Stanford University Professor Ronald A. Howard first defined decision analysis as a profession in 1964. Over the ensuing decades, Howard has supervised many doctoral theses on the subject across topics including nuclear waste disposal, investment planning, hurricane seeding, and research strategy. Decision analysis (DA) is a systematic, visual, and quantitative decision-making approach where all aspects of a decision are evaluated before making an optimal choice.
Understanding decision analysis
Fundamentally, decision analysis enables organizations to evaluate or model the potential outcomes of various decisions so they can choose the one with the most favorable outcome. The tool assesses all relevant information and incorporates aspects of training, economics, psychology, and various management techniques.
Another part of decision analysis requires the business to examine uncertainty around a decision. Uncertainty is measured by probability. In other words, what are the chances the outcome will occur? From this point, the organization can make a decision based on the value and likelihood of success of a decision. Alternatively, it can base the decision on the likelihood of failure and its corresponding impact.
Decision analysis is extremely valuable in the project planning stage and during periodic reviews of project progress by senior management. Since most projects are characterized by decisions made with high uncertainty, decision analysis has multiple applications. For one, the analysis helps project teams obtain accurate activity duration estimates. Decision analysis also assists in risk analysis, “what-if” analysis, and subproject terminating in a research and development context.
The Significance of Decision Analysis
Decision Analysis is significant for several reasons:
1. Informed Decision-Making
It provides a framework for making decisions based on rigorous analysis, ensuring that choices are well-informed rather than relying solely on intuition or guesswork.
2. Risk Assessment
Decision Analysis helps assess and quantify the risks associated with different choices. This is particularly crucial in high-stakes decisions with potential consequences.
3. Resource Allocation
It aids in efficient resource allocation by guiding organizations in choosing the best courses of action, thereby optimizing the use of resources.
4. Improved Problem-Solving
Decision Analysis offers structured problem-solving techniques that can be applied to complex and multifaceted decision scenarios.
5. Strategic Planning
It supports strategic planning by helping organizations make choices that align with their long-term objectives and goals.
Steps in Decision Analysis
Conducting Decision Analysis involves several key steps to systematically evaluate options and make informed choices. Here’s an overview of the essential steps:
1. Define the Decision Problem
Clearly define the decision problem or the question that needs to be answered. This step involves specifying the objectives, criteria, and constraints of the decision.
2. Identify Alternatives
Generate a list of possible alternatives or courses of action that can be taken to address the decision problem. Ensure that all relevant options are considered.
3. Specify Decision Criteria
Identify the criteria or factors that will be used to evaluate and compare the alternatives. Criteria can include cost, time, quality, risk, and any other relevant considerations.
4. Collect Data and Information
Gather data and information related to the alternatives and criteria. This may involve conducting research, surveys, or assessments.
5. Quantify Uncertainty
Assess the uncertainty associated with the decision by quantifying probabilities and uncertainties related to outcomes, especially in situations where uncertainty is high.
6. Build Decision Models
Construct decision models that represent the relationships between alternatives, criteria, and outcomes. Decision models can take the form of decision trees, influence diagrams, or other appropriate structures.
7. Analyze the Decision
Use the decision models to evaluate and compare the alternatives. This involves calculating expected values, considering sensitivity analysis, and assessing trade-offs.
8. Make the Decision
Select the alternative that best aligns with the objectives and criteria established in the first steps. The chosen alternative should have the highest expected value or utility.
9. Implement the Decision
Put the chosen alternative into action. This step may involve allocating resources, assigning responsibilities, and initiating the chosen course of action.
10. Monitor and Review
Continuously monitor the outcomes of the decision and assess whether they align with the expected results. Adjustments may be necessary based on real-world feedback.
11. Communicate the Decision
Effectively communicate the decision and its rationale to stakeholders, ensuring that all relevant parties are informed and on board.
Real-World Applications of Decision Analysis
Decision Analysis is widely applied in various sectors and industries:
Case Study 1: Business
In the business world, Decision Analysis is used to make strategic decisions about investments, product development, market entry, and resource allocation. It helps organizations choose the most profitable and sustainable paths forward.
Case Study 2: Healthcare
In healthcare, Decision Analysis supports clinical decision-making, treatment planning, and healthcare resource allocation. Physicians use it to assess treatment options and make recommendations based on patient outcomes.
Case Study 3: Environmental Management
Environmental agencies and organizations use Decision Analysis to evaluate the environmental impact of projects, policies, and regulations. It helps in making decisions that balance economic development with environmental conservation.
Case Study 4: Project Management
Project managers apply Decision Analysis to assess project risks, choose project management methodologies, and make critical decisions regarding project scope, scheduling, and resource allocation.
Case Study 5: Public Policy
Government agencies use Decision Analysis to evaluate policy options, budget allocation, and the potential consequences of different policy choices. It supports data-driven policymaking.
Limitations and Considerations
While Decision Analysis is a valuable tool for making informed decisions, it is important to consider its limitations and potential challenges:
1. Data Availability
The quality and availability of data can significantly impact the accuracy and reliability of the analysis. Incomplete or biased data can lead to flawed decisions.
2. Complexity
Decision Analysis can become complex in situations with numerous alternatives, criteria, and uncertainties. Managing this complexity requires careful modeling and analysis.
3. Subjectivity
Some aspects of Decision Analysis, such as assigning probabilities or utility values, may involve subjective judgments, which can introduce bias.
4. Implementation Challenges
Implementing the chosen alternative may face challenges, especially when it requires significant resource allocation or changes in organizational practices.
5. Uncertainty
While Decision Analysis can quantify uncertainty to some extent, it cannot eliminate it entirely. There will always be inherent uncertainties in complex decisions.
How does decision analysis work in practice?
The decision analysis process can be explained in the following steps.
1 – Identify the problem
What is the problem to be solved or the decision to be made?
Once this has been determined, a list of possible options should be devised. For instance, a non-profit that receives a large endowment may have several ways they can put the money to good use.
2 – Research options
Each choice or option must then be researched, with any relevant data set aside to develop a decision model later in the process. Data may be quantitative or qualitative, depending on the context.
It is important to consider each outcome in terms of its costs, risks, benefits, and probability of success or failure.
3 – Create a framework
To allow the business to properly assess its options, an evaluation framework must be created.
One way to achieve this is by using key performance indicators (KPIs) to measure and indicate progress. For example, a business looking to expand may stipulate that each potential new market causes a minimum increase in monthly sales volume.
Like the research from the previous step, KPI data may be qualitative or quantitative.
4 – Develop a decision model
Now it is time to combine the framework with a decision model. One of the most popular decision analysis models is the decision tree, where each choice has branches representing different outcomes.
Influence diagrams can also be used when there is a high amount of uncertainty around a decision or goal.
5 – Calculate the expected value
The expected value (EV) is the weighted average of all potential decision outcomes. To calculate the expected value, multiply the probability of each outcome occurring by the resulting value – sometimes referred to as the expected payoff. Then, sum the expected values for each decision.
For example, consider a large architectural firm that designs stadiums. During a public tender process, the firm submits two designs which the city council must evaluate for viability. For the sake of this article, we will call them Design A and Design B.
The city council determines that Design A, once completed, has a 55% chance of a $350 million valuation and a 25% chance of a $275 million valuation. The expected value of Design A is (0.55 x 350,000,000) + (0.25 x $275,000,000) = $261.25 million
On the other hand, Design B has a 20% chance of being valued at $400 million and a 60% chance of being valued at $290 million. The expected value of Design B is (0.20 x 400,000,000) + (0.60 x 290,000,000) = $254 million.
In this instance, the council should choose Design A.
Case Studies
- Pharmaceutical Company: New drug development versus other R&D projects.
- Identify the problem: Whether to invest in the development of a new drug or allocate resources to other R&D projects.
- Research options: Analyze the success rate of similar drugs, potential market size, regulatory challenges, and the competitive landscape.
- Create a framework: Use KPIs like potential ROI, time-to-market, and success rate of clinical trials.
