysis and modeling, aggregating risks contrasts sharply with the reductionist approach used by managers to isolate specific individual critical uncertainties for customized management. Although sometimes used with options, the identification of critical risks and contingency planning are not developed operationally in risk management to provide managers guidance concerning how to use flexibility to improve project performance.

As prescribed by defense agencies and industry, strategic management provides general guidelines for project flexibility. But the processes, methods, and tools for developing flexible strategic plans for projects and adapting to changes have not been operationalized adequately to manage dynamic project uncertainty. Pre-project planning and risk management also do not provide operational processes to proactively use flexibility to manage projects. An operational process for options use in projects would reflect managerial practice.

Options as Tools for Strategically Managing Dynamic Uncertainty

One means of achieving flexibility to address dynamic complexity is to delay committing to a strategy until uncertainty resolves, new information becomes available, and the better strategy is clearer (Gupta and Rosenhead, 1968). For example, Ward, Liker, Cristiano, and Sobek (1995) report how delaying the selection of automobile systems creates competitive advantage for Toyota in time-to-market and quality. Options structure managerial flexibility into delayed opportunities without obligations to change strategies to improve asset performance. Options add value by allowing managers to capture more benefits or shift risks depending on how one or more uncertain parameters behave. For example, a contract clause permitting the termination of the contract if a critical technology is not developed provides the government with an opportunity (but not an obligation) to terminate depending on how the technology feasibility uncertainty is resolved. Option taxonomies have been based on the nature of the managed asset, the objective of the management (risk mitigation or increased benefits), the timing of delayed strategy selection decisions and uncertainty evolution, and actions taken on strategies (e.g. abandon, expand, switch, etc.). Trigeorgis (2000) and others categorize and describe these classifications. Contracting guidelines for options have been codified in federal acquisition regulations (FAR subpart 17.2) for some types of options and many economic valuation models have been developed as a basis for strategy selection. A managerial process approach is adopted here as the basis for investigating options potential, challenges, and research needs. This approach focuses on managerial decision making, planning, design and management and contrasts to the existing approaches which focus on valuation (Trigeorgis, 1996; Amram and Kulatilaka, 1999) and strategic advantage (Courtney et al., 1997; Andersen, 1999) only. 

Despite the extensive use of options by acquisition project practitioners (Miller and Lessard 2000) few options processes are described in the literature, and most of those are normative. Based on fieldwork at the project described later, a typical process begins when a manager recognizes (perhaps through risk assessment) that the value of a managed asset may be significantly impacted by how an unpredictable parameter behaves in the future. For example a defense project manager may recognize that the costs and planned development of a specialized weapon (the asset) may depend on the design expertise of a critical vendor after the end of an existing contract (the uncertain parameter). Managers recognize the need for flexibility when a possible resolution of uncertainty (a scenario) using a basic strategy generates a performance scenario to be avoided (e.g. large costs) or captured (e.g. improved product performance). In the example the depletion of design expertise could increase future costs beyond budgets, constrain weapon development, or both. Alternative strategies that could improve performance under specific uncertainty resolution scenarios are then designed. Examples for the weapon system that could improve performance include contracting for the right to continue design after the current contract or guaranteeing employment of critical employees. Option designs include specific decision rules for implementation that describe the conditions that trigger a change in strategy. Options are implemented by monitoring the uncertain parameters, analyzing them as necessary to determine the status of the trigger, and changing strategies if indicated.

Options are valuable only when their benefits exceed their costs. Options can provide a variety of benefits, including improvements in economic performance, stronger strategic position, broader managerial perspectives, expanded planning processes, and increased productivity. Options also generate costs, most visibly in financial terms. Both initially purchasing the opportunity to choose strategies in the future and implementing strategy changes can incur costs. In the weapon example an extension clause might increase the contract price and include computer upgrades at government expense if the clause is invoked. The economic valuation of options as a basis for strategy selection has been a primary focus of options research (Howard 1976, Trigeorgis 1996). Valuation methods for real assets compare net asset values with a specific option strategy to asset values using a rigid strategy to estimate option values.

Many acquisition project managers recognize the value of flexibility in managing dynamic uncertainties and use options. However, the practice is rarely structured into the frameworks developed by options theoreticians. Theories of managing uncertainty and valuing options in particular do not reflect options practice by acquisition planners. This may be due to the complex, multi-dimensional nature of actual option settings, the difficulties of integrating widely varying data types for formal analysis, and the resulting informal and tacit processes used by practitioners. The process gap between options theory and options practice limits the description, evaluation, and improvement of options use practices.

