YAAR UNIVERSITY

FACULTY OF ENGINEERING
DEPARTMENT OF INDUSTRIAL ENGINEERING

GRADUATION PROJECT

PINAR SU PRODUCTION GROUP

PROJECT by
HALIL CANER KARAOGLU
DOGUKAN ANIL YILDIZ
CENGIZHAN AYDIN

PROJECT SUPERVISORS

OMER OZTURKOGLU

ADALET ONER

ZMR, 2014





i


Table of Contents
1. INTRODUCTION .................................................................................................................................. 1
2. GENERAL INFORMATION ABOUT PINAR SU ....................................................................... 1
3. MICRO SYSTEMS OF PINAR SU .................................................................................................. 4
3.1.1. Water of Display Parameters .......................... Error! Bookmark not defined.
3.1.2. Layout of Facility ............................................................................................. 6
3.1.3. Processes in Production Lines .......................................................................... 7
4. OBSERVATIONS AND SYMPTOMS ................................................................................................. 9
4.1. Analysis of Production Processes ............................................................................................ 9
4.2. SYMPTOMS ................................................................................................................................ 10
4.2.1. Analysis of Downtimes in Production Line 1 and Line 2 .............................. 12
5. PROBLEM DEFINITION, TARGET & CRITICAL SUCCESS FACTORS .................. 13
5.1. TARGET & CRITICAL SUCCESS FACTORS .............................................................. 13
6. LITERATURE REVIEW AND MODELING PERSPECTIVE ........................................... 14
6.1. Literature Review........................................................................................................................ 14
6.2. Modeling Perspective .............................................................................................................. 19
7. PLAN OF THE PROJECT ................................................... Error! Bookmark not defined.
8. RISK AND RISK AVOIDANCE ........................................... Error! Bookmark not defined.
9. CONCLUSION ............................................................................. Error! Bookmark not defined.



ii


List of Tables

Table 1 : Monthly Distribution Graph of Water Coming From Mountain ................. 4
Table 2 : Monthly Distribution Graph of Water Coming From Mountain for PC ... 5
Table 3 : Monthly Distribution Graph of Water Coming from Mountain for Pet
Bottle ................................................................................................................................................... 5
Table 4: Process Flow Diagram ................................................................................................. 8
Table 5: Ratios of Downtimes ................................................................................................. 12



































iii





List of Figures


Figure 1: Pinar Su Facilities in Turkey .................... Error! Bookmark not defined.
Figure 2: Market Share of Pinar Su ........................... Error! Bookmark not defined.
Figure 3: Pet and Polycarbonate sales ..................... Error! Bookmark not defined.
Figure 4: Exporting Sales in Pnar Su ....................... Error! Bookmark not defined.
Figure 5: Pet & Glass Bottle Segmentations and Polycarbonate.6
Figure 6: Layout of Pinar Su Aydn/Bozdoan Facility.7
Figure 7: Fishbone Diagram of Breakdown of Machines..11
Figure 8: Arena Simulation Model Draft..15






1



1. INTRODUCTION


Pnar Su San ve Tic A.S. is located in Bozdoan / AYDIN and their field of
activity is spring water bottling and bottle production. Pnar Su is a member of the
Yaar group which one of Turkeys biggest and most highly respected corporate
groups. Pinar Su established and started first non-recyclable packaged water
production of Turkey in 1984 in Izmir, Menderes. Annual production reached
100.000 tones after the 12 year production in Menderes so facility has moved to
Bozdoan facility, one of the top-notch waters in the world in terms of taste in 1996.
Therefore, there are 155 employees and 2 engineers are working in Pnar Su
Bozdoan.
2. GENERAL INFORMATION ABOUT PINAR SU

Sales and distribution network of dealers located all over Turkey. Pinar Su
has 3 facilities that located in Adapazar, Isparta and Aydn as shown as Figure 1.
Water which Pnar Su obtains from its Bozdoan, Gkeaa, Akaaa springs is
supplied to customers in Turkey and in the nearly thirty countries to which the
company exports. Mineral water supplied to the customer under the most natural and
hygienic conditions. Natural mineral water sourced from the Bozdoan, Gkeaa
and Akaaa springs supplied to market in all packaging formats. Bozdoan facility
is occupying a area of 64.000 m2 and its the biggest facility of Pinar Su. Annual
production capacities of Bozdoan, Gkeaa, Akaaa are 620.000, 607.000 and
210.000 tones respectively.