- Develop a decision model: Use a decision tree to map out each phase of drug development and the probabilities of success/failure at each stage.
- Calculate the expected value: Estimate potential revenues from the new drug against the costs and risks, and compare with the expected returns from other R&D projects.
- Oil and Gas Exploration: To drill or not to drill.
- Identify the problem: Assess the potential profitability of drilling in a new location.
- Research options: Investigate geological data, historical successes in similar terrains, and current oil market conditions.
- Create a framework: KPIs might include estimated oil reserves, drilling costs, potential environmental impact, and breakeven oil price.
- Develop a decision model: A decision tree that weighs the cost of drilling, potential oil yield, and market prices.
- Calculate the expected value: Factor in the potential revenues from oil sales against drilling and operational costs.
- Venture Capital Investment: To invest in a particular startup or not.
- Identify the problem: Determine if the startup presents a lucrative investment opportunity.
- Research options: Assess the startup’s business model, market potential, competition, and team competency.
- Create a framework: KPIs can include projected ROI, market share potential, and scalability.
- Develop a decision model: Use a decision tree to model potential growth trajectories and associated risks.
- Calculate the expected value: Weigh potential returns against the investment amount and associated risks.
- Supply Chain Management: Selecting a supplier based on cost versus reliability.
- Identify the problem: Decide between a cheaper supplier with a less reliable track record or a slightly more expensive but trusted supplier.
- Research options: Examine past performance, reviews, and testimonials of the suppliers.
- Create a framework: KPIs might encompass delivery timelines, defect rates, and communication efficiency.
- Develop a decision model: Decision tree comparing potential disruptions or benefits from each supplier.
- Calculate the expected value: Factor in potential costs from supply chain disruptions against savings from the cheaper supplier.
- Environmental Policy: Renovate an old water treatment plant or build a new one.
- Identify the problem: Determine the most efficient and sustainable water treatment solution for the city.
- Research options: Compare the costs, efficiency, and environmental impact of renovation versus new construction.
- Create a framework: KPIs could include water treatment capacity, operational costs, and environmental compliance levels.
- Develop a decision model: Decision tree analyzing the long-term benefits and costs of both options.
- Calculate the expected value: Evaluate the long-term savings and benefits against initial costs.
- Real Estate Developer: Build a residential complex or a commercial complex on a newly acquired land.
- Identify the problem: Decide the best utilization of the newly acquired land for maximum profit and sustainability.
- Research options: Assess the local demand for housing versus commercial spaces, potential rent or sale prices, and local infrastructure.
- Create a framework: KPIs might include potential ROI, occupancy rates, and maintenance costs.
- Develop a decision model: Decision tree considering construction costs, potential revenue, and long-term market trends.
- Calculate the expected value: Compare potential profits from both residential and commercial complexes against construction and maintenance costs.
- Tech Company: Launch a new software product or improve an existing one.
- Identify the problem: Determine where to allocate resources for product development.
- Research options: Analyze market demand for new features, competition, and feedback on the current product.
- Create a framework: KPIs can include user adoption rates, customer retention, and potential market share.
- Develop a decision model: Decision tree weighing the costs of new development versus enhancement and potential market reception.
- Calculate the expected value: Contrast potential sales and subscription revenues against development and marketing costs.
- Agricultural Producer: Invest in organic farming or continue with traditional methods.
- Identify the problem: Determine the farming method that will yield maximum profit and sustainability.
- Research options: Study market demand for organic produce, cost implications, and potential yield differences.
- Create a framework: KPIs might encompass crop yield per acre, market prices, and long-term soil health.
- Develop a decision model: Decision tree comparing the short-term and long-term costs and benefits of both farming methods.
- Calculate the expected value: Evaluate potential premium prices for organic produce against increased costs and potential yield variations.
- Automobile Manufacturer: Introduce electric vehicles (EVs) or improve fuel efficiency in traditional vehicles.
- Identify the problem: Decide on the product direction in light of changing environmental regulations and market demands.