Options in Procurement for the National Ignition Facility

The National Ignition Facility (NIF) is being developed by Lawrence Livermore National Laboratory (LLNL) under contract with the U.S. Department of Energy (DoE). DoE’s goal is to generate the thermonuclear conditions created in nuclear explosions in a laboratory setting. NIF will be the world’s largest experimental fusion facility, using 192 lasers to compress and heat a small capsule of material to fusion ignition. The project budget is $2.248 billion to be spent over approximately eleven years (Moses 2002). Project success depends on several large simultaneous research and development efforts to produce unique subsystems. 

Data was collected about the use of options to manage uncertainty at NIF by observing four public presentations on the project by DoE and LLNL management, reviewing project documents, interviewing the DoE project manager and LLNL project and procurement managers, and visiting the site twice, including tours of the facility while under construction. NIF managers were found to use options (although they do not typically use that term) to manage many of the large uncertainties inherent in the project. The LLNL project manager attributed the management team’s frequent use of flexibility (including options) to their focus on project objectives instead of specific solutions. This allows managers to identify multiple potential strategies and scenarios to success (Moses, 2001). These strategy: scenario sets were used to design options. Several principles for managing uncertainty guided procurement at NIF. Examples include having two or more vendors for all major components to reduce the risk of inflated prices by a sole supplier and LLNL avoiding a manufacturing role to reduce the risks due to uncertain project funding and schedules (Moses, 2002). LLNL preferred to contribute in its areas of strength (scientific expertise and funding) and focused vendor efforts on their strengths (technology development and manufacturing).

Laser Glass Procurement: An Example of Applying Options in Technology Development

The NIF laser glass production strategy illustrates the use of options to address a common but important acquisition project question: How many parallel development efforts should be supported? Many factors impacted laser glass production in addition to the options strategy. Some of these factors will be identified to illustrate the complexity of options use in practice. However we focus on the options aspects of laser glass procurement, as the details of other aspects or their impacts on laser glass production are beyond the scope of this paper.

NIF will spend more than $350 million to produce over 3,000 pieces of laser glass, weighing about 150 pound each4. Laser glass begins as slabs of very high quality glass called “blanks.” The large volume of blanks and project schedule and budget required a production rate thirty times larger and five times cheaper than was used on prototype lasers, requiring the development of a new glass production technology and manufacturing facilities. Glass vendors could not justify funding the development. Therefore NIF invested in glass production technology development (Campbell, 2001). The development of a high-volume continuous-melting glass production process included at least two critical uncertainties; whether the technology could make the glass and whether the quality of the glass would be acceptable. The threat posed by these uncertainties was that, if development efforts failed in either way the project could be delayed too far to meet its deadline and would incur very high unbudgeted costs. Although LLNL had established relationships with experienced laser glass vendors, none could guarantee successful development a priori. Therefore, it became clear during laser glass procurement planning in 1994 that alternatives to a one-vendor strategy should be considered.

LLNL considered two types of procurement strategy for glass production technology development. A base strategy would invest in a single production development effort, helping as possible and hoping for a successful development. An alternative strategy would simultaneously invest in two independent development efforts by two glass producers, increasing the likelihood that at least one effort would be successful. If only one effort was successful this strategy allowed LLNL to abandon the failed development effort, use the successful one, and avoid the consequences of having no successful glass production system at the end of the initial investment period5. The managed asset is the NIF project and the dynamic uncertainty is the likelihood of a vendor successfully developing a feasible glass production technology with the required quality. The cost of the basic strategy is the cost of investing in one vendor (approximately $12 million). Investing in multiple vendors would purchase NIF opportunities to proceed with successful vendors at two points in time, when each of the uncertainties were resolved. The option costs are the funds required to invest in a second vendor up to the uncertainty resolution times (approximately $12 million each). The strategy can be structured as two options to abandon an unsuccessful vendor when the technology feasibility and glass quality uncertainties resolve. Does the one-vendor or two-vendor strategy best serve NIF?

