Theo Simu
Theo Simu
Four phases:
-Problem formulation, and
setting objective and overall
design (important)
(step 1 to 2).
-Modeling building and data
collection (step 3 to 7)
-Running of the model
(step 8 to 10).
-Implementation (step 11 to
12).
An iterative process.
Explain why it is important
it is the first step in the study,
we must understand the
problem that we're simulating,
if we understand it in the
wrong way, we simulate
wrong and the model is
invalid
Question: How to simulate random events that we cannot come up with the probability like in the given
examples?
Simulating random events that we cannot come up with the probability for can be challenging, but there are
several approaches that can be used.
One approach is to use a simulation technique called Monte Carlo simulation. This involves generating a large
number of random samples from a distribution that approximates the unknown probability distribution. The
simulated outcomes can then be used to estimate the probabilities and statistics of interest.
Another approach is to use empirical data to estimate the probability distribution. If historical data or
experimental data is available, it can be used to estimate the probability distribution and simulate future events
based
"Discrete" refers to a simulation model where events occur at distinct points in time. "Stochastic" means that the
model incorporates randomness or probability into the system being simulated. "Dynamic" indicates that the
model changes over time based on the outcomes of previous events and actions. Together, these terms describe
a simulation modeling approach that involves modeling complex systems as a sequence of discrete events with
probabilistic outcomes that evolve over time.
1. Entities: Objects or items within the system being modeled that move through the system and change
states as they interact with other entities or resources.
2. Events: Occurrences that cause changes in the system state, such as the arrival of an entity or the
completion of a process.
3. Queues: Areas where entities wait for access to resources or events to occur before they can proceed.
4. Resources: Items that entities need to complete their processes or actions, such as machines, workers, or
materials.
5. Time: Simulation models typically involve the passage of time, often represented in discrete increments,
such as minutes or hours.
6. Statistics: Metrics used to evaluate the performance of the system being modeled, such as throughput,
waiting times, and utilization rates.
Together, these concepts form the basis for building a discrete-event simulation model that can help answer
questions about how a system behaves under different scenarios and conditions.
Concept in Discrete-Event Simulation:
● Entities:
- dynamic objects in the simulation, which move around in the system, affect the state of the system
- most entities represent real things in a simulation, but you also can create “fake” ones
● Attributes:
- a common characteristic of all entities, but with a specific value that can differ from one entity to
another
- attribute values are tied to specific entities
Activities:
- A duration of time of specified length, which is known when it begins (although it may be specified
deterministic or in statistical distribution)
- This duration is characterized and defined by the modeler
Events: An instantaneous occurrence that changes the state of a system
What are the system state variables in the Single-Channel Queue?
The Single-Channel Queue (also known as the M/M/1 queue) is a common model in queuing theory and
discrete-event simulation. The system state variables for this model are:
1. Number of customers in the system: This variable tracks the number of customers currently in the
system, including those who are waiting in the queue and those being served by the single server.
2. Server status: This variable indicates whether the server is idle or busy serving a customer.
3. Queue length: This variable measures the number of customers waiting in the queue for service.
4. Arrival times: This variable records the time each customer arrives to the system.
5. Service times: This variable measures the time it takes the server to complete service for each customer.
These variables are dynamic and change over time as customers arrive, join the queue, are served by the server,
and exit the system. Analyzing these variables can help to evaluate system performance and identify areas for
improvement.
- Write down what time it is now on one of your attributes (for other computations later)
- Stay with the server for a duration that the service requires
1. Event-scheduling: In this approach, the simulation model maintains a list of future events and their
scheduled times. The system state is updated only when an event occurs, and any subsequent events that
depend on the updated state are added to the event list. This approach can be more efficient than
continuously updating the system state, as it avoids unnecessary computations.
2. Process interaction: In this approach, the simulation model explicitly models the interaction between
different processes or entities in the system being simulated. Each process may have its own logic for
making decisions and taking actions based on the current state of the system. The system state is
updated continuously as each process interacts with others or with external events.( ARENA)
Or
Both approaches have their strengths and weaknesses, and the choice of which one to use depends on the
nature of the system being modeled and the goals of the simulation study. Event-scheduling is often used for
systems where events occur infrequently and there are long periods of idle time, while process interaction is
more suitable for systems with frequent interactions between entities or where the behavior of individual entities
is important.
Bonus:
Random numbers and random variates are both used in simulation modeling to introduce stochasticity or
randomness into the system being modeled. However, there is a difference between these two concepts:
1. Random numbers: These are numbers generated by a random number generator (RNG), which typically
produce a sequence of uniformly distributed values between 0 and 1. Random numbers can be used as
building blocks for generating other types of random variables.
2. Random variates: These are values generated from a specific probability distribution, such as the normal
distribution or the exponential distribution. Random variates are derived from random numbers by
applying an appropriate transformation that maps the uniform distribution onto the desired distribution.
In other words, random numbers are raw inputs used to generate random variates, which in turn are used to
model the behavior of random variables in the system being simulated.
Give some examples of random OUPUTS used to evaluate a queuing system?
In a queuing system, random outputs are used to measure the performance of the system and evaluate the
effectiveness of different queue management policies. Some examples of random outputs used in queuing
system evaluation include:
1. Waiting time: The amount of time a customer spends waiting in the queue before being served.
2. Queue length: The number of customers waiting in the queue at any given time.
3. Service time: The amount of time it takes for a server to serve a customer.
4. Utilization: The percentage of time that a server is busy serving customers.
5. Arrival rate: The rate at which customers arrive at the queuing system.
6. Abandonment rate: The percentage of customers who leave the queue without being served.
7. System throughput: The total number of customers served by the system over a given time period.
8. Idle time: The amount of time that the server is idle and not serving customers.
By analyzing these random outputs, simulation modeling can help identify bottlenecks in the queuing system,
assess the impact of different queue management policies, and optimize the system to improve customer service
levels and reduce wait times.
Simulation modeling can be used to evaluate different inventory management policies and strategies. The
following are some examples of random outputs that might be used to evaluate the performance of an inventory
management system:
1. Inventory level: This is the amount of inventory on hand at any given time. Random fluctuations in
demand, lead times, and other factors can cause inventory levels to vary over time.
2. Stockouts: This occurs when inventory levels drop below a certain threshold and can result in lost sales or
production delays. The occurrence and duration of stockouts can be modeled using random variates.
3. Reorder point: This is the inventory level at which a replenishment order is placed. The optimal reorder
point depends on the uncertainty in demand, lead times, and other factors, which can be modeled using
random inputs.
4. Order quantity: This is the amount ordered when a replenishment order is placed. The optimal order
quantity depends on the costs associated with ordering and holding inventory, as well as the uncertainty
in demand and lead times.
5. Total cost: This includes the costs of ordering, holding, and potentially backordering or expediting
inventory. The total cost can be computed for different inventory management policies and evaluated
using random inputs to assess robustness and sensitivity to different scenarios.
By evaluating these random outputs, simulation modeling can help identify the strengths and weaknesses of
different inventory management policies and strategies and inform decision-making in real-world systems.