EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

3318. Behavioural Real Options Analysis of Projects using Machine Learning

Invited abstract in session TC-7: Behavioural OR meets Information systems, stream Behavioural OR.

Tuesday, 12:30-14:00
Room: 1019 (building: 202)

Authors (first author is the speaker)

1. Marius Wittke

Abstract

Investment projects typically involve flexibility of action, also known as managerial flexibility. Managerial flexibility has a value that can be quantified using Real Options Analysis. In order to exercise managerial flexibility, information is required. Unlike financial options, there is typically no objective market information available for real options. Instead, the decision maker usually has noisy information. This means that the current value and relevant factors affecting the project have to be estimated and therefore decisions have to be made on the basis of imperfect information. These estimates are subjective and can be affected by cognitive biases. Cognitive biases are systematic errors in judgement and decision making. As a result, cognitive biases can lead to poor decisions, which can reduce the value of managerial flexibility. In this context, the question arises as to how wrong decisions due to cognitive biases can be avoided. Since cognitive biases can be interpreted as a deviation from rational action, it would be helpful to be able to identify an optimal policy as a benchmark. However, this can be challenging, especially with noisy information. Machine learning algorithms could be useful in deriving an optimal policy. The basic idea is that wrong decisions lead to opportunity costs that can be minimised by an algorithm based on a simulation model. Neural Networks and Reinforcement Learning could be used to make optimal decisions.

Keywords

Status: accepted

Back to the list of papers