American Institute for Aeronautics and Astronautics 1 PROBABILISTIC MODELING OF TECHNOLOGY DEVELOPMENT FOR USE IN PROJECT PLANNING AND MANAGEMENT

M. Largent,D. Mavris

Abstract:† From the dawn of human flight, the use of new technology has been necessary for aircraft to fly faster, further, and higher. That need continues today, making technology development a critical factor in new aerospace systems development. This study introduces some of the detailed formulation behind a process developed to assist the planning and management of a development project for a new technology. The process is based on risk management and uses probabilistic analysis, project network techniques, and Bayesian statistics in order to provide information useful for decision making. The six steps in the process focus on identification and reduction of performance uncertainty, as well as identification of cost and schedule risk and uncertainty. The process introduces some ways in which analysis can interact with expert opinion in order to produce meaningful system-level comparisons of technologies. The iterative nature of the process helps to deal with uncertainty in the technology and outside influences which affect the technology’s development. INTRODUCTION/MOTIVATION The development of technology is an important part of fielding a new aerospace system, and is a large driver in whether the system meets its cost, schedule, and performance goals. A GAO study on best practices in management of technology development found that insufficient technology development for subsystems of the Comanche helicopter has contributed to the product development project being over budget and behind schedule. As such, an important part of the design and planning process is getting an accurate look at * Graduate Research Assistant, AIAA Student Member † Boeing Professor for Advanced Design, Director ASDL, AIAA Senior Member Copyright © 2002 by Largent and Mavris. Published by the American Institute of Aeronautics and Astronautics, Inc. with permission. what is required to mature a new technology and the future benefit gained from that technology. NASA defines technology development as being (italics added for emphasis), “a process of testing and analysis that progressively reduces the programmatic risk of selecting that technology for an application and increases the readiness of that technology for use in a mission.” Reduction of programmatic risk can be seen as a reduction of the risk of a technology being too expensive or not being ready in time to be used for an application. An increase of readiness means a reduction in the uncertainty associated with the technology’s performance and a shifting of the performance (as much as possible) towards desirable values. This reduction in performance uncertainty and shift of performance values is typically performed through planning and carrying out activities such as experiments, analysis, and prototyping. As the development project progresses, the activities get more detailed in focus, and the technology’s true performance is determined. These activities, then, are the focus of technology development. Since the activities that are planned deal with the uncertain performance of a technology, it follows that the programmatics of those activities also have an element of uncertainty. The time and money necessary to complete development activities cannot be known exactly, and the possible ranges of the schedule and budget values can actually be quite large. This introduces the risk of not meeting deadlines or budget limits, an unacceptable situation in today’s budget and schedule-conscious environment. Combining the importance of reduction in performance uncertainty with the resulting cost and schedule risk results in a situation where all three measures must be examined together. This study builds off of earlier work to introduce the details behind a process that could assist in planning and managing a technology development project in order to achieve beneficial levels of performance and performance uncertainty and keep project risks at an acceptable level.

Engineering