2


Figure 1 : Pinar Su Facilities in Turkey

Pinar Su is producing 3 main products as pet bottles, glass bottles and
polycarbonate in Turkey. Their pet segmentations are 0.33, 0.5, 1, 1.5, 5, 10 lt. Glass
bottle products are 0.33 and 0.75 lt. Polycarbonates are producing as only 19lt.
Increasing investments of multinational firms like Nestle, Danone, Coca Cola
is a sign for the growth of the market. Trend in the market to increase the number of
spring sources in order to optimize logistic cost. Pnar Sus competitive advantages
are superior quality standards, logistical strengths, a talent for keeping a close watch
on national and international customer trends and preferences and transforming them
into marketable products. In this market conditions, Pinar Su market share is %7 and
this market share is success against global companies. Small scale local producers
which occupying %48 as shown as in figure 2 are causing a fragmented market
structures.


3


Figure 2 : Market Share of Pinar Su

Pet bottle sales and Polycarbonate sales are biggest parts of market as shown
as figure 3. According to sales of pet and glass bottles, pet bottles occupying
approximately %33 of whole sales. Polycarbonate sales percent is approximately
%67.Pinar Su sales are increasing every year within market share in routine
conditions. Sales can possibly change according to lack of spring water amount. It
could be reason to ups and downs in sales.


Figure 3 : Pet and Polycarbonate Sales


Pnar Su was exported to Germany for the first time in 1985. As shown as
Figure 4, Germany has the biggest pie as % 46. England and other countries are other
important customers of Pnar Su.


4


Figure 4 : Exporting sales in Pnar Su


3. MICRO SYSTEMS OF PINAR SU

In Aydn-Bozdoan, there are totally 8 production lines, 6 of them producing
pet bottle and 2 of them producing glass bottle. Line 1,2,3 are very critical because of
0.5 and 0.33 pet bottle production. These two segmentations are occupying most of
all production. In order to reach customer demands, line 1&3 must be working in
every shift. Line 2 is back-up production line for line 1&3. Other lines are producing
other segments of pet bottle, glass bottles and 19L polycarbonates.



Table 1 : Monthly Distribution Graph of Water Coming From Mountain


0
5000
10000
15000
20000
25000
30000
Tonnes
Tonnes

5



Table 2 : Monthly Distribution Graph of Water Coming From Mountain for PC



Table 3 : Monthly Distribution Graph of Water Coming from Mountain for Pet Bottle


In these graphs we analyzed the amount of spring water. Therefore we
analyzed pet bottles and PC respectively. The amount of spring water is increasing
between April-October months due to weather conditions. In other times, the amount
of water is not makes a big difference. The interval of amount of water is 5000
tonnes and 14000 tonnes per year. The most important thing is in these graphs, the
0
2000
4000
6000
8000
10000
12000
14000
16000
PC Tonnes
PC Tonnes
0
2000
4000
6000
8000
10000
12000
14000
16000
PET Tonnes
PET Tonnes

6

amount of water make a big increase according to months. In these months every day
is so precious for company.


In Aydin Bozdogan facility, there are 6 pet bottle lines, 2 glass bottle lines
and 2 polycarbonate lines. These production lines are producing 9 different
segmentation products. These segmentations are 0.33lt, 0.5lt, 0.75lt, 1lt, 1.5lt, 5lt,
10lt and 19lt polycarbonate. First 5 lines are producing only pet bottles. Line 6 and 8
are producing glass bottles. Line 7 and 9 are producing 19L polycarbonates.
Segmentations of pet bottles are 0.33lt, 0.5lt, 1lt, 1.5lt, 5lt and 10lt. Glass bottle
segmentations are 0.33lt and 0.75lt. Best-seller segmentations of Pinar Su are
0.33lt&0.5lt pet bottles and 19L polycarbonates.



