SIMULATION SYSTEM REPORT

Case Study:
BUBUK KOPI DAS, PADANG

BY:
Group 1
Arfan Maulana 1610931005
AlifaSalsabila Putri 1610931026
Gilang Farhan Anugrah 1610932007
BimaSyiqta Kurniawan 1610932025

Lecturer:
Ir. ElitaAmrina, Ph. D, IPM

DEPARTMENT OF INDUSTRIAL ENGINEERING

FACULTY OF ENGINEERING
ANDALAS UNIVERSITY
2019
 TABLE OF CONTENTS

TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
SUMMARY

CHAPTER I INTRODUCTION
1.1 Background
1.2 Problem Formulation
1.3 Objectives
1.4 Scopes of the Study
1.5 Assumptions

CHAPTER II THEORITICAL FRAMEWORK

2.1 Production Process
2.2 System Modeling
2.3 Simulation Modeling
2.4 Random Variable and Probability Distribution
2.5 Statistical Testing
2.6 Verification and Validation

CHAPTER III METHODOLOGY

3.1 Location and Time of Study
3.2 Data Collection
3.3 Study Stages

CHAPTER IV MODEL DEVELOPMENT

4.1 System Description
4.2 Determining Probability Distribution
4.3 Development ARENA Model
4.4 Model Verification
 4.5 Model Validation

CHAPTER V SYSTEM IMPROVEMENTS AND DISCUSSIONS

5.1 System Improvements
5.2 Discussions

CHAPTER VI CONCLUSIONS
6.1 Conclusions
6.2 Suggestions

REFERENCES

APPENDICES
 LIST OF TABLES

Table 2.1 Types of Probability Distribution

Table 2.2 Common Statistical Test
Table 5.1 Comparison of Two Scenarios
 LIST OF FIGURES

Figure 3.1 Research Methodology Flowchart

Figure 4.1 Flowchart of Production System ini Bubuk Kopi Das
Figure 4.2 Mixing Process Probability Distribution
Figure 4.3 Roasting Process 1 Probability Distribution
Figure 4.4 Roasting Process 2 Probability Distribution
Figure 4.5 ARENA Model
Figure 4.6 Model Verification
Figure 4.7 Model Validation
 CHAPTER I
INTRODUCTION

This chapter consists of background, problem formulation, objectives,

scopes of the study, and the assumptions of the simulation system project.

1.1 Background

In general, the aim of a manufacturing industry is to produce goods

economically so that they can make a profit and can deliver products on time. In
addition, the manufacturing industry also wants the production process to
continue and develop so that the company's survival is guaranteed. Nowadays
companies are also demanded to be more competitive so that they are able to
compete for existing markets. One step to realize this is through the development
of operational and processing systems by eliminating unnecessary stages of
operation.

Changes in lifestyle and people's behavior in daily life, increasingly affect

the community's interest in a variety of food industry products including coffee
drinks, at this time coffee drinks are no longer drinks that are only consumed by
parents, seen by the widespread expansion of cafes and coffee shops nowadays.
Coffee in Indonesian society has become a lifestyle that is consumed in various
opportunities both for young people and parents with all social levels of society.
So that the coffee industry is one industry that is quite promising. Coffee can be
obtained in various places such as in stalls and in supermarkets. Types of coffee
are made various kinds, based on the typical coffee beans from each region
throughout Indonesia.

The coffee industry is actually classified as small and medium industries

or home industries, but the industry can be developed into a labor-intensive
industry that is able to increase the income of employers and the surrounding
workforce. The coffee industry basically only requires skills in processing coffee
beans. Nowadays, the coffee industry in Padang City has circulated many types of
local coffee brands. One of them is Bubuk Kopi Das.

Bubuk Kopi Das is a manufacturing industry that produces ready-to-sell

coffee powder. The manufacturing process of this industry consists of drying
process, roasting process, mixing process, grinding process, and packaging
process. Some processes in this manufacturing industry require quite a long time,
so to produce several packs of coffee powder takes around three days. For
example, grinding process and cooling process after mixing and grinding require 8
hours each. So there are queues in several production processes which would
disrupt other processes that will have an impact on the company's income.