NIF managers considered the two-vendor strategy attractive for both economic and non-economic reasons (e.g. generating competition between vendors). Despite a plethora of factors that influenced strategy attractiveness, the analysis that valued the option and drove strategy selection centered on the following comparison of strategy: scenario sets. If a single vendor was selected the development might succeed. But if the single vendor failed the costs to the project in time, money, and political consequences would prevent the project from meeting its targets. In contrast, if two vendors were selected none, one, or two could succeed. The likelihood of two failures was considered low. Exactly one success would allow NIF to exercise its option by abandoning the unsuccessful vendor, and two successes would provide manufacturing and pricing flexibility in addition to NIF's minimum needs. The avoided costs of project failure if investments were made in two vendors were (informally) estimated to greatly exceed the additional cost of investing in a second vendor (0.5% of the project budget), even if the avoided costs were discounted at any reasonable rate to account for the time value of money and other benefits. Therefore the option was considered more valuable than its cost. Based on this reasoning, in 1994 DoE and LLNL selected a two-vendor strategy and in 1995 contracted with two vendors to support parallel development efforts without further commitments by LLNL or DoE. This strategy purchased two options to abandon a failed development (if one occurred) and continue to support a successful development (if one occurred) when the technology feasibility and quality uncertainties resolved. 

The uncertainty about technology viability was resolved in early 1999 when both vendors successfully produced pilot runs of glass using continuous-melting processes. Due largely to the remaining quality uncertainty NIF chose to not abandon either vendor. Quality uncertainty was resolved near the end of 2000 when both vendors also demonstrated the ability to generate the required glass quality. NIF chose again to continue with both vendors. Because both vendors succeeded NIF purchased valuable production and pricing flexibility with which it can manage other project uncertainties (e.g. schedule). The costs avoided remain significant, albeit less than those saved in case of a development failure. The NIF laser glass production option illustrates how options have been used to increase project value and the difficulty of rigorously addressing relatively simple but important procurement questions in practice. 


An acquisition strategist might ask several questions concerning the NIF laser glass procurement strategy. What is the optimal number of vendors for NIF to invest in? Precluding clairvoyant planning that could have perfectly predicted the success of a single vendor, it appears that the NIF management chose the right strategy. But if, in the extreme, the likelihood that a single vendor would succeed was believed to be 99% and the added cost of a second vendor was very high, perhaps a single-vendor strategy would have been preferred and the option should not have been purchased. On the other hand, what if both vendors had failed? Perhaps NIF should have invested in more than two vendors. Would strategies in addition to those considered have added even more value to the project? How do acquisition planners know whether all potentially valuable strategies have been identified? How do planners design strategies? More generally, how do project and option structures and development processes impact project value and strategy selection?

Answers to these questions are not obvious or easily obtained. Strategy analysis depends largely on the probabilities of success, costs, and their analysis. Researchers have proposed methods of economically valuing staged parallel development strategies that may be applicable to the multiple-vendor, staged development problem described here (Ding and Eliashberg, 2002). But the simplifying assumptions required for mathematical tractability prevent these models from realistically modeling the complexities inherent in managing uncertainty in large development projects. A theory of options practice is needed to relate current options theories and practice. 

NIF’s use of options can be assessed with existing uncertainty management theories. Laser glass procurement for NIF used some of the methods and tools prescribed by strategic management. NIF managers surveyed the project environment to identify uncertainties that threatened project success (e.g. undeveloped glass production technology). Managers also assessed NIF’s internal strengths (e.g. science) and weaknesses (e.g. manufacturing) as a part of strategy development. However there was no evidence found that the team identified all potentially valuable strategies or was able to test how well the selected strategies addressed the uncertainties compared to alternate strategies. Laser glass procurement in the NIF project focused on flexibility (ala DoD guidelines), using a reductionist approach to isolate and manage key individual uncertainties. The observed use of options on the NIF project does not appear to closely match options theory. This supports our hypothesis that a structured process for the design, assessment, and use of options that resembles practice is needed to improve options use.

This paper has described dynamic uncertainties as a particularly difficult challenge in strategically planning large complex defense projects. Traditional methods and tools for managing uncertainty were reviewed and found unable to adequately address dynamic uncertainties. Options are described as a framework for structuring the management problem, potential strategic responses, and valuation for strategy selection to proactively manage dynamic uncertainty. An example of the use of options for procurement at a large defense project illustrates the potential benefits and challenges of using options in practice and deficiencies in options theories for application. The lack of structured processes and tools for designing, valuing, and implementing options by practitioners limits their assessment and improvement. These tools and processes would also integrate flexible strategies with existing project planning and management tools and thereby expand the strategic project planning domain. Applying such tools and processes would increase the value of large complex defense projects.

