Decision Analysis Example: How to Use Decision Analytics

Decision analysis formalizes the decision-making process by breaking down every component of a business decision and assessing all of its possible outcomes. Decision analysis is a crucial aspect of risk analysis, operations research, new product launches, and optimization strategies. It ensures a company considers all the relevant information before making an important decision. For example, businesses might use decision analysis when weighing a hefty investment decision.

Several decision-making tools enable the decision analysis process by providing a visual representation of every aspect of the decision in question. Decision trees, influence diagrams, and expected value equations help businesses make strategic decisions by comparing every possible outcome and its associated risks, opportunities, trade-offs, and utility functions. Decision-making tools can be straightforward or can involve highly complex algorithms.

Decision analysis is a quantitative and systematic approach to making optimal decisions. The specifics of the decision analysis process can vary depending on the problem it is solving, but generally, the following steps are involved in every decision analysis process.

• Identify the problem. When you face a complex decision, the first step is to identify factors that make your decision difficult so you can seek the right decision support. For example, a company making changes to its pricing strategy needs to consider all the possible effects of the change and whether or not a change would help them achieve a desirable outcome.
• Conduct data analysis. Descriptive and predictive analytics can help businesses make informed decisions. These business intelligence metrics provide insight into the past and forecast future trends or possibilities. Analysts collect and model this data using a decision analysis tool that enables their company to plan a course of action.
• Use decision modeling. Influence diagrams and decision trees can help you visualize the problem and identify possible solutions throughout the decision analysis process. Influence diagrams are beneficial for providing a high-level overview of a decision problem. They involve assigning every variable of the decision to a box or circle. An arrow connects each decision node to a possible outcome, uncertainty, or alternative. Decision trees represent the decision problem in the shape of a tree. Each “branch” represents a possible outcome of the decision. These models are most common due to their simplicity. They allow decision analysts to model and evaluate every variable of a decision problem and find the most favorable outcome.
• Calculate the expected value. Calculating the expected value (EV) helps determine the average outcome of each business decision and avoid uncertain outcomes. To find the expected value of a decision, you must evaluate the probability of each outcome and assign it a numerical or monetary value. If you’re familiar with Bayes’ theorem, you might take a bayesian approach to determining the probability. Alternatively, once you’ve collected and modeled all the relevant data, you can input that information into the following equation to calculate the expected value: EV = (Probability A x Expected Profit A) + (Probability B x Expected Profit A).

Consider this real-world example of decision analysis in action: A restaurant considers opening a second location in Chicago or New York. Opening in either city will involve different costs and varying rates of success. For risk assessment purposes, the project management team hires a risk management firm to gather data and input it into a decision tree. The decision tree shows the relevant monetary values they will use to calculate the expected value.

With regard to Chicago, the data shows a twenty percent chance of success and an eighty percent chance of failure. With regard to New York, there’s a forty percent chance of success and a sixty percent chance of failure. The firm estimates it will cost $1 million to open in Chicago and $3 million to open in New York. Furthermore, they expect a profit of $15 million in Chicago and a possible loss of $2 million. In New York, they expect a profit of $30 million and a possible loss of $9 million.

Next, the firm runs these numbers through the expected value equation:

• EV (Chicago)= (0.2 x $15,000,000) + (0.8 x -$2,000,000) = $1,400,000
• EV (New York)= (0.4 x $30,000,000) + (0.6 x -$9,000,000) = $6,600,000

To calculate the net gain and net loss for each location, the firm must subtract the expected value from the upfront costs:

• Chicago: $1,400,000 - $1,000,000 = $400,000
• New York: $6,600,000 - $3,000,000 = $3,600,000

These calculations determine New York will yield the most favorable results, meaning the stakeholders of the restaurant should decide to open their second location in New York instead of Chicago.

MasterClass at Work is the learning platform to help unlock the full potential of your employees and inspire a learning lifestyle in everyone. From negotiation to mindfulness to baking—these are just a few things your teams can explore.