One method of analysis that can be used to identify problems in

production facilities is to use a simulation approach. Choi et al. (2002) uses
simulations to identify bottlenecks and evaluate machine performance and
production data for more efficient production supervision. From the simulation
results, the area that becomes the bottleneck in the system and the optimum
number of assembly machines must be added to the area. In addition, from the
simulation it is also known that the low utilization rate of personnel at several
work stations so it is recommended that the production system be made automatic
or semi-automatic.

Basically, improvements made to a subsystem will greatly affect the

overall systems, moreover these improvements are made in a production process
where the processes are interrelated. With the simulation approach, the overall
condition of the system will be known, so that the information obtained can be
used as an indicator to determine the policy to be taken.

Therefore, it is necessary to do system modeling in the Bubuk Kopi Das

manufacturing industry in order to be able to carry out an effective and efficient
production process.
1.2 Problem Formulation

Based on the background above, the problem formulation of this project is

how to design a simulation system model for the Bubuk Kopi Das industry so that
the production process can run effectively and efficiently.

1.3 Objectives

The purpose of this simulation system project is to design a simulation

system model in Bubuk Kopi Das company which is valid and can be
implemented to the company.

1.4 Scopes of the Study

Scopes of problems in this study include:

1. Data collection is carried out for 22 working days.
2. Solution scenarios produced only as a suggestion to improve
manufacturing productivity, while the implementation of the solution
depends on the company’s policy.

1.5 Assumptions

The assumptions that used in this research are:

1. The same packaging type for all coffee powders produced.
2. The number of machines used in the model matches the number of
machines commonly used during production.
 CHAPTER II

THEORITICAL FRAMEWORK

This chapter consists of related literatures of simulation project that consist

of system modeling, simulation modeling, probability distribution, statistical
testing, and verification and validation of the model.

2.1 Production Process

The production process is the process of transformation or a set of

activities that transform a collection of inputs (human resources, materials,
energy, information, etc.) into output products (finished products or services) that
have added value. Manufacturing systems issue processes from raw materials to
finished products through operations. These operations contain a combination of
equipment and varying degrees of automation. In a discrete manufacturing
system, items from objects that are run through a series of activity processes,
buffer queues, storage areas to finished products manufacture. (Groover, 2000).

The manufacturing process can be divided into two main types of

processes: processing operations and assembly operations. Operation of the
process of changing work materials from one form into another that is part or
product, while the operation of discussing two or more components into parts or
products. (Groover, 2000).

2.2 System Modeling

System modeling is the process of developing abstract models of a system,

with each model presenting a different view or perspective of that system. System
modeling has generally come to mean representing the system using some kind of
graphical notation, which is now almost always based on notations in the Unified
Modeling Language (UML). However, it is also possible to develop formal
(mathematical) models of a system, usually as a detailed system specification.
(Sommerville, 2010).

Models are used during the requirements engineering process to help

derive the requirements for a system, during the design process to describe the
system to engineers implementing the system and after implementation to
document the system’s structure and operation. It may develop models of both the
existing system and the system to be developed: (Sommerville, 2010)
1. Models of the existing system are used during requirements engineering.
They help clarify what the existing system does and can be used as a basis
for discussing its strengths and weaknesses. These then lead to
requirements for the new system.
2. Models of the new system are used during requirements engineering to
help explain the proposed requirements to other system stakeholders.
Engineers use these models to discuss design proposals and to document
the system for implementation. In a model-driven engineering process, it is
possible to generate a complete or partial system implementation from the
system model.

The most important aspect of a system model is that it leaves out detail. A
model is an abstraction of the system being studied rather than an alternative
representation of that system. Ideally, a representation of a system should
maintain all the information about the entity being represented but unfortunately,
the real world (also known as the universe of discourse) is utterly complex so
weed to simplify. An abstraction consciously simplifies and picks out the most
evident characteristics. It may develop different models to represent the system
from different perspectives. For example:
1. An external perspective, where you model the context or environment of
the system.
2. An interaction perspective where you model the interactions between a
system and its environment or between the components of a system.
3. A structural perspective, where you model the organization of a system or
the structure of the data that is processed by the system.
4. A behavioral perspective, where you model the dynamic behavior of the
system and how it responds to events.
(Sommerville, 2010).

