Академический Документы
Профессиональный Документы
Культура Документы
▪ Best-Case/Worst-Case Analysis
▪ What-if Analysis
▪ Simulation
Best-Case/Worst-Case Analysis
▪ Best case - plug in the most optimistic values for each of the
uncertain cells.
▪ Worst case - plug in the most pessimistic values for each of
the uncertain cells.
▪ This is easy to do but tells us nothing about the distribution of
possible outcomes within the best and worst-case limits.
Possible Performance Measure Distributions Within a Range
If RAND( ) falls
in the interval: Return value:
0.0 to 0.16667 1
0.16667 to 0.3333 2
0.3333 to 0.5 3
0.5 to 0.66667 4
0.66667 to 0.83333 5
0.83333 to 1 6
Simulation Example
▪ To solve this problem we will simulate the demands for 15 days and
calculate profits from the sales of each day
Simulation Example (continued….)
▪ Variables/Parameters
𝑄 − 𝐷 × 𝑆, 𝑖𝑓 𝑄 ≥ 𝐷
𝑆𝑎𝑙𝑣𝑎𝑔𝑒 𝑓𝑟𝑜𝑚 𝑠𝑎𝑙𝑒 𝑜𝑓 𝑠𝑐𝑟𝑎𝑝 =
𝐸𝑙𝑠𝑒 0,
𝐶𝑜𝑠𝑡 𝑜𝑓 𝑛𝑒𝑤𝑠𝑝𝑎𝑝𝑒𝑟 = 𝑄 × 𝐶
Simulation Example (continued….)
▪ Inputs to the simulation model are the order quantity (decision variable
determined by the newspaper seller), the daily demand which is uncontrollable and
stochastic, and the cost and revenue factors which are set values and do not
change.
▪ Since it does not matter when the customers show up during the day, the problem
can be simulated through a static and stochastic model
▪ We first need to choose a value for the decision variable 𝑄 and then set up the
simulation where the demand is generated from the probability distribution in
Table
Random Digit Assignment For the Daily Demand
Of Newspaper
▪ Interval estimates can be calculated to determine how far off the calculated
Assuming that the average profit and standard deviation calculated are
independent and normally distributed the following equation can be used to
calculate the half width of a confidence interval for a given level of confidence
𝑡𝑛−1,𝛼 2 𝑆
√𝑛
𝑛 → 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑟𝑒𝑝𝑙𝑖𝑐𝑎𝑡𝑖𝑜𝑛
𝑠 → 𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛
𝛼 → 𝑃𝑟𝑜𝑏𝑎𝑏𝑖𝑙𝑖𝑡𝑦 𝑓𝑎𝑙𝑙 𝑜𝑢𝑡𝑠𝑖𝑑𝑒 𝑡ℎ𝑒 𝑐𝑜𝑛𝑓𝑖𝑑𝑒𝑛𝑐𝑒 𝐼𝑛𝑡𝑒𝑟𝑣𝑎𝑙
𝑡𝑛−1,𝛼 2→𝑉𝑎𝑙𝑢𝑒 𝑓𝑟𝑜𝑚 𝑇 𝑡𝑎𝑏𝑙𝑒
Simulation For Purchase Of 80 Newspaper
▪ Different options
– Within the spreadsheet using Rand() and other functions (limited)
– Embed code in the macro
– Add-ons
▪ Crystal Ball, Simtools, @Risk etc.
What is @RISK?
RiskDiscrete({20,21,22,23},{.15,.35,.45,.05}) INT(RiskUniform(20,24))
0.50 0.50
0.40 0.40
0.30 0.30
0.20 0.20
0.10 0.10
0.00 0.00
20 21 22 23 20 21 22 23