Acknowledgements: The authors gratefully acknowledge the financial support for this research provided by the Defense Acquisition University under contract N00244-01-C0037. The authors also thank Jim Anderson, DoE project manager of NIF, and LLNL managers of NIF who contributed information about the project and their time.



Amram, M. & Kulatilaka, N. (1999). Real options. Harvard Business School Press

Andersen, T.J. (1999). Strategizing in the new competitive energy market. The Electricity Journal. August/September, 42-48

Baldwin, C.Y. and Clark, K.B. (2000). Design Rules, The Power of Modularity. MIT Press, Cambridge, Massachusetts

Campbell, J. 2001. Interview with author. December 13, 2001. Lawrence Livermore National Laboratory. Livermore, Ca.

Chapman, C. & Ward, S. (1997). Project Risk Management. John Wiley & Sons. New York.

Construction Industry Institute (1995). Pre-Project Planning Handbook, Austin, Tx.

Courtney, H., Kirkland, J. and Viguerie, P. (1997). Strategy under uncertainty. Harvard Business Review. November-December, 67-79

DeMeyer, A., Loch, C.H. & Pich, M.T. (2002). A framework for project management under uncertainty. Sloan Management Review

Department of Defense (2001). Risk Management Guide for DoD Acquisition. Defense Acquisition University. Defense Systems Management College. Fort Belvoir, Va.

Department of Navy. (2001) .Acquistion Strategy Decision Guide. Accessed May 15, 2001.

Department of the Navy. (1998) Top Eleven Ways to Manage Technical Risk. PAVSO P-3686. Office of the Assistant Secretary of the Navy (RD&A).

Ding and Eliashberg (2002) Structuring the New Product Development Process. Management Science. 48(3): 343 - 363.

Ford, D., Lander, D. and Voyer, J. (2002) "The Real Options Approach to Valuing Strategic Flexibility in Uncertain Construction Projects" Construction Management and Economics. 20(4): 343-352.

Gupta, S.K. and Rosenhead, J. (1968). Robustness in sequential investment decisions. Management Science, 15(2), B-18-B29

Howard, R.A. (1976). The Science of Decision-Making, Readings in Decision Analysis, Stanford Research Institute, 1976, 149-164

Miller, R. & Lessard, D. (2000). The strategic management of large engineering projects. The MIT Press. Cambridge, Ma.

Mintzberg, H. (1978). Patterns in Strategy Formation. Management Science. 24(9), 934-948

Mintzberg, H. (1995). The strategy process. Prentice Hall Europe. London.

Moses, E. (2001). Interview with author. December 13, 2001. Lawrence Livermore National Laboratory. Livermore, Ca.

Moses, E. (2002) "NIF Overview/Status" presentation to National Research Council Committee on Oversight and Assessment of Department of Engergy Project Management. February 6, 2002.

Porter, M.E. (1980). Competitive strategy. The Free Press

Sayles, L.R. and M.K. Chandler. (1971). Managing Large Systems. Harper & Row, Publishers, New York

Senge, Peter M. (1990) The Fifth Discipline, The Art and Practice of the Learning Organization. Doubleday, New York:

Sterman, John D. (1994) "Learning in and about Complex Systems," System Dynamics Review, 10(2-3): 291-330

Trigeorgis, L. (1996). Real Options, Managerial Flexibility and Strategy in Resource Allocation. The MIT Press. Cambridge, Ma.

Ward, A., Liker, J.K., Cristiano, J.J. and Sobek II, D.K. (1995). “The second Toyota paradox: How delaying decisions can make better cars faster”. Sloan Management Review, Spring Issue, 43-61

4 The parallel development of a French facility similar to NIF that would also need laser glass increased demand and schedule uncertainty.

5 Other alternatives such as adding investment or postponing the decision beyond the initial investment period provided additional options but are not considered here for clarity.

NIF: PMI Project of Year (2010)

  • All rights reserved © 2014 - SPC Solutions Sustainability and Project Controls Solutions
  • Designed and Executed by Aynur Ugurlu