Figure 5 : Pet & Glass Bottle Segmentations and Polycarbonate

Line 1 has 50 tones production capacity per hour. Line 2 has 34 tones
production capacity for 0.5 lt segmentation and 16 tones production capacity for
0.33lt segmentation per hour. Line 3 has 89 tones capacity for 1.5lt production
capacity per hour and line 4 has 30 tones production capacity for 10lt per hour.


3.1.1. Layout of Facility

Total production and warehouse area is 64,000m2 in Bozdogan facility and
enclosed area is 17.000m2. 8 production lines are located side by side. The ninth line
is located behind of these lines and next to shipping dock. Production lines and
warehouse are nested. Line 1 and Line 2 are the longest lines of production lines.

7

Other lines lengths are equal each other. All lines are starting after inflating process
and finish with palletizing, stretching processes. Warehouse area starts from
palletizing and palletizing process is final process of production. Transportation
between lines is easy because all lines are parallel. This situation provides early
intervention ability to workers in production cases.

Lengths of production lines can change by workers. Current layout is optimal
layout according to facilitys workers. Length of line 1 & 2 is 11.526 cm. Length of
lines can decrease or increase by workers according to production needs.



Figure 6 : Layout of Pinar Su Aydn/Bozdoan Facility

3.1.2. Processes in Production Lines

Totally there are 13 stages to produce water in Bozdogan facility. Production
can be classified as 3 main processes in facility. First main process is transfer of
spring water, second main process is processing pet preforms (raw material) and
other main process is basic processes of water bottles. These 3 main processes are
common in every segment production.

8


Table 4: Process Flow Diagram


Spring water is transferring from mountain to facilitys total 600 tones water
tanks with pipelines every single day. Total water amount can change every day
according to spring water source. Weather conditions are only reason to affects to
spring water amount. Distance from mountain to facility is 8 km and it provides big
advantage to facility in order to making production. Quality of spring water is
another advantage of Pinar Su. After the transferring, spring water waits in water
tank for a while. Duration of waiting time is up to demands and production plans.
The water which waiting in water tank goes to ozonation and final process is
distribution of water to production lines.



9

Meanwhile in raw material area, pet preform are coming with boxes and
transferring to inflating machine by workers. Inflating machine never starves. Pet
preform raw materials are always being on hand. Pet preforms goes to two type of
oven. Temperatures of these ovens are 40 C and 60 C. After the inflating processes
for pet preforms, pet preforms transforms to pet bottles and they are preparing for
washing process. In washing process, bottles turns 360 C and goes to detailed
washing via heliozoan machine. Detailed washing is doing by well water.

At the beginning of third main processes, spring water meets with pet bottles
and goes to filling process. Remaining steps of production is common for all lines.
After the filling process, bottle goes to capping process via conveyor. Then following
processes labeling, shrinking, palletizing and stretching are finishing production and
products are transferring to warehouse with forklifts.
4. OBSERVATIONS AND SYMPTOMS

4.1. Analysis of Production Processes

Filling-Inflating-Capping: These operations are processing in common
machine and average 15.000 pet bottles are producing in this machine per hour.
Another important detail is defective pet bottle units do not pass 25 units during the
shift. Rotation speed of filling machine is 4000 cycle greater than inflating machine
and filling amount directly proportional with inflating amount. This system avoids
collapse in production line. Capping is last process of machine.

Labeling: Rotation speed of labeling machine is maximum 18.000 cycle per
hour. Minimum value of rotation is 5000 cycle. In this process, bottles are coming
one by one then all bottles are going to shrinking process according to pallet size.

Shrinking: %95 of shrinking processes are processing as 24 bottle pallet size.
Machine is shrinking 560 packages per hour and 4200 packages in one shift.
Rotation speed of machine is maximum 24.000 cycle. Duration of bottles shrinking
is 39 seconds average.

10


Palletizing & Stretching: Palletizing and stretching is last process of all
production lines. There are two types of stretching. One of them are product
returned, the other of them is stretch machine returned. Duration of any palletizing
process is average 7 minutes 21 seconds and duration of any stretching process is
average 1 minute 54 seconds in normal conditions for 24 bottle pallets. 1 pallet is
including 7 level and 2016 pet bottles. Segmentation change takes two shift and it
happens one day in month.

4.2. SYMPTOMS

We analyzed firstly breakdown of machines and stated as important
observation. Fishbone diagram can easily show causes of breakdown of machines as
seen in Figure 7.

Figure 7 : Fishbone Diagram of Breakdown of Machines.









11



Fishbone Diagram shows 6 main reason to breakdowns and alternative reasons to
corresponding main reasons. Main reasons are possibly classified as method,
management, human, environmental factors, machine and others.

Method failures are:
Machine setup
Cleaning.

Machine setup times are very critical for production. It can affect to next or previous
steps of production. Increasing machine setup times absolutely would decrease
production total speed and amount of outputs. Cleaning is dividing into two type.
First type is routine cleaning. Every machine is cleaning in every 2 weeks.


Management failures are:
Shift Change
Training

Shift Change: There are 3 shifts in Pinar Su Bozdogan. Shift are categorized as A, B,
C. Shift duration is 8 hour. Working production lines can possibly change according
to shifts. A and B shifts are more intense than night shift C. In A and B shifts are
producing more products.

Training: Pinar Su company is giving training sessions to workers at least 2 in a
month. Training sessions are very important for productivity of facility.

Human failures are vacations of workers, day-off and breaks in shift.

Machine failures are most important failure of production system. Problems
are categorized as maintenance of facility, breakdowns of machines and producing
defective parts. Most effective failure is breakdown of machines. As well as,
defected products is very important problem for facilities. On the other hand, Filling

12

machine able to miss bottles to filling. Consequently, this situation causes to lose
customer satisfaction.

4.2.1. Analysis of Downtimes in Production Line 1 and Line 2

Primarily, we collected shift reports from facilitys workers. According to this
shift reports, we found out total amount of breakdowns of line 1 and line 2 between 1
February 2014 and 31 March 2014.
According to our analysis, there were two important downtimes category
which we classified as method and machine. Method breakdowns are occurring
mostly in line 2. Machine breakdowns are occurring mostly in line 1. We showed
this result in our chart as seen in below.
Table 5: Ratios of Downtimes



Machine breakdowns are occurring mostly in category which we named as
machine in fishbone diagram. These machine breakdowns are happening mostly in
inflating machine of line 1. Biggest reason of this inflating machine breakdown is
lack of air pressure and insufficient torque.

Method breakdowns are causing stops in production. Routine cleaning
operation is constant but machine setups can change according to demands. They
0
10
20
30
40
50
60
70
Machine Method Other
Causes of Downtimes -%
Line 1 Line 2

13

need to change product segment and it takes two shift long. Machine setups are
taking long time. Machine setup is biggest problem of method category and it mostly
happens in line 2.

5. PROBLEM DEFINITION, TARGET & CRITICAL SUCCESS
FACTORS
Water amount in production lines depends on the arrival of water from
the spring water source. Our most important goal is determine failures period time.
Under the lights of symptoms, our problem is to determine the amount of
produced products from each line when a set of input such as flow rate of water from
source, mean time of failure given. Arena software simulation program will help us
to solve this problem. Productivity of production systems would be measured
through our arena simulation model. Our next critical step is putting values, data and
results of analyzes into the simulation model respectively. According to simulation
results, we can forecast our next interference.
In this project, we examine efficiencies of production lines, utilization of line
processes and determine optimal amount of produced goods. Determining amount of
production in lines can help to increase profits of factory and decrease loss of raw
materials. At the same time, we examine machine breakdowns, machine setups and
other stoppings of production lines. Within this information, we set up a Arena
Simulation model to reach our goals.
5.1. TARGET & CRITICAL SUCCESS FACTORS
Within this project, in line with the observations and symptoms, we determined
these issues,
Inefficient use of production lines
Low profitability from production lines
Long setup times and machine breakdowns
The problem is we look to increasing the efficiency of production lines and the
profitability from the lines.


14

6. LITERATURE REVIEW AND MODELING PERSPECTIVE


6.1. Literature Review
SIMULATION OF MANUFACTURING SYSTEMS
ABSTRACT
This paper discusses how simulation is used to design new manufacturing systems and to
improve the performance ofexisting ones. Topics to be discussed include:manufacturing issues
addressed by simulation, simulation software for manufacturing applications, techniques for
building valid and credible models, and statistical considerations.


MANUFACTURING ISSUES ADDRESSED BY SIMULATION

The following are some of the specific issues that simulation is used to address in manufacturing:

The need for and the quantity of equipment and personnel

Number, type, and layout of machines for a particular objective
Requirements for transporters, conveyors, and other support equipment (e.g.,
pallets and
fixtures)
Location and size of inventory buffers
Evaluation of a change in product volume or mix
Evaluation of the effect of a new piece of equipment on an existing
manufacturing system
Evaluation of capital investments
Labor-requirements planning
Number of shifts


Performance evaluation

Throughput analysis
Time-in-system analysis
Bottleneck analysis

15



Evaluation of operational procedures

Production scheduling
Inventory policies
Control strategies [e.g., for an automated guided vehicle system
(AGVS)]
Reliability analysis (e.g., effect of preventive maintenance)
Quality-control policies

The following are some of performance measures commonly estimated by simulation:


Throughput
Time in system for parts
Times parts spend in queues
Queue sizes
Timeliness of deliveries
Utilization of equipment or personnel




SIMULATION SOFTWARE FOR MANUFACTURING APPLICATIONS

Historically, simulation packages were classified to be of two major types, namely,
simulation languages and applications-oriented simulators. Simulation languages were general in
nature and model development was done by writing code. Simulation languages provided, in
general, a great deal of modeling flexibility, but were often difficult to use. On the other hand,
applications-oriented simulators were oriented specifically toward a particular class of
applications and a model was developed by using graphics, dialog boxes, and pull-down menus.
Simulators
were sometimes easier to learn and use, but might not have been flexible enough for some
problems.

However, in recent years vendors of simulation languages have attempted to make their
software easier to use by employing a graphical model-building approach. A typical scenario
might be to have a palette of model building icons located on one side of the computer screen.

16

The icons are selected from the palette with a mouse and placed on the work area. The icons are
then connected to indicate the flow of entities through the system of interest.

Finally, one double-clicks on an icon to bring up a dialog box where detail is added. On the
other hand, vendors of simulators have attempted to make their software more
flexible by allowing programming in certain model locations using an internal pseudo-language.
In at least one simulator, it is now possible to modify existing modeling constructs and to create
new ones. Thus, the distinction between simulation languages and simulators has really become
blurred. Based on the above discussion, we will now say that there are two types of simulation
packages. A general purpose simulation package can be used for any application, but might have
special features for certain ones (e.g., for manufacturing or process reengineering).
Examples of general-purpose simulation packages are Arena, AweSim, Extend, GPSS/H, Micro
Saint, MODSIM III, SIMPLE++, SIMUL8, SLX, and Taylor Enterprise Dynamics Developer. On
the other hand, an application soriented simulation package is designed to be used for a certain
class of applications such as manufacturing, health care, or call centers. Examples of
manufacturing-oriented simulators are Arena Packaging Edition, AutoMod, AutoSched, Extend +
MFG, ProModel, QUEST, Taylor Enterprise Dynamics Logistics Suite, and WITNESS.


DEVELOPING VALID AND CREDIBLE SIMULATION MODELS
A simulation model is a surrogate for actually experimenting with a manufacturing system, which
is often infeasible or not cost-effective. Thus, it is important for a simulation analyst to determine
whether the simulation model is an accurate representation of the system being studied, i.e.,
whether the model is valid. It is also important for the model to be credible; otherwise, the results
may never be used in the decision-making process, even if the model is valid.
The following are some important ideas/techniques for deciding the appropriate level of model
detail (one of the most difficult issues when modeling a complex system), for validating a
simulation model, and for developing a model with high credibility:

o State definitively the issues to be addressed and the performance measures for
evaluating
a system design at the beginning of the study.
o Collect information on the system layout and operating procedures based on
conversations
with subject-matter experts (SMEs).
o Delineate all information and data summaries in an assumptions document, which
becomes the major documentation for the model.
o Interact with the manager (or decision-maker) on a regular basis to make sure that
the correct problem is being solved and to increase model credibility.
o

17

o Perform a structured walk-through (before any programming is performed) of the
conceptual simulation model as embodied in the assumptions document before an audience
of SMEs, managers, etc.

o Use sensitivity analyses to determine important model factors, which have to be
modeled carefully.
o Simulate the existing manufacturing system (if there is one) and compare model
performance measures (e.g., throughput and average time in system) to the
comparable measures from the actual system.




STATISTICAL ISSUES IN SIMULATING MANUFACTURING SYSTEMS
Since random samples from input probability distributions drive a simulation model of a
manufacturing system through time, basic simulation output data (e.g., times in system of parts)
or an estimated performance measure computed from them (e.g., average time in system from the
entire simulation run) are also random. Therefore, it is important to model system randomness
correctly and also to design and analyze simulation experiments in a proper manner. These topics
are briefly discussed in this section.



Simulation of Manufacturing Systems


Modeling System Randomness

following are some sources of randomness in
simulated manufacturing systems:

Arrivals of orders, parts, or raw materials
Processing, assembly, or inspection times
Machine times to failure
Machine repair times
Loading/unloading times
Setup times




18

In general, each source of system randomness needs to be modeled by an appropriate probability
distribution, not what is perceived to be the mean value. Note that sources of randomness
encountered in practice are rarely normally distributed. A detailed discussion of simulation input
modeling is given in Chapter 6 of Law and Kelton (1999).


Design and Analysis of Simulation Experiments
Because of the random nature of simulation input, a simulation run produces a statistical estimate
of the (true) performance measure not the measure itself. In order for an estimate to be
statistically precise (have a small variance) and free of bias, the analyst must specify for each
system design of interest appropriate choices for the following:


o Length of each simulation run
o Number of independent simulation runs
o Length of the warmup period, if one is appropriate



We recommend always making at least three to five independent runs for each system
design, and using the average of the estimated performance measures from the individual runs as
the overall estimate of the performance measure. (Independent runs means using different random
numbers for each run, starting each run in the same initial state, and resetting the models
statistical counters back to zero at the beginning of each run.) This overall estimate should be
more statistically precise than the estimated performance measure from one run. Note that
independent runs (as compared to one very long run) are required to obtain legitimate and simple
variance estimates and confidence intervals.

For most simulation studies of manufacturing systems, we are interested in the long-run (or
steady-state) behavior of the system, i.e., its behavior when operating in a normal manner. On
the other hand, simulations of these kinds of systems generally begin with the system in an
empty and idle state. This results in the output data from the beginning of the simulation run not
being representative of the desired normal behavior of the system. Therefore, simulations are
often run for a certain amount of time, the warmup period, before the output data are actually
used to estimate the desired performance measure. Use of the warmup-period data would bias the
estimated performance measure. A comprehensive treatment of simulation output-data
analysis can be found in Chapter 9 of Law and Kelton
(1999).