2.3 Simulation Modeling

The model is defined as a representation of a quantitative qualitative

system that represents a process or event, which can clearly illustrate the
interaction between various important factors that will be observed. The model
was developed to conduct development investigations that allow it to be applied to
real systems or to determine the effect of different policies. The purpose of many
studies of systems is to predict how the system will work before the system is
built. As an alternative, prototypes are sometimes built for testing, but this is very
expensive and time-consuming. Even with existing systems, it is impossible or
practical to experiment with real systems. So the study of systems is usually done
with a system model. The model is not only a substitute for the system, but also a
simplification of the system.

Simulation is the process of planning a model of a real system and

experimenting with that model with the aim of understanding the behavior of the
system or evaluating various strategies to operate the intended system. In some
cases, it is important to observe a system to try to get a picture of the relationships
between various components, or to estimate performance under new conditions
under consideration.

Simulation models can be divided into (Law and Kelton, 2000):

1. Static or dynamic
Static simulation model, represents a system at a certain time. One
of the most common types of static simulation uses random numbers to
solve problems, usually stochastic, and the passage of time has no role.
 Dynamic simulation models, represent a system that changes with
time, for example simulations of CNC machines that work 40 hours per
week.
2. Deterministic or stochastic
The deterministic simulation model assumes no variability in the
model parameters and, therefore, does not involve random variables. If a
deterministic model is run on the same input value, it will always produce
the same value. The output of once running a deterministic simulation
model is the real value of the performance of the model.
Stochastic simulation model, containing one or several random
variables to explain the processes in the observed system. The output of
the stochastic simulation model is random and therefore only estimates of
the actual characteristics of the model. Then, it takes several runs of the
model, and the results are only an estimate of the expected performance of
the model or system being observed.
3. Continuous or discrete
In a continuous simulation model, variable conditions change
continuously, for example, fluid flow in a pipe, or flight of an aircraft,
variable position and speed conditions change continuously with each
other.
In a discrete simulation model, variable conditions change only at a
number of points (certain, which can be calculated) in time. Most of the
manufacturing systems are modeled as dynamic, discrete, stochastic event
simulations and use random variables to model the range of arrivals,
queues, processes, etc.

The elements of a simulation model are as follows.

1. Entities, these are system elements that are simulated and can be identified
and processed individually, for example machines in factories, vehicles,
people or anything that changes the state during simulation time.
Interactions between these entities shape system behavior.
2. Attributes, each entity will have one or more attributes that carry
additional information about the entity. These attributes have uses which
include dividing entities into classes, for example car manufacturing which
is distinguished by color in the final assembly sequence. In this case the
color is an attribute of the car body. Other uses are controlling entity
behavior, for example the priority attribute used to establish queuing
discipline.
3. Variables, reflects a characteristic of the whole model, not of specific
entities.
4. Resources, a resource can have several units of capacity, example seats at
a table in a restaurant.
5. Queues, place for entities to wait when they can’t move on (maybe since
the resource they want to seize is not available)
6. Statistical accumulators, variables that “watch” what’s happening, depend
on output performance measures desired.

2.4 Random Variable and Probability Distribution

A random variable is a numerical description of the outcome of a

statistical experiment. A random variable that may assume only a finite number or
an infinite sequence of values is said to be discrete; one that may assume any
value in some interval on the real number line is said to be continuous. For
instance, a random variable representing the number of automobiles sold at a
particular dealership on one day would be discrete, while a random variable
representing the weight of a person in kilograms (or pounds) would be
continuous. (Anderson, 2019).

The probability distribution for a random variable describes how the

probabilities are distributed over the values of the random variable. For a discrete
random variable, x, the probability distribution is defined by a probability mass
function, denoted by f(x). This function provides the probability for each value of
the random variable. In the development of the probability function for a discrete
random variable, two conditions must be satisfied: (1) f(x) must be nonnegative
for each value of the random variable, and (2) the sum of the probabilities for each
value of the random variable must equal one. (Anderson, 2019).

