Areas Of Simulation Application Waiting lines/service Inventory management Production & manufacturing systems Supply chain systems Service operations Environmental & resource analysis
Building a Simulation Model Define the problem, objectives and variables Random variables (source of uncertainty) Decision variables Create a model to represent the problem Design the logic between the variables (formulas, flowcharts) Collect data for the probability distributions Create a computer program or template
Building a Simulation Model (continued) Validate the model Check formulas and logic especially in copied rows Design “what-if” strategies for the decision variables Run the model MANY times for each possible “what-if” strategy to see what would happen on average, in the worst case as well as in the best case scenario.
Monte Carlo Simulation Use random numbers that have an equal likelihood of being selected (lottery balls) to select numbers from a probability distribution Use these values to observe how a model of a system performs over time
Distribution Of Demand Cases demanded Frequency of Probability of per week, x Demand Demand P(x) 14 20 0.20 15 40 0.40 16 20 0.20 17 10 0.10 18 10 0.10
Roulette Wheel Of Demand 90 x = 18 x = 14 20 80 x = 17 x = 16 x = 15 60
Russell/Taylor Oper Mgt 3/e Random Number Table 39 65 76 45 45 19 90 69 64 61 73 71 23 70 90 65 97 60 12 11 72 18 47 33 84 51 67 47 97 19 75 12 25 69 17 17 95 21 78 58 37 17 79 88 74 63 52 06 34 30 © 2000 by Prentice-Hall Inc Russell/Taylor Oper Mgt 3/e
Generating Demand From Random Numbers Demand Ranges of Random Numbers x r 14 0-19 15 20-59 r = 39 16 60-79 17 80-89 18 90-99
15 Weeks Of Demand Average demand = 241/15 = 16.1 cases per week Week r Demand (x) 1 39 15 2 73 16 3 72 16 4 75 16 5 37 15 6 02 14 7 87 17 8 98 18 Week r Demand (x) 9 10 14 10 47 15 11 93 18 12 21 15 13 95 18 14 97 18 15 69 16 = 241 Average demand = 241/15 = 16.1 cases per week