19

6.2. Modeling Perspective

20



21

Figure 8 : Arena Simulation Model Draft
Variables;
Rated speed of machines Xi;
The amount of spring water Y;
Finished goods P
Parameters;
Machines(Line 1) Rated Speed per hour
Inflating 12000
Filling 14000
Labeling 12000
Shrinking 14400
Palletizing 14400

Machines(Line 1) Rated Speed per hour
Inflating 10000
Filling 14500
Labeling 14500
Shrinking 30 pallet/min

Table6: Parameters of machines in Line 1

Note: One pallet is formed by 7 floor and one pallet is including 2016 bottle. ( for
0.5lt pet bottle )
These data were taken from the company. And we observed these machines
and they suit the standard numbers. Because of the inflating machines be a first
machine and its rated speed is the lowest rated speed, there is no any capacity
problem for other machines.



22

7. SOLUTION METHODOLOGY



In order to solve this case, we made some literature research then we decided
to make simulation model. Within this simulation software, we could know that
actual rate of water which is transferring from spring water and how many tones of
water should transfer to which lines. And also we could do throughput analysis,
time-in-system analysis and bottleneck analysis. We had to follow some principles to
reach our goal. These principles are arrivals of orders, parts, process times of
machines in lines, machine times to failure, machine repair times and setup times of
machines. Our research showed us that we had to use discrete event simulation for
our project. Discrete event simulation has three main advantages in our project.

Risk avoidance Hypothetical or potentially dangerous systems can be
studied without the financial or physical risks that may be involved in
building and studying a real system
Repeatability The ability to study different systems in identical
environments or the same system in different environments
Control Everything in a simulated environment can be precisely monitored
and exactly controlled


After the stating our parameters, we clarified our variables for the model. Our
three critical variables are rated speed of machines, the amount of spring water, and
produced goods per lines. These 3 variables are significative factors of simulation
model. Defining one of these variables, produced goods would also explain our result
of project. Cause by this way, we can easily state throughput of lines.






23

In order to complete these requirements, we needed to fill some blanks of
simulation model. First goal was collecting data of production lines. As its
mentioned in 4.2, our first step was defining breakdown rates of production lines. We
collected all daily production reports of February, March and June months. These
daily production data are including breakdowns, machine setups, total amount of
water which used in production lines and amount of water extravagance as shown
Figure 9.






Figure 9: Daily Production Report




24


Figure 10: Daily Breakdown Report

In this reports, our priority was find out breakdowns rate. This would help us
in simulation model because we put them to model as machines failures and within
the failures, we could find real throughput level. We selected all breakdowns from
daily production reports and copied them to blank worksheet. This separation process
would help us to analyze data more easily. We opened all single days and shifts of
reports and completed separation process. After the separation, we separated them
again and classified them as processes as shown Figure 11.



Figure 11: Production Report Review

25




We did this separation process for 8 production lines and separate them into
their processes as 7 worksheets. One worksheet includes process, reason and time of
breakdowns. These worksheets classified as process names (Bottle Fabrication,
Ozonation, Labeling, Shrinking, Inflating, Palletizing and Line). We classified
general breakdowns (which not related to specific processes) as Line. These
worksheets breakdown frequencies are more than other processes so we eliminated
processes which happened at most 5 times in 3 months. Finally, we draw a chart of
these downtimes analysis for Line 1 and Line 2. We picked first two lines because
their downtime ranges are more than other lines. These charts show us percentages of
downtimes of Line 1 and Line 2 as seen as Figure 12.


Figure 12: Percentages of Downtimes.


Our next step was calculating up times and down times. Up time means time
between first breakdown and next breakdown of each process. On the other hand;
Downtime is repair time of each process. (Figure 13) These analyzes are playing
very critical role in simulation model. Before inserting our breakdown data to model,
we had one more step to go. We had to convert this data to arena simulation
softwares language. Input analyzer tool was eligible way to this step. We put our up
time and down time data to input analyzer and we got a histogram of our data.
0
5
10
15
20
25
30
35
40
45
50

26

After following next steps, we found distributions of our data. (Arena ->
Tools -> Input analyzer -> Fit -> Fit all).Then, we collected all input analyzer data of
production lines in one excel sheet. That excel sheet shows distributions and p-values
of data according to chi-square test as shown as Figure 14 and we showed input
analyzer results in figures 15 and 16.