A continuous random variable may assume any value in an interval on the

real number line or in a collection of intervals. Since there is an infinite number of
values in any interval, it is not meaningful to talk about the probability that the
random variable will take on a specific value; instead, the probability that a
continuous random variable will lie within a given interval is considered.
(Anderson, 2019).

In the continuous case, the counterpart of the probability mass function is

the probability density function, also denoted by f(x). For a continuous random
variable, the probability density function provides the height or value of the
function at any particular value of x; it does not directly give the probability of the
random variable taking on a specific value. However, the area under
the graph of f(x) corresponding to some interval, obtained by computing
the integral of f(x) over that interval, provides the probability that the variable will
take on a value within that interval. A probability density function must satisfy
two requirements: (1) f(x) must be nonnegative for each value of the random
variable, and (2) the integral over all values of the random variable must equal
one. (Anderson, 2019).

The probability distribution that commonly used in simulation as follows.

(Vose Software, 2019).
Table 2.1 Types of Probability Distribution
Type of
Definition Graph
Distribution
Because of its ability to take on a
wide variety of shapes, this
distribution is often used as a
rough model in the absence of
Beta
data. The beta is often used to
represent random proportions
such as the proportion of defective
items in a lot.
This distribution is used in
situations in which an activity
occurs in successive phases and
Erlang each phase has an exponential
distribution. It is often used to
represent the time required to
complete a task.
This distribution is often used to
model inter event times in random
Exponential arrival and break-down processes,
but it is generally inappropriate
for modeling process delay times.
This is often used to represent the
time required to complete some
Gamma
task (e.g., a machining time or
machine repair time).

The flexibility of the Johnson

Johnson distribution allows it to fit many
data sets.
This is used in situations in which
the quantity is the product of a
Lognormal
large number of random
quantities.
This distribution is used in
situations in which the central
Normal limit theorem applies, i.e.,
quantities that are sums of other
quantities.
This distribution is a discrete
distribution that is often used to
Poisson model the number of random
events occurring in a fixed
interval of time.
 It is commonly used in situations
in which the exact form of the
distribution in not know, but
Triangular
estimates (or guesses) for the
minimum, maximum and most
likely are available.

This distribution is used when all

Uniform values over a finite range are
considered to be equally likely.

The weibull distribution is widely

Weibull used in reliability models to
represent the lifetime of a device.

2.5 Statistical Testing

After looking at the distribution of data and perhaps conducting some

descriptive statistics to find out the mean, median, or mode, it is time to make
some inferences about the data. Inferential statistics are the set of statistical tests
researchers use to make inferences about data. These statistical tests allow
researchers to make inferences because they can show whether an observed
pattern is due to intervention or chance. There is a wide range of statistical tests.
The decision of which statistical test to use depends on the research design, the
distribution of the data, and the type of variable. In general, if the data is normally
distributed, parametric tests should be used. If the data is non-normal, non-
parametric tests should be used. Below is a list of just a few common statistical
tests and their uses. (CYFAR, 2019).
Table 2.2 Common Statistical Test (CYFAR, 2019)
Type of Test Use
Correlation: these tests look for an association between variables
Pearson Correlation Tests for the strength of the association between two continuous variables

Tests for the strength of the association between two ordinal variables (does not rely on
Spearman Correlation
the assumption of normally distributed data)

Chi-Square Tests for the strength of the association between two categorical variables
Comparison of Means: these tests look for the difference between the means of variables
Tests for the difference between two variables from the same population (e.g., a pre-
Paired T-Test
and posttest score)
Tests for the difference between the same variable from different populations (e.g.,
Independent T-Test
comparing boys to girls)
Tests for the difference between group means after any other variance in the outcome
ANOVA
variable is accounted for (e.g., controlling for sex, income, or age)
Regression: these tests assess if change in one variable predicts change in another variable
Tests how change in the predictor variable predicts the level of change in the outcome
Simple Regression
variable
Tests how changes in the combination of two or more predictor variables predict the
Multiple Regression
level of change in the outcome variable
Non-Parametric: these tests are used when the data does not meet the assumptions required for parametric tests
Tests for the difference between two independent variables; takes into account
Wilcoxon Rank-Sum Test
magnitude and direction of difference
Tests for the difference between two related variables; takes into account the
Wilcoxon Sign-Rank Test
magnitude and direction of difference
Tests if two related variables are different; ignores the magnitude of change—only
Sign Test
takes into account direction