- Research options: Analyze market trends for EVs, technological advancements, and potential government incentives.
- Create a framework: KPIs can include sales volume, profit margins, and brand perception.
- Develop a decision model: Decision tree weighing the R&D costs, production costs, and potential market share of both options.
- Calculate the expected value: Compare potential profits from EVs and improved traditional vehicles against development and production costs.
- Tourism Board: Invest in promoting local tourism or attract international tourists.
- Identify the problem: Determine where marketing efforts and investments will yield maximum tourist influx and revenue.
- Research options: Assess the current state of local versus international tourism, accessibility, and attractions.
- Create a framework: KPIs might include tourist footfall, average spend per tourist, and hotel occupancy rates.
- Develop a decision model: Decision tree analyzing the potential reach and effectiveness of local versus international marketing campaigns.
- Calculate the expected value: Weigh potential revenues from increased tourism against marketing and infrastructure investment costs.
Key takeaways:
- Decision analysis is a systematic, visual, and quantitative decision-making approach where all aspects of a decision are evaluated before making an optimal choice.
- Decision analysis is used in the project planning stage and during periodic reviews of project progress by senior management. The approach is especially suited to project management where there is often uncertainty around decision outcomes.
- Decision analysis occurs via five steps: identify the problem, research options, create a framework, develop a decision model, and calculate the expected value. At the heart of this process are the decision tree framework and the calculation of expected value.
Key Highlights
- Understanding Decision Analysis: Decision analysis is a systematic, visual, and quantitative approach to decision-making that evaluates all aspects of a decision before making an optimal choice. It involves assessing relevant information, considering uncertainty, and incorporating various disciplines like training, economics, psychology, and management techniques.
- Application of Decision Analysis: Decision analysis is valuable in project planning and project progress reviews, especially for projects with high uncertainty. It helps obtain accurate activity duration estimates, conduct risk analysis, perform “what-if” analysis, and make decisions in a research and development context.
- The Decision Analysis Process:
- Identify the Problem: Define the decision or problem to be solved and generate a list of possible options.
- Research Options: Gather relevant data for each option, considering costs, risks, benefits, and probability of success or failure.
- Create a Framework: Develop an evaluation framework using key performance indicators (KPIs) to measure progress and assess options.
- Develop a Decision Model: Use decision trees or influence diagrams to create decision models that represent different outcomes for each choice.
- Calculate the Expected Value: Calculate the weighted average of all potential decision outcomes to determine the expected value (EV) for each option. Select the option with the highest expected value.
- Decision Analysis Example: For instance, a large architectural firm submitting stadium designs for a public tender has Design A with a 55% chance of a $350 million valuation and a 25% chance of a $275 million valuation. The expected value of Design A is $261.25 million. Design B, with a 20% chance of $400 million and a 60% chance of $290 million, has an expected value of $254 million. Based on expected values, the city council should choose Design A.
