SMS Module 1 Flashcards
Simulation Definition
Simulation is the imitation of a real world process or system over time. It helps us to study the behavior of a system as it evolves.
It can be done either by hand or using a computer using the mathematical and symbolic relationships between entities of the system.
Appropriate Use Cases For Simulation
- To study and experiment on complex systems and understand their internal interactions.
- It can be used to experiment on systems with new policies or designs prior to implementation.
- It can be used to verify analytical solutions.
- Knowledge gained from simulation can be great value towards improving the system under investigation.
- Valuable insight can be gained about which variables are important and how they interact from simulations.
Inappropriate Use Cases For Simulation
- When a problem can be easily solved using common sense.
- When a problem can be easily solved using analytical solutions.
- It’s easier to experiments in real life.
- Development cost of the model exceeds that of performing experiments.
- If there is less or no data is available.
- If the system behavior is too complex or cannot be defined.
Advantages Of Simulation
- New policies, designs, decisions etc can be explored and tested without disrupting the ongoing real system.
- New designs and systems can be tested without committing actual or physical resources.
- Insight can be gained about which variables are important to the performance of the system.
- Insight can be gained about various interactions between variables in the system.
- Hypothesis about how or why phenomena occur can be tested for feasibility.
- Bottleneck analysis can be performed to gain insights about how, why and where processes are delayed.
Disadvantages Of Simulation
- Model building requires special training, it is an art learned over time and through experience.
- Simulation modeling and analysis can be time consuming and expensive.
- Two models constructed by two competent individuals may have similarities but will never be the same.
- Results obtained from simulation maybe hard to interpret, it is hard to determine if observation is result of interrelationships in the system or pure randomness.
Steps In A Simulation Study
- Problem Formulation
- Setting of objective and overall project plan
- Model conceptualization
- Data collection
- Model translation
- Verification
- Validation
- Experimental Design
- Production runs and analysis
- More runs
- Documentation and reporting
- Implementation
Refer Simulation Study Diagram
Step 1 : Problem Formulation
- Every study should begin with a problem statement.
2. The analyst and policy makers both must clearly understand the problem statement and agree with it’s formulation.
Step 2 : Setting Of Objective And Overall Project Plan
- The objective indicates the questions to be answered by simulation.
- The overall project plan must include :
- 1 Statement of alternative systems
- 2 Method for evaluating effectiveness of alternatives
- 3 Cost of study
- 4 The number of days or time required and also the manpower or number of people involved in the study
Step 3 : Model Conceptualization
- Construction of a model is as much art as it is science.
- Modeling is enhanced by it’s ability to :
- 1 Abstract essential features of a problem
- 2 Select and modify basic assumptions that characterize the system
- 3 Enrich and elaborate the model until useful results are obtained
Step 4 : Data Collection
There is constant interplay between construction of a model and collection of necessary input data.
Step 5 : Model Translation
- Real world system result in complex models that require a lot of computation and storage.
- They can be programmed using simulation languages or special purpose software.
Step 6 : Verification
- The simulation program or model and it’s performance needs to be verified after development.
- It is verified only if the input parameters and logical structure are correctly represented.
Step 7 : Validation
- A model is validated to determine that a model is an accurate representation of the real system.
- It is achieved through calibration of the model, which is an iterative process of comparing the model to actual system behavior and identifying differences between the two.
Step 8 : Experimental Design
- The alternative designs that are to be simulated must be determined.
- For each experimental design, decisions need to be made such as :
- 1 Length of initialization period
- 2 Length of simulation runs
- 3 No. of replications to be made of each run
Step 9 : Production Runs And Analysis
They are used to estimate measures of performance for the system designs being simulated.
Step 10 : More Runs
Analyst determines if additional runs are required and what additional experiments that need to be performed based of past runs.
Step 11 : Documentation And Reporting
Two types of documentation
- Program documentation : Used by analysts to understand how the model/program operated.
- Process documentation : History of the simulation project, analysis results, alternatives, experiment results etc
Step 12 : Implementation
- Success of the implementation depends on previous steps.
- If the model user has been thoroughly involved in the process, the likelihood of a good implementation will have been increased.
Entity
An entity is an object of interest in a system.
Ex : Departments, orders, parts etc are entities in a factory system.
Attribute
An attribute denotes the property of an entity.
Ex : Quantities of each order, number of machines in each department.
Activity
Activities represent a time period of specified length.
Ex : Manufacturing process, service time
System State
It is defines as a collection of variables necessary to describe a system at any time, relative to the objective of study.
Ex : Busy or idle state of server
Event
An event is defined as an instantaneous occurrence that may change state of system.
- Endogenous : Activities and events occurring within a system.
- Exogenous : Activities and events occurring in an environment that affects the system.
Event Notice
A record of an event to occur at current time or future along with necessary data to execute the event.
Ex : Arrival of customer at clock 8 with customer number as data.
Delay
Duration of time of unknown length. The duration is known only when it ends.
Ex : Customer waiting in line for service to begin.
FEL
Future Event List - list of event notices for future events ordered by time of occurrence.
Clock
A variable representing simulated time.
System
A collection of entities that interact together over time.
Ex : Bank, grocery store
Average Waiting Time
Total time customers wait in queue / Total no. of customers
Probability Of Waiting
No. of customers who wait / Total no. of customers
Average Service Time
Total service time / Total no. of customers
Probability Of Idle Server
Total idle time of server / Total simulation time
Average Time Between Arrival s
Sum of all time between arrivals / No. of arrivals - 1
Average Waiting Time For Those Who Wait In Queue
Total time customers wait in queue / Total no. of customers who wait
Average Time Customer Spends In System
Total time customers spend in system / Total no. of customers