2.6 Verification and Validation

Models created must pass verification and validation to guarantee the

correctness of the model and output (Sargent, 2011). Verification is an activity
that ensures the model is made correctly (Sargent, 2011). There are several
methods for verifying such as (Güneş, 2012):
1. Visually test by others.
2. Create a flow chart indicating the activity model when it is run.
3. Using the debugger.

According to Harrel (2004), there are several techniques for conducting

verification, including the following:
1. Re-examining the model, can be done bottom-up, which is doing a unit
check and process logic used in the model.
2. Check the output produced in each process in the model using trace. There
are 3 types of trace settings:
a. Trace off, is a list of events that are not displayed during the simulation
system.
b. Step, is a way of making a list of events appear for each step and
usually the time displayed is the length of time.
c. Continuous, is a way to make a list of events appear continuously
according to the time used in the simulation model.
3. Observe the animation of the model being run, whether the behavior of the
system is in accordance with the desired model.
4. Compile errors or debugging on the simulation model.

Validation is a measurement of an activity, a plan, or a transaction to

prove that the measured matter is correct, complete, applied as desired, and
deliver the results as desired (Web Finance, Inc., 2017). Validation is the activity
of comparing the model and the nature of the model to the real system, some
methods that can bedone for validation i.e. (Güneş, 2012):
1. High face validity, that is to ensure the visual model represents the
realsystem when the model is run.
2. Validate model assumptions, i.e. test the working stage of the model
againstthe real system and the reliability of data along with statistical
analysis,
3. Validate input-output transformation, i.e. using some of the historical
inputdata to compare to the data used to create real system models and
outputs.

According to Harrel (2004), validation techniques that can be used are as

follows:
1. Observing animation on the model that is run, comparing behavior on the
model with behavior in the real system according to the knowledge of
others about the system.
2. Compare models with real systems by running real models and systems in
the same conditions.
3. Do a comparison between the output model and the output on the real
system.
4. Perform sensitivity analysis, namely by making changes to the input
values to determine the effect on the behavior that occurs in the system or
system output.
 CHAPTER III
METHODOLOGY

This chapter consists of the identity of the project study, data collection
from the company, and the study stages to achieve objectives of the project study.

3.1 Location and Time of Study

The location of Bubuk Kopi Das company is located on Jl. ParakAnau

Raya No.21, Tabing. This coffee powder industry was founded in 1970 by Mr.
Das (Alm.). At present, Bubuk Kopi Das company is run by Mr. Algi. The
process of collecting data related to this project was carried out for 22 days.

3.2 Data Collection

There are four cycle in each day in this company. Every process has arrival
time, starting process time, process time, and finishing. For roasting process,
grinding process, and packaging process are started in the beginning of the day,
because these process takes a long time. Mixing process and set a side process of
the coffee beans and coffee powder, the started time begin in every finished
previous process. In the roasting process, the amount of coffee beans decrease
about 22% of all the amount, so it does the grinding process that decrease about
5% in every process.

The raw data that have been taken from the inventory, are the real data
taken from the survey in the company. the coffee beans arrives in the factory
every week with amount of coffee beans of 4 - 5 tons. This company have 5 days
works with worker time from 08.00 until 17.00.

The data that have been collected for 22 days, are tabulated that shown in
Appendices.
3.3 Study Stages

This point explanation of the stages are carried out in the research,
methods, and techniques used in this study, including data collection, the object of
research, and application of theory in this research so as to facilitate in
determining the method used to solve the problems in this study.

1. Preliminary Study

Preliminary studies are an early stage in conducting research to find out

the scope of the problems to be researched. It can be done with field studies that
are conducted to obtain information related to the work system, management, and
production of the company as well as problems supporting the research. Field
study conducted by survey, interview and direct discussion with the Manager of
Bubuk Kopi Das, Mr. Algi and with some of the workers in the company. Based
on the results of the interviews and discussions that have been done, it is known
that in the production process Bubuk Kopi Das there are some queuing process, so
it needs to be designed a simulation modeling as a tool to improve the
effectiveness of the production processes.

2. Literature Study

In this study, literature study conducted to obtain concepts that support

problem solving in research by collecting theories related to the simulation
modeling, probability distribution, etc.

3. Problem Formulation

The problem formulation of this project is how to design a simulation

system model for the Bubuk Kopi Das industry so that the production process can
run effectively and efficiently.
4. Objectives

The objective of this simulation system project is to design a simulation

system model in Bubuk Kopi Das company which is valid and can be
implemented to the company.

5. Data Collection

The collection of data consist of several stages that is identification of

waste and identification of solution scenarios. The data collected consists of
primary and secondary data, where:
1. Primary data is data obtained from observations and research directly in
the field. Primary data in the form of observation and questionnaire data.
a. Observation
The primary data collection is done by observing directly at the
research site and requesting information and interviewing operators or
technicians directly involved operationally.
b. Questionnaire
The use of a questionnaire is a way of collecting data using a list of
questions or a list of fields of the researched object.
2. Secondary data is data that is not directly observed by researchers. This
Data is the company's documentation.

6. Data Processing

Data collected, then processed to be used in research. The stages of data

processing carried out in the study are system modeling, simulation modeling,
statistical testing, verification and validation, improve a scenario, and develop the
simulation model.
7. Development of Simulation Model

In this stage is describe and explain the improvement scenarios of

simulation project that may be implemented to the company.

8. Discussions

Discussion is carried out to determine whether the scenario developed in

the simulation model is feasible to be applied in company conditions.

9. Conclusions

Contains the conclusions obtained from research related to increasing the

company's business operations based on the objectives, as well as suggestions for
companies or next researcher. Flowchart methodology of the study is shown in
Figure 3.1.
 Start

Preliminary Study
It can be done with field studies that are conducted to obtain
information related to the work system, management, and production
of the company as well as problems supporting the research.

Literature Study
Conducted to obtain concepts that support problem solving in
research by collecting theories related to the simulation modeling,
probability distribution, etc.

Problem Formulation
How to design a simulation system model for the Bubuk Kopi Das
industry so that the production process can run effectively and
efficiently.

Objective
To design a simulation system model in Bubuk Kopi Das company
which is valid and can be implemented to the company.

Data Collection

Secondary Data
Primary Data
Data that is not directly
Data obtained from
observed by researchers.
observations and research
This Data is the company's
directly in the field.
documentation.

Data Processing
The stages of data processing carried out in the study are system
modeling, simulation modeling, statistical testing, verification and
validation, improve a scenario, and develop the simulation model.

Development of Simulation Model

In this stage is describe and explain the improvement scenarios of
simulation project that may be implemented to the company.

Discussions
Discussion is carried out to determine whether the scenario
developed in the simulation model is feasible to be applied in
company conditions.

Conclusions
Contains the conclusions obtained from research related to increasing
the company's business operations based on the objectives, as well as
suggestions for companies or next researcher.

End

Figure 3.1 Research Methodology Flowchart

 CHAPTER IV
MODEL DEVELOPMENT

This chapter consists of the processing of the model will be developing,

such as system description, determining the probability distribution, development
ARENA model, and verification and validation of the model.

4.1 System Description

The production processes in Bubuk Kopi Das company will be develop in

a model that will be simulated later on. The description of the system is shown in
Figure 4.1.

Start

Coffee beans are dried in the sun.

Waiting
Weather Bad
(Queue)

Good

Roasting Process.

Mixing Process.

Set a side.

Grinding Process.

Smoothness Waiting
Rough
Level (Queue)

Smooth

Set a side.

Packaging Process.

End

Figure 4.1 Flowchart of Production System ini Bubuk Kopi Das

 The production process in Bubuk Kopi Das start when the coffee beans
come to the factory. It will be dried in the sun for 4 – 8 hours. If the weather are
not good enough to dried the coffee beans, it will be stored again to the storage
and dry it in the next day. But if the weather is good, the coffee beans will be store
to the sacks.

In roasting process, there are two machines that have capacity of each
machine is 80 kg. The process in this stage take time about 1,5 hours. After that,
the roasted coffee beans moves to the mixing process. This process only have one
machine that have capacity 400 kg. Coffee beans are dropped from the roasting
production floor through wind-fired funnels, so that they will be separated
between the beans and the charcoal shell.Coffee beans that have been mixed are
set a side for about 8 hours in sacks with a capacity of 50 kg.

The next process is grinding process. This is the process where the coffee
beans will be grind until it reaches the desired level of smoothness that take 8
hours process with eight machines that have 50 kg capacity of each machine.
Damage conditions that are often experienced by grinding machines are crushed
stone that wears quickly and is separated from each other. It causes the coffee
powder produced is still not smooth enough. Then this coffee powder will be put
back into the grinding machines. Then, if the coffee powder has been ground, it
will be left for 8 hours in a container which will then be carried by the crane to the
next production floor.

The last process in this factory is packaging process. This process uses a
total of 6 people to wrap coffee powder in packaging. On average each worker can
complete 60 kg per day. This finished product will be stored in a warehouse and
ready to be distributed.
4.2 Determining Probability Distribution

The distribution of the raw data drawn in Data Analyzer. Each of the
process has the probability distribution of each of them. For the mixing process,
roasting process 1 and 2, the probability distribution is Normal distribution, Beta
distribution, and Lognormal Distribution, respectively. The graph are shown in
the Figure 4.2, Figure 4.3, and Figure 4.4 below.

Figure 4.2 Mixing Process Probability Distribution

Figure 4.3 Roasting Process 1 Probability Distribution

 Figure 4.4 Roasting Process 2 Probability Distribution

4.3 Development ARENA Model

The model is developing in ARENA Software is based on the system

description from the survey as shown as Figure 4.5 below.

Figure 4.5 ARENA Model

4.4 Model Verification

Verification is carried out to guarantee that the simulation model is built in

the right way. In the model that has been built, verification is done by debugging
the simulation model and checking the report produced. When doing debugging,
the simulation model can run according to the desired model without error. It can
be seen in Figure 4.6.

Figure 4.6 Model Verification

4.5 Model Validation

In the model that has been built, validation is done by making a

comparison between output in the model existing and the real system from the
survey. The output data in the real system is 8280 for 22 days and the output in
the model is 12,946. We do the model validation using SPSS Software with One-
Sample Test and α = 95%. The output of SPSS are shown in Figure 4.7 below.

Figure 4.7 Model Validation

 The t value is -6.191, which gives us a p-value of .000. This is going to be
a significant result for any realistic alpha level. A standard alpha level is .05 and
.000 is smaller that .05, so we reject the null hypothesis which asserts there is no
difference between output of the real system and the model existing. Because the
output difference between both of them is 4,666 that declare there is a big
differences so the model of the system is valid enough.
 CHAPTER V
SYSTEM IMPROVEMENTS AND DISCUSSIONS

This chapter consist of the improvement of the system and the discussion
about the project.

5.1 System Improvements

The output report in Resources explain that grinding machine has average
utilization of 93.68% and the worker packaging has average utilization of 90.34%.
The limitation of the utilization that shown the process are good or bad is below
80%. It needs improvement in the system. So, the improvement that we built is
the adding of grinding machine, because this process takes a long time and will
affect the next process with the average queue time is 2.6438. Besides, we also do
another improvement when we adding grinding machine, we do additional
packaging workers, hire 6 people because the process of packaging has average
queue time 1.8842.

