Decisions are a part of our everyday life. They can include a low-stake decision like buying a cup of coffee before your drive into the office, or a high-stake decision like buying a house. As engineering professionals, your project work requires a multitude of decisions and the opinions of stakeholders throughout the process. Decisions made in a project setting often require a more structured approach to produce a rational and auditable methodology for determining a choice between competing options.

A decision analysis process can help build consensus among stakeholders, consider a wide range of options, identify potential risks, and develop a plan with specific actions. There are several that are commonly used in project management including the Kepner-Tregoe method and Multiple Accounts Analysis. The advantage of a decision analysis process is it can bring together informed people in many fields (fields of interest are called accounts) and can include social, environmental, technical, and cost aspects of a project. These fields of interest often have competing requirements and different risk profiles which the process describes in plain language and evaluates from different stakeholder viewpoints, often in a workshop setting.

Define the Problem

The first step when conducting a decision analysis is defining the problem. During this step, you will develop a thorough description of the situation, its purpose, and identify a core team of stakeholders from a variety of fields (geotechnical, environmental, operations, finance, etc.) who can contribute. The core team should determine the purpose of the decision and any constraints that will guide the scope of the process. It is during this step that you will consider multiple criteria, identify the account structure (social, environmental, technical, cost, etc.), and define assumptions.

Establish the Objective

Once you have developed an understanding of the problem, you can establish the objectives of the decision. Start by compiling a list of objectives and divide them into “musts” and “wants”. "Must” objectives are criteria that must be met for an alternative to be successful (e.g. regulatory criteria). “Want” objectives provide the means of differentiating between options (e.g. maximize opportunity to reach passive care) and do not need to be met for an alternative to succeed.

Identify Alternatives

In this step, you will identify alternatives to meet the decision. These can be achieved through a matrix of elements to develop various options which are discussed throughout the process. If any alternatives do not fulfill all the “must” objectives, screen them out.

Engage Stakeholders

Now, compare the alternatives to the defined objectives. (This is typically undertaken in a workshop setting with key stakeholders and a capable facilitator guiding the process.) Rank each alternative based on its ability to achieve the objectives. For this step, alternatives must meet all the “must” objectives and are evaluated against the “want” objectives. Assess your results by conducting sensitivity analyses and a risk assessment to reduce the overall risk to as minimal as possible.

Make your Decision

Decide on an alternative based on how it ranks above other alternatives, its costs, and risks. Once a decision has been made, develop an action plan, documenting the approach and rationale of the decision, to move the project forward. This step usually includes a forward high-level work plan for the project.

Tips when conducting a decision analysis:

• Be clear on the situation appraisal and problem analysis prior to undertaking the decision analysis.
• Have the right people in the room to make decisions. Identify and include key stakeholders to increase the success of the decision analysis process.
• Achieve consensus at each step.
• An objective framing workshop is a useful way to engage decision makers and identify policy, strategic and tactical objectives.
• Characterize each alternative with a supporting body of knowledge enough to compare each option without prior judgement.
• It can be as simple or complicated as it needs to be. But do not over-complicate. Not all the answers are needed to make an informed decision.