Decision Analysis

Decision analysis is a method for making decisions involving uncertainty and risk. Decision analysis helps structure a problem, removing extraneous detail while preserving important information, to help analyze any decision. This process links facts, judgments, and decisions in such a way to help solve problems and communicate the results and rationale to others.

Decision trees and influence diagrams are powerful ways to break complex problems down into their component parts for easier analysis. We work with you to identify the points in the problem where your company has the ability to make decisions, the range of possible alternatives, as well as areas of uncertainty. Once the problem is structured, we assign values to decision outcomes and probabilities. Given this information, we can help you determine the optimal path to pursue.

Case Study 1: A small company was developing Product A and later had the opportunity to acquire another product, Product B. The company asked Objective Insights to model the decision, which included alternatives such as acquire Product B, don’t acquire Product B, develop Product A, and out-license Product A. The company was concerned about (uncertain about) immediate-term funding, mid-term funding, and the results of Product B’s clinical trials. The best course of action was determined to be “Acquire Product B.”Decision Tree Example

Case Study 2: One of our clients for whom we were conducting a forecasting project was trying to decide on a regulatory filing strategy. The client could either file sooner with a narrower label or wait for an additional trial to complete (with attendant uncertainty as to outcome) and then file with a broader label, better marketing message, and higher price. Based on the outcomes of the decision analysis, we recommended filing sooner and submitting a supplemental application if results of the ongoing trial were positive.

For more information, please download the Decision Analysis Overview (pdf).

Return to Decision Analysis & Modeling