5.2 Discussions

After we develop the scenario to improve the production productivity,

when we only add the grinding machine, the average utilization of grinding
machine 1 is 70.45% and the grinding machine 2 is 47.59% with average queue
time for this scenario are 0.7494 and 0.2347. But, the problem in packaging
process still bad with the average utilization is 90.34 and the average queue time
is 3.7078. So, when we do the other improvement, which are add the grinding
machine and hire 6 packaging workers, the average utilization of both grinding
machine are 58.82% and 69.04% with average queue time are 0.2306 and 0.4729.
Besides, the worker packaging 1 utilization is 53,90% and worker packaging 2
utilization is 56,55% with average queue time are 0.4043 and 0.4273. The
comparison between these two scenarios are tabulated in Table 5.1 below.
Table 5.1 Comparison of Two Scenarios
Problem Queue in grinding process Queue in packaging process
Average Utilization 93.68% 90.34%
Average Queue
2.6438 1.8842
Time
Improvement Scenario 1
Action Add grinding machine
Average Utilization 70.45% (grinding machine 1);
90.34%
Scenario 1 47.59% (grinding machine 2)
Average Queue 0.7494 (grinding machine 1);
3.7078
Time Scenario 1 0.2347 (grinding machine 2)
Scenario 2
Add grinding machine and hire 6 packaging workers
Average Utilization 58.82% (grinding machine 1);
53.90% (worker 1); 56.55% (worker 2)
Scenario 2 69.04% (grinding machine 2)
Average Queue 0.2306 (grinding machine 1);
0.4043 (worker 1); 0.4273 (worker 2)
Time Scenario 2 0.4729 (grinding machine 2)
 CHAPTER VI
CONCLUSIONS

This chapter consist of conclusions and suggestion of this project.

6.1 Conclusions

The simulation model may be applicable to the company that give

suggestion the company adding the grinding machine and additional of packaging
workers to decrease the average utilization and average queue time of them. This
model has been verification and validate with the real data and the output of the
system. For the system improvement, we suggest to add grinding machine and
hire 6 more packaging workers.

6.2 Suggestions

The suggestions that can be suggest to the company or the next research as
follows.
1. The model of this simulation system of the company’s production process
may applicable to the company to improve manufacturing productivity.
2. The company may add the grinding machine to the manufacturing process
and do additional of packaging worker.
3. The next research can study a good simulation model so that it can be
applied to the company.
 REFERENCES

Anderson, David R., Dennis J. Sweeney., Thomas A. Williams., Random

variables and probability distributions, Encyclopædia Britannica, Inc.,
https://www.britannica.com/science/statistics/Random-variables-and-
probability-distributions (online accessed November 2019).

Choi, S. D., Anil R. Kumar, and AbdolazimHoushyar (2002), “A Simulation

Study of an Automotive Foundry Plat Manufacturing Engine Blocks.”
Proceedings of 2002 Winter Simulation Conference, pp 1035-1040.

CYFAR Web., (2019), Types of Statistical Test, University of Minnesota,

https://cyfar.org/types-statistical-tests (online accessed November 2019).

Groover, Mikell P. (2001), Automation, Production Systems, and CIM, 2nd

edition. Prentice Hall, New Jersey, USA.

Güneş, M., (2012), Chapter 10: Verification and Validation of Simulation Models,
FreieUniverstat Berlin.

Harrel, C., Ghosh, B., dan Bowden, R., (2004), Simulation Using ProModel,
SecondEdition, The McGraw-Hill, Inc., NY.

Law, Averill M. and W.D. Kelton (2000), Simulation Modelling and Analysis,
3rd edition, MC Graw-Hill, USA.

Sargent, R. (2011). Verification and Validation od Simulation Models.

Proceedings of the 2011 Winter Simulation Conference, IEEE.

Sommerville, Ian., (2016). Software Engineering, 10th ed., Chapter 5, USA:

Courier Westford.

Vose Software, (2019), Continuous Distribution, Model Risk Help,

http://www.vosesoftware.com/ModelRiskHelp/index.htm#Distributions/C
ontinuous_distributions/Erlang_distribution.htm (online accessed
November 2019).

WebFinance, Inc., (2017), Business Dictionary,

http://www.businessdictionary.com(online accessed November 2019).
APPENDIX