Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
48 Cards in this Set
- Front
- Back
It is reasonable to expect good outcomes to occur more frequently when using a ____________ approach to decision making. |
structured |
|
3 characteristics of of decision problems? |
1. Alternatives - set of alternatives is available 2. Criteria - there is criteria for making the decision 3. States of Nature - events that are out of the decision makers control |
|
What is included in a decision matrix? |
1. Decision 2. Alternatives 3. States of nature 4. Payoff values |
|
The possible outcomes in a payoff matrix are _______ _______ and _______ _______. |
mutually exclusive and collectively exhaustive |
|
Describe what is involved in maximax decision. |
1. determine the max payoff in each row (alternative) 2. select the alternative with the best payoff |
|
Maximax is appropriate when the decision maker can _________ the ________ _________ __________. |
survive the worst case payoff |
|
Describe what is involved in maximin decision. |
1. determine the minimum payoff in each row (alternative) 2. select the alternative with the best minimum payoff |
|
Maximax is _____ _____ but _____ _____. |
high risk but high reward |
|
Maximin is _____ _____ but _____ _____. |
low risk but low reward |
|
What is the minimax decision rule process? |
1. determine the max of each state of nature (column) 2. subtract each payoff from each column maximum |
|
What feature in excel can be used to automate the sensitivity analysis process? |
data table |
|
It is important to accompany any decision analysis with a __________ __________. |
sensitivity analysis |
|
Under what tab can you find Data Tables under? |
What-if Analysis |
|
What are tree diagrams composed of? |
1. Branches 2. Nodes |
|
What are the 3 nodes on trees diagrams and their corresponding shapes? |
1. Decision node - square 2. Event node - circle 3. Terminal node - triangle |
|
How do you rollback a decision tree? |
1. start from the right and work left 2. calculate the EMV at each probability node 3. choose alternative with best EMV at each decision node |
|
What is involved in a contingent decision? |
perform test/analysis and wait for results before making the next decision |
|
A spreadsheet can be seen as a relationship between _____ variables and _____ measures. |
input variables and performance measures |
|
Risk is a function of what? |
- uncertainty in the outcome of a decision - magnitude of the potential losses |
|
2 major issues with with using the expected value of a varying variable in a spreadsheet?
|
1. bottom line performance measure is non-linear 2. potential variability in the performance measure |
|
3 methods of risk analysis? Which of these are the most common? Which is the most powerful? |
1. Best/Worst-case scenario (most common) 2. What-if analysis (most common) 3. Simulation (most powerful) |
|
How is a best/worst-case analysis conducted? |
- calculate base case (most likely scenario) - calculate the best and worst case for uncertain input cells |
|
What is the appeal of the best/worst-case analysis? What is the main problem with this analysis? |
- it is very easy to do - it gives no indication of distribution - it gives no indication of how far the actual outcome might be from the expected value |
|
How is a What-if analysis conducted? |
change the values of uncertain input variables to see what happens to bottom line performance measure |
|
3 major flaws with What-if analysis? |
1. Bias - independant variables may be biased if based on managers judgement 2. Time - thousands of what-if analyses may be needed for accurate picture of variability in performance measure 3. Little value for management - insight may be of little value to top management |
|
How is a simulation analysis performed? |
Repeatedly and randomly generate sample values for each uncertain inpupt variable and then compute the bottom-line performance measure |
|
Why are random variables used in a Simulation analysis? |
- they are assigned in a non-biased way - spreadsheet user doesn't have to determine them |
|
What is the most difficult aspect of decision making? |
uncertainty |
|
Simulation develops a model to evaluate a system numerically over some _____ _____ of interest? |
time period |
|
2 major categories of Simulation? Which kind did we learn? |
1. Monte Carlo (we learned this) 2. Systems |
|
Costs associated with having too much inventory? |
1. Material costs 2. Handling & storage 3. Holding costs 4. Spoilage (& potential salvage) |
|
Costs associated with having too little inventory? |
1. Lost sales & profits 2. Lost goodwill 3. Rush order costs |
|
How can we replicate uncertainty with regards to daily demand of a product? |
generate random numbers |
|
What does the RAND excel function do? |
generates a random number between 0 and 1 with every value being equally likely to occur |
|
What button do you press to generate new random values? |
F9 |
|
How do we automate random number assignment in the excel demand example? |
using VLOOKUP |
|
If demand > stocking level then you have? demand < stocking level? demand = stocking level? |
- shortage costs - excess costs - no extra costs |
|
What type of graph would you use in excel to display a frequency distribution? |
histogram |
|
How do excel add-ins make model building easier? |
1. Scalability - 10,000 replications as easy as 100 replications 2. Automatically create histograms 3. Automatically calculates various statistics |
|
To perform simulation in a spreadsheet, what must we place in each cell that represents an uncertain, independant variable? |
random number generator |
|
What are the 2 types of theoretical distributions built into excel? |
1. Continuous 2. Discrete |
|
What is the results tab in excel RSP used for? |
to statistically monitor cells during a simulation |
|
What is a proportion confidence interval used to determine? |
the range of values which the actual proportion is expected to fall |
|
Data that has been collected using real world observations is called what? |
emperical distribution |
|
Why don't organizations use observed data for simulations? |
observed data may not cover all possibilities (extremes) |
|
What does a seed value do? |
the exact same string of random numbers will be used each time you run a simulation |
|
What sampling method is used in the RSP simulation options for this course? |
Latin Hypercube |
|
What happens if you neglect to set the seed value and/or sampling method in the RSP simulation options? |
the curve for your output will not be smooth |