Premium
7.3.2 Optimization of Research and Development Investment Strategies
Author(s) -
Gormley Kevin,
Fishenden James C.,
Scherer William T.
Publication year - 2004
Publication title -
incose international symposium
Language(s) - English
Resource type - Journals
ISSN - 2334-5837
DOI - 10.1002/j.2334-5837.2004.tb00587.x
Subject(s) - schedule , heuristics , operations research , interdependence , investment (military) , computer science , warrant , monte carlo method , expected return , risk analysis (engineering) , mathematical optimization , economics , business , engineering , finance , mathematics , portfolio , politics , political science , law , operating system , statistics
An enterprise must make up‐front investments in applied research and development (R&D) to mature technologies to a point where they can be successfully deployed. These decisions are based on expected return from the investment and the cost, schedule and technical risks facing each R&D project. Periodically, the decisions are revisited to see if changes in project performance and cost/schedule/risk estimates warrant a change in project funding. Common approaches to this problem involve either Monte Carlo simulation of outcomes for a decision policy, or optimization of decision variables based on some objective. This paper outlines an approach to R&D project selection where both simulation and optimization are used to determine an investment strategy, given project interdependencies and uncertainties in utility, cost, schedule and technical performance. The problem is formulated as a discrete sequential stochastic decision problem and uses heuristics to narrow the solution space and find a reasonable (though non‐optimal) solution.