Decision Analysis | Description | Analysis | Implications | Applications | Examples |
---|---|---|---|---|---|
1. Define the Decision Problem (DDP) | Decision Analysis begins by clearly defining the decision problem or choice to be made. | – Describe the decision problem, its objectives, and the key alternatives under consideration. – Identify the relevant stakeholders and their interests. – Establish a decision timeframe and any relevant constraints. | – Provides a clear and unambiguous understanding of the decision context. – Ensures alignment with organizational goals and stakeholder expectations. | – Evaluating the selection of a new product to develop within a technology company. – Assessing the choice of a location for a new manufacturing facility. | Decision Problem Definition Example: Defining whether to enter a new market segment by launching a new product. |
2. Identify Decision Criteria (IDC) | Identify and define the criteria that will be used to evaluate and compare the alternatives. | – List and describe the specific decision criteria that are relevant to the problem. – Distinguish between quantitative and qualitative criteria. – Assign weights or importance values to each criterion to reflect its relative significance. | – Ensures that all relevant factors and dimensions are considered during the analysis. – Allows decision-makers to prioritize criteria based on their importance. | – Assessing potential real estate investment options based on criteria like location, cost, and potential ROI. – Evaluating job candidates for a critical position using criteria such as skills, experience, and cultural fit. | Decision Criteria Identification Example: Defining criteria like profitability, market demand, and environmental impact for a product launch decision. |
3. Generate Decision Alternatives (GDA) | Generate a set of potential alternatives or options that could address the decision problem. | – Brainstorm and identify a range of alternatives that have the potential to achieve the decision objectives. – Ensure that the alternatives cover different approaches or courses of action. – Avoid prematurely eliminating options to maintain creativity and diversity. | – Explores various possibilities and approaches for addressing the decision problem. – Enables a comprehensive analysis of the pros and cons of each alternative. | – Developing product design alternatives for a consumer electronics company. – Considering different investment strategies for a financial portfolio. | Decision Alternatives Generation Example: Generating product launch alternatives, including different target markets and pricing strategies. |
4. Assess Uncertainty and Risks (AUR) | Identify uncertainties and risks associated with each alternative and the decision environment. | – Identify and describe sources of uncertainty and risks that may affect the outcomes of each alternative. – Assess the likelihood and potential impact of each uncertainty or risk event. – Consider the time dimension and how uncertainties may evolve over time. | – Highlights potential challenges and uncertainties that may impact the decision. – Allows for the incorporation of probabilistic analysis, sensitivity analysis, and risk mitigation strategies. | – Evaluating the investment risk associated with different financial instruments. – Assessing the risks of market volatility in a portfolio optimization decision. | Uncertainty and Risks Assessment Example: Analyzing the potential risks of supply chain disruptions in the manufacturing process for a product. |
5. Perform Multi-Criteria Evaluation (MCE) | Evaluate and compare the alternatives using the defined decision criteria. | – Apply the decision criteria to assess each alternative systematically. – Use quantitative and qualitative information to rate or score each alternative for each criterion. – Aggregate the scores to obtain an overall evaluation for each alternative. – Consider the impact of uncertainty and risks in the evaluation. | – Provides a structured and comprehensive evaluation of the alternatives based on multiple criteria. – Facilitates a transparent and defensible decision-making process. | – Selecting a supplier for a manufacturing company based on criteria like cost, quality, and reliability. – Choosing a location for a new retail store based on factors such as demographics, competition, and accessibility. | Multi-Criteria Evaluation Example: Scoring different software vendors for a business’s IT solution based on criteria like functionality, cost, and vendor reputation. |
6. Make Informed Decisions (MID) | Based on the analysis, make informed decisions about which alternative to pursue. | – Consider the results of the multi-criteria evaluation alongside other qualitative and strategic factors. – Decide on the preferred alternative based on its overall evaluation and alignment with goals and priorities. – Develop an implementation plan and monitor the decision’s progress. | – Facilitates data-driven and rational decision-making that accounts for multiple dimensions. – Ensures that the selected alternative aligns with organizational objectives and stakeholder interests. | – Deciding whether to invest in a new product development based on its overall evaluation score. – Choosing a merger or acquisition target based on a comprehensive evaluation of candidates. | Decision-Making Example: Selecting a supplier for a manufacturing company after considering cost, quality, and risk factors. |
Connected Analysis Frameworks
Failure Mode And Effects Analysis
Related Strategy Concepts: Go-To-Market Strategy, Marketing Strategy, Business Models, Tech Business Models, Jobs-To-Be Done, Design Thinking, Lean Startup Canvas, Value Chain, Value Proposition Canvas, Balanced Scorecard, Business Model Canvas, SWOT Analysis, Growth Hacking, Bundling, Unbundling, Bootstrapping, Venture Capital, Porter’s Five Forces, Porter’s Generic Strategies, Porter’s Five Forces, PESTEL Analysis, SWOT, Porter’s Diamond Model, Ansoff, Technology Adoption Curve, TOWS, SOAR, Balanced Scorecard, OKR, Agile Methodology, Value Proposition, VTDF Framework, BCG Matrix, GE McKinsey Matrix, Kotter’s 8-Step Change Model.
Main Guides: