A robust algorithm for project portfolio selection problem using real options valuation

Abstract

Purpose The principal concern of organization managers in the global rivalry of commerce environment is how to select the project portfolio among available projects. In this matter, organizations should consider the uncertainty intrinsic in the projects regarding an appropriate valuation technique within an optimization framework. In this research, the purpose of this paper is to formulate using a robust optimization algorithm to deal with the complexities and uncertainty inherent in the construction of the project portfolio. Design/methodology/approach First, a general mathematical formulation is presented, which in compound real options valuation is highlighted. This formulation gives managerial flexibility by correcting the deficiency of traditional discounted cash flow technique that excludes any form of flexibility. Then, considering a limitation on budget of the organization, an integer programming formulation to maximize the n-fold compound options for project portfolio selection is proposed. Finally, a robust optimization model is developed along with the robust combinatorial optimization algorithm, which is effective for solving problems under uncertainty. Findings Sensitivity analysis showed that projects in later phases of development, having survived several phases of pre-clinical and clinical tests, are worth more because they are more likely to pertain to business. However, the investment costs related to each project during development phases limit the number of projects that a company can bring to their final portfolio. Additionally, the analysis of conservatism level represented how project managers can quite easily determine their risk attitude and the corresponding portfolio composition. From a managerial point of view, the proposed framework is very useful because it requires only financial estimates. Hence, the proposed decision support tool can assist research and development (R&D) project managers in the pharmaceutical industry for making decisions. Originality/value The first is the application of the n-fold compound options on portfolio of R&D projects and the employment of compound options value of a project portfolio as an objective function. The second one is a mathematical formulation of these concepts and solving it by the robust combinatorial optimization algorithm. The literature is lacking in the application of the robust combinatorial optimization algorithm to R&D project portfolio selection based on the generalized n-fold compound option model of Cassimon et al. (2004). Every framework from calculation of the n-fold compound option to solving robust combinatorial algorithm is programmed in Matlab software, since it can be used as a business support tool.

Application of robust optimization for a product portfolio problem using an invasive weed optimization algorithm

“…The resulting model was solved using a robust combinatorial algorithm. The results showed that project details can have a deep impact on even the simplest selection of project portfolio under uncertainty [24]. Qu et al (2017) formulated the problem of multi-objective portfolio optimization with the objective of reducing investment risk and increasing stock return.…”

An Improved Artificial Intelligence Based on Gray Wolf Optimization and Cultural Algorithm to Predict Demand for Dairy Products: A Case Study

“…The values of all these weights range from zero to one. The next step after determining the network weights is to obtain a set of predictions (Y), compare them with their counterparts among real observations (O), and measure the difference in terms of the mean squared error [16], which is presented in Eq. (3).…”

“…Therefore, we will save the first three best solutions and force other search agents (omegas) to update their position according to the position of the best search agents based on Eqs. (10)(11)(12)(13)(14)(15)(16).…”

“…The prediction process was initially calculated using the available data (the target data) for the forecasting each month between 2014 and 2017. For these predictions, the coefficient of determination (R2) and mean square error [16] were calculated and are presented in…”

Prioritizing investments in rapid response vaccine technologies for emerging infections: A portfolio decision analysis

“…In all these studies the notion of uncertainty is partly conflated with that of risk of project failure, or inversely, probability of success (PoS). However, several studies introduce measures of uncertainty that capture variance of R&D portfolio performance more broadly, such as: Value at Risk (VaR) or Conditional Value at Risk (CVaR) [34], fuzzy value [43], reward/loss ratios [38,44,46] or value probability thresholds [34,38,40,[44][45][46]48,49], variance of portfolio value distribution [39,42,44,50,51], semivariance below or above portfolio value thresholds [42], or covariance of portfolio value, cumulative probability distribution of portfolio value and Gini criteria [41]. A final set of methods emerging from the health economic literature attempt to measure the impact of this variance on the probability that a portfolio is chosen [42,52,53].…”