12.4 Decision Making Under Uncertainty
5 min read•july 30, 2024
Engineering economics often involves making decisions in uncertain environments. This chapter explores methods for quantifying and managing uncertainty in engineering projects. From to Monte Carlo simulations, these tools help engineers make informed choices when faced with incomplete information.
Understanding risk-return trade-offs is crucial in engineering decision-making. This section delves into techniques for assessing and balancing potential risks against expected returns. By applying these concepts, engineers can optimize project outcomes and manage uncertainties effectively.
Uncertainty in Engineering Economics
Sources and Types of Uncertainty
Top images from around the web for Sources and Types of Uncertainty
Engineering Method – Electrical and Computer Engineering Design Handbook View original
Is this image relevant?
Uncertainty in risk engineering: concepts View original
Is this image relevant?
The Decision Making Process | Organizational Behavior and Human Relations View original
Is this image relevant?
Engineering Method – Electrical and Computer Engineering Design Handbook View original
Is this image relevant?
Uncertainty in risk engineering: concepts View original
Is this image relevant?
1 of 3
Top images from around the web for Sources and Types of Uncertainty
Engineering Method – Electrical and Computer Engineering Design Handbook View original
Is this image relevant?
Uncertainty in risk engineering: concepts View original
Is this image relevant?
The Decision Making Process | Organizational Behavior and Human Relations View original
Is this image relevant?
Engineering Method – Electrical and Computer Engineering Design Handbook View original
Is this image relevant?
Uncertainty in risk engineering: concepts View original
Is this image relevant?
1 of 3
- Uncertainty in engineering economic decisions stems from incomplete information about future outcomes and their probabilities
- Sources of uncertainty in engineering projects encompass market conditions, technological changes, regulatory environment, and project-specific risks
- Two main types of uncertainty affect engineering decisions
- Aleatory uncertainty arises from inherent randomness (weather patterns affecting construction timelines)
- Epistemic uncertainty results from lack of knowledge (unknown geological conditions in a mining project)
- Uncertainty significantly impacts engineering economic decisions by affecting project costs, revenues, and overall feasibility
- Example: A new manufacturing plant's profitability depends on uncertain future demand for its products
Methods for Addressing Uncertainty
- Sensitivity analysis examines how changes in input variables affect project outcomes
- Example: Analyzing how different oil prices impact the profitability of an offshore drilling project
- Scenario analysis evaluates project performance under different possible future states
- Example: Assessing a renewable energy project under scenarios of high, medium, and low government subsidies
- Probabilistic approaches incorporate probability distributions of uncertain variables
- Example: Using to model the combined effect of uncertain material costs, labor productivity, and market demand on a construction project's budget
Quantifying Uncertainty
Probability and Statistical Concepts
- Probability theory provides a framework for quantifying the likelihood of uncertain events in engineering economic analysis
- Key probability concepts include:
- Sample space represents all possible outcomes of an uncertain event
- Events are subsets of the sample space
- Probability distributions describe the likelihood of different outcomes (discrete or continuous)
- Statistical methods analyze uncertain data in engineering economics:
- Descriptive statistics summarize and describe data characteristics
- Inferential statistics draw conclusions about populations based on sample data
- Hypothesis testing assesses the validity of claims about population parameters
- Measures characterize uncertain variables in engineering economic analysis:
- Central tendency measures include mean (average value), median (middle value), and mode (most frequent value)
- Dispersion measures include variance (average squared deviation from the mean) and standard deviation (square root of variance)
Advanced Techniques for Uncertainty Analysis
- Monte Carlo simulation models complex systems with multiple uncertain variables
- Example: Simulating project completion time by considering uncertainties in task durations, resource availability, and potential risks
- updates probabilities as new information becomes available
- Example: Refining cost estimates for a novel technology project as prototype testing provides more data
- Value at Risk (VaR) quantifies the potential loss in value of an investment over a specific time period
- Example: Calculating the maximum expected loss on a portfolio of engineering projects with 95% confidence over a one-year horizon
Decision Making Under Uncertainty
Decision Tree Analysis
- graphically represent the sequence of decisions and chance events in a decision-making process under uncertainty
- Components of a decision tree include:
- Decision nodes represent points where a choice must be made
- Chance nodes represent uncertain outcomes
- Branches show possible decisions or outcomes
- Terminal nodes display final outcomes
- (EV) calculation multiplies each possible outcome by its probability and sums these products
- Example: Calculating the expected value of a new product launch by considering different market scenarios and their probabilities
- Optimal decision path determination involves working backwards from terminal nodes, calculating the expected value at each chance node
- Example: Choosing between expanding a manufacturing facility or outsourcing production based on expected values of each option
Advanced Decision-Making Techniques
- Sensitivity analysis applied to decision trees assesses the impact of changes in probabilities or outcome values on the optimal decision
- Example: Evaluating how changes in the probability of technical success affect the decision to invest in a new R&D project
- Real Options Analysis, an extension of decision tree analysis, values flexibility in engineering projects under uncertainty
- Example: Valuing the option to abandon a mining project if mineral prices fall below a certain threshold
- incorporates decision-makers' risk attitudes (risk-averse, risk-neutral, risk-seeking) into the analysis
- Example: Using exponential utility functions to model a company's in evaluating different investment opportunities
Risk vs Return Trade-offs
Quantifying Risk and Return
- Risk in engineering economics represents the potential for negative outcomes or variations from expected results
- Return signifies the potential benefits or profits from an engineering project or investment
- The risk-return trade-off principle states that higher potential returns generally accompany higher levels of risk
- Methods for quantifying risk include:
- Variance measures the spread of possible outcomes around the expected value
- Standard deviation provides a measure of risk in the same units as the original data
- Coefficient of variation allows comparison of risk across investments with different expected returns
- Value at Risk (VaR) estimates the maximum potential loss over a specified time period and confidence level
Risk Management Strategies
- Risk attitudes influence decision-making and can be incorporated into analysis through utility theory
- Example: A risk-averse company may choose a project with lower but more certain returns over a high-risk, high-return alternative
- Portfolio theory and diversification strategies manage risk in engineering economic decisions involving multiple projects or investments
- Example: Balancing a portfolio of energy projects across different technologies and geographical regions to reduce overall risk
- Risk mitigation strategies in engineering projects include:
- Insurance protects against specific risks (property damage, liability)
- Contingency planning develops response strategies for potential risks
- Risk transfer through contracts shifts certain risks to other parties (contractors, suppliers)
- Example: Using fixed-price contracts to transfer cost overrun risks to contractors in a large infrastructure project
Key Terms to Review (16)
Bayesian Analysis: Bayesian analysis is a statistical method that applies Bayes' theorem to update the probability of a hypothesis as more evidence or information becomes available. It combines prior knowledge with new data, allowing for a more flexible approach to decision making under uncertainty, which is particularly valuable when dealing with incomplete or varying information.
Decision Trees: Decision trees are visual representations that help in making decisions by illustrating possible outcomes, risks, and rewards associated with different choices. They provide a clear framework to evaluate potential options and their consequences, making them especially useful in uncertain situations. By breaking down complex decisions into simpler parts, decision trees allow for systematic analysis of various scenarios, which can greatly aid in economic evaluations and multi-criteria decision-making processes.
Environmental Uncertainty: Environmental uncertainty refers to the lack of predictability and the presence of unknown factors that can affect decision-making processes. This uncertainty can arise from various sources such as economic conditions, technological changes, political stability, or competitive dynamics, making it challenging for organizations to plan effectively and allocate resources efficiently.
Expected Value: Expected value is a fundamental concept in probability and decision-making that represents the average outcome of a random variable when considering all possible values and their associated probabilities. It helps in quantifying the potential benefits or costs associated with different choices, allowing for more informed decisions under uncertainty. By calculating expected values, individuals and organizations can evaluate options based on their likely outcomes, making it a crucial tool in risk assessment and strategic planning.
Heuristics: Heuristics are mental shortcuts or rules of thumb that simplify decision-making processes, especially when faced with uncertainty or complex situations. They help individuals make quick judgments and solve problems efficiently without extensive information processing. While heuristics can be helpful, they may also lead to biases and errors in reasoning.
Market uncertainty: Market uncertainty refers to the unpredictable fluctuations and potential risks in a market environment that can affect the behavior of consumers, producers, and investors. This uncertainty can arise from various factors, including economic conditions, regulatory changes, technological advancements, and shifts in consumer preferences. Understanding market uncertainty is crucial for making informed decisions when faced with incomplete information.
Maximin criterion: The maximin criterion is a decision-making strategy used in situations of uncertainty, where the focus is on maximizing the minimum possible payoff. This approach prioritizes caution by selecting the option that has the best worst-case outcome, making it particularly relevant when outcomes are unpredictable and risks need to be minimized.
Minimax regret: Minimax regret is a decision-making criterion used under uncertainty that aims to minimize the maximum potential regret associated with various choices. It involves evaluating possible outcomes for each decision and determining the worst-case scenario, allowing decision-makers to choose the option that has the least potential for regret, based on their risk preferences.
Monte Carlo Simulation: Monte Carlo Simulation is a statistical technique used to model and analyze complex systems by generating random samples from probability distributions to understand the impact of risk and uncertainty on outcomes. This method allows for a comprehensive exploration of possible scenarios, making it a valuable tool in various fields, including systems engineering and decision-making processes.
Prospect theory: Prospect theory is a behavioral economic theory that describes how people make decisions under risk and uncertainty, highlighting that individuals value potential losses and gains differently. This theory challenges the traditional economic assumption of rationality, showing that people are loss-averse, meaning they prefer to avoid losses rather than acquiring equivalent gains. It emphasizes how framing choices and potential outcomes can significantly affect decision-making processes.
Risk aversion: Risk aversion refers to the preference of individuals or organizations to avoid uncertainty and potential losses when making decisions. People who are risk-averse tend to choose options with lower risk and more predictable outcomes, even if it means forgoing higher potential gains. This mindset influences decision-making processes, especially in scenarios where uncertain outcomes can significantly impact success or failure.
Risk premium: A risk premium is the extra return that investors require to hold a risky asset instead of a risk-free asset. This concept highlights the trade-off between risk and return, where higher levels of uncertainty associated with an investment necessitate a higher expected return to entice investors to take on that risk.
Robust optimization: Robust optimization is a mathematical approach used to make decisions under uncertainty, ensuring solutions remain effective even when faced with unpredictable variations in input parameters. This method is particularly valuable in complex environments where data may be incomplete or unreliable, allowing decision-makers to account for worst-case scenarios while still optimizing performance.
Sensitivity Analysis: Sensitivity analysis is a method used to determine how different values of an independent variable will impact a particular dependent variable under a given set of assumptions. It helps in identifying how sensitive an outcome is to changes in input parameters, which is essential for making informed decisions and optimizing processes.
Stochastic optimization: Stochastic optimization is a mathematical approach used to find the best possible solution in problems that involve uncertainty and randomness. It combines traditional optimization techniques with probabilistic models to account for the inherent variability in parameters or outcomes, making it suitable for decision-making in complex environments where not all information is known.
Utility Theory: Utility theory is a framework used to understand how individuals make choices based on the satisfaction or pleasure derived from different outcomes. It connects the concept of preferences with decision-making processes, helping to model how people evaluate risks and rewards when faced with uncertainty or multiple criteria. This theory is crucial for analyzing decisions in economics and various fields, as it provides insight into how preferences influence choices.