Figure 13: Down time and Up Time Excel Sheet




27



Figure 14: General distributions of breakdowns of lines.






Figure 15: Input Analyzer for Line 3 ( Down Time)



28


Figure 16: Input Analyzer for Line 3 ( Up Time)





During the input analyzer work, determining p-value was very important
because p-values are giving an idea about our data. We only needed to reach ( p <
0.005) equation in input analyzer results. Assume that our data analyses and results
in input analyzer are our hypothesis, due to p < 0.005 equation, we accept %95
confidence level, while we were matching equation value, we had a proof to accept
our hypothesis. Otherwise, if we couldnt match, we had to reject our hypothesis.
Input analyzer results were showing also Kolmogorov-Smirnov test values but chi-
square method is more reliable for our model. Within the input analyzer, we also
defined ranges, intervals, mean and standard deviations of data.


Our input analyzer showed us important information about our down times
and up times data. Most of down times and up times are fitting with Beta
distribution. In order to confirm our hypothesis, we used Minitab application. We
tried to match our input analyzer equation to Minitab as shown as Figure 16&17.

29



Figure 17: Distribution of down time of Line 3 on Minitab.


30




Figure 17: Distribution of up time of Line 3 on Minitab.

31

8. ARENA SIMULATION MODEL


Arena simulation model is last and most important step of our project. Within
this model, we aimed to reach our goals as its mentioned in literature review and
problem definition. In our case, our model has two main model structure. First model
structure is flow process model and its representing the water which transferring
from spring water. Other model is representing 8 production lines of Pnar Su.
Congregating of two model is representing transfer of water to production lines. This
is the main idea of our model. Another reason to pick this model was structure of
production. Pnar Su production is product-layout type production.







32




Flow process is beginning of the model. We analyzed coming rate of spring
water according to months and put them to Regulator module respectively. Due to
our analyzed reports and daily production reports, we could define actual water rate
of each production lines. This information would help us to answer these questions.
Which line needs more water? or Which line shouldve produce more goods?
After inserting the spring water data, we used Flow module to represent inflating
of water tanks. There are two water tanks in facility and capacities of them are 600
tonnes. Spring water is flowing to these tanks directly and from these tanks, water
goes to ozonating room. After the ozonating water, its delivering to production lines
by distributor. These processes are represented with in our model.




Other model is about conveyors. We put Create module for each
production lines. We inserted arrival date of water from tank. Then we put two
Process module to represent line failures. We put into the these modules
distributions of machine failures.













33





















11. CONCLUSION


During the graduation project period, we got the chance to observe lots of
information about industrial engineering activities and the working system of Pnar
Su Facility. It is a great experience for us to see the production mechanism of a live
production company. Also this period provided us a chance to apply the theoretical
knowledge gained from school in real working environment.
We have met employees and engineers belonging to other engineering
disciplines. The most outstanding factor that separates industrial engineers from
other engineers is systematic thinking ability and the difference in the point of view.
This ability provides us to see the lack of production process and suggest
improvements.

34



12. REFERENCES


Banks, J., J. S. Carson, and B. L. Nelson. 1996. Discrete-
Event System Simulation. 2d ed. Upper Saddle River,
New Jersey: Prentice-Hall.

Law, A. M. 1998. How to Select Simulation Software.
Tucson, Arizona: Averill M. Law & Associates.
Law, A. M., and W. D. Kelton. December, 1999.
Simulation Modeling and Analysis. 3d ed. New York:

McGraw-Hill.
Law, A. M., and M. G. McComas. Secrets of Successful
Simulation Studies. Industrial Engineering 22: 47-48,
51-53, 72.
Proceedings of the 1999 Winter Simulation Conference
P. A. Farrington, H. B. Nembhard, D. T. Sturrock, and G. W. Evans, eds.
Averill M. Law
Michael G. McComas
13. APPENDCES