Uncertainty and Risk: What Is It?

Uncertainty:

Simply put, uncertainty arises when information about future events is flawed or imperfect. In such cases, the lack of certainty stems from the unreliability of the available information. This unreliable information hampers our ability to accurately predict or understand the future outcome of a given event. From an enterprise perspective, uncertainty can arise during various scenarios in the process, such as when estimating product demand, the arrival time of products, stock returns, and interest exchange rates, among other factors.

Despite uncertainty being a phenomenon that can arise at any point in the business process, it can be mitigated or reduced through the application of standardized work processes. Hence, standardization is critical when attempting to reduce output variability. More importantly, understanding the appropriate actions to take when uncertainty has been identified is crucial because any business decision made under uncertainty involves risk.

Risk:

Each industry has its unique way of defining risk. Generally, risk refers to the uncertainty or variability in potential results. In the business context, it represents the possibility of experiencing unfavorable outcomes or potential losses. Therefore, completely eliminating or avoiding risk from any process is impractical. However, recognizing, assessing, and understanding risk is pivotal in enterprise risk management. The more we comprehend the types of risks we face, the more informed decisions we can make.

Risk Analysis:

Risk analysis encompasses a comprehensive assessment of potential threats linked to a project or business. This evaluation involves estimating and understanding the various factors that may pose risks. Risk assessment often involves gauging the probability and potential impact of adverse outcomes. Increased chances of incurring losses typically signify a higher degree of risk.

Risk Analysis Methods:

Three key methods used to analyze risk in a corporate environment are:
- Best Case / Worst Case Analysis
- What-if Analysis
- Simulation

Best Case Analysis involves using the most optimistic values for uncertain factors within a model. Worst Case Analysis employs the most pessimistic values for these uncertain inputs. Base Case Analysis relies on the most probable or likely values for such uncertain factors. However, despite being the most commonly favored method among managers, the Base Case Analysis approach has its drawbacks, often leading to pitfalls due to the “flaw of averages.”

Moreover, despite the advantage of the Best Case / Worst Case Analysis, a major disadvantage of this method is that it does not estimate the likelihood of occurrence for any of the outcomes.

What-if Analysis employs multiple values for uncertain inputs, encompassing the most optimistic, the most pessimistic, and the most probable values in an attempt to comprehend potential outcomes.

Simulation Analysis (stochastic (probabilistic) simulation) utilizes probability distributions to offer a comprehensive projection of the primary uncertainties encountered in decision-making processes. Probability distributions furnish a detailed representation of uncertainty, encapsulating both the most probable outcome (central tendency) and the potential variation (variability) from this likelihood. Consequently, values for uncertain inputs are usually chosen at random in an unbiased manner within this framework.

Example3:

Should a product line continue in the next quarter?

Let’s assume that a major purchasing order by a current client is on the table. However, there is significant market competition, and a decision must be made without knowing if the purchase will occur. Moreover, there is uncertainty regarding the general level of sales apart from the single major purchase.

Scenario 1 - Should the major purchase occur

Inputs:

# The input: 

SellingPriceUnit <- 10
DiscountedPriceunit <- 9
ProductionCostUnit <- 5

FixedCost <- 500000 
GeneralSaleUnits <- 75000 #best Guess
MajorSaleUnits <- 50000

Inputs <- data.frame(
  Item = c("Selling Price per Unit", "Discounted Price per Unit", "Production Cost per Unit", "Fixed Cost", "General Sale Units", "Major Sale Units"),
  Value = c(SellingPriceUnit, DiscountedPriceunit, ProductionCostUnit, FixedCost, GeneralSaleUnits, MajorSaleUnits)
)

print(Inputs)
##                        Item  Value
## 1    Selling Price per Unit     10
## 2 Discounted Price per Unit      9
## 3  Production Cost per Unit      5
## 4                Fixed Cost 500000
## 5        General Sale Units  75000
## 6          Major Sale Units  50000

Output:

# Revenues: 
RevGeneralSales <- SellingPriceUnit * GeneralSaleUnits
RevMajorSale <- DiscountedPriceunit * MajorSaleUnits
TotalRevenues <- RevGeneralSales + RevMajorSale

# Expenses: 
VariableCost <- ProductionCostUnit * (GeneralSaleUnits + MajorSaleUnits)
TotalExpenses <- FixedCost + VariableCost

# Profit/Loss
Profit <- TotalRevenues - TotalExpenses

# Reporting:
Reporting <- data.frame(
  Category = c("Revenues", "Revenues", "Revenues", "Expenses", "Expenses", "Profit/Loss"),
  Item = c("RevGeneralSales", "RevMajorSale", "TotalRevenues", "VariableCost", "TotalExpenses", "Profit"),
  Amount = c(RevGeneralSales, RevMajorSale, TotalRevenues, VariableCost, TotalExpenses, Profit)
)

print(Reporting)
##      Category            Item  Amount
## 1    Revenues RevGeneralSales  750000
## 2    Revenues    RevMajorSale  450000
## 3    Revenues   TotalRevenues 1200000
## 4    Expenses    VariableCost  625000
## 5    Expenses   TotalExpenses 1125000
## 6 Profit/Loss          Profit   75000
# Therefore, with the major porche, the estimated profit is $75,000

Scenario 2 - Should the major purchase not occur:

Inputs:

# The input: 

SellingPriceUnit <- 10
DiscountedPriceunit <- 9
ProductionCostUnit <- 5

FixedCost <- 500000 
GeneralSaleUnits <- 75000
MajorSaleUnits <- 0

Inputs <- data.frame(
  Item = c("Selling Price per Unit", "Discounted Price per Unit", "Production Cost per Unit", "Fixed Cost", "General Sale Units", "Major Sale Units"),
  Value = c(SellingPriceUnit, DiscountedPriceunit, ProductionCostUnit, FixedCost, GeneralSaleUnits, MajorSaleUnits)
)

print(Inputs)
##                        Item  Value
## 1    Selling Price per Unit     10
## 2 Discounted Price per Unit      9
## 3  Production Cost per Unit      5
## 4                Fixed Cost 500000
## 5        General Sale Units  75000
## 6          Major Sale Units      0

Output:

# Revenues: 
RevGeneralSales <- SellingPriceUnit * GeneralSaleUnits
RevMajorSale <- DiscountedPriceunit * MajorSaleUnits
TotalRevenues <- RevGeneralSales + RevMajorSale

# Expenses: 
VariableCost <- ProductionCostUnit * (GeneralSaleUnits + MajorSaleUnits)
TotalExpenses <- FixedCost + VariableCost

# Profit/Loss
Profit <- TotalRevenues - TotalExpenses

# Reporting:
Reporting <- data.frame(
  Category = c("Revenues", "Revenues", "Revenues", "Expenses", "Expenses", "Profit/Loss"),
  Item = c("RevGeneralSales", "RevMajorSale", "TotalRevenues", "VariableCost", "TotalExpenses", "Profit"),
  Amount = c(RevGeneralSales, RevMajorSale, TotalRevenues, VariableCost, TotalExpenses, Profit)
)

print(Reporting)
##      Category            Item  Amount
## 1    Revenues RevGeneralSales  750000
## 2    Revenues    RevMajorSale       0
## 3    Revenues   TotalRevenues  750000
## 4    Expenses    VariableCost  375000
## 5    Expenses   TotalExpenses  875000
## 6 Profit/Loss          Profit -125000
# Therefore, without the major purchase, the estimated profit is $-125,000. Thus, it is evident that, whether the company gets the major purchase or not does a huge impact on the decision of whether it should continue the product line

  1. Financial Analyst (https://www.linkedin.com/in/tklvchv/)↩︎

  2. Disclaimer: The content provided represents my personal in-class notes for the MET AD616 lecture at Boston University (https://www.bu.edu/met/) within the MSc. Applied Business Analytics program. These notes are solely an interpretation of the class material and should be regarded as guidance or supplementary information for individuals interested in Enterprise Risk Analysis. It’s crucial to clarify that they do not represent the official stance of Boston University or its educational curriculum. These notes are shared with the intent of contributing to accessible education, but they should not be misconstrued as an official statement or viewpoint held by the university or its faculty. It’s essential to understand that these notes are based on personal interpretation. Thank you for your interest and support. I hope you find this information beneficial in your pursuit of knowledge and understanding.↩︎

  3. Carraway, R. L, Analyzing Uncertainty: Probability Distributions and Simulation, 2005.↩︎