ENGINEERING SIMULATION

Table of Contents

Student ID Page 1
EXECUTIVE SUMMARY...........................................................................................................................2
1. INTRODUCTION.................................................................................................................................2
2. WORKING METHODOLOGY................................................................................................................4
2.1 Petri Net......................................................................................................................................4
2.2 Modelling.....................................................................................................................................5
3. REVENUE AND BUDGET.....................................................................................................................7
4. OPTIMISATION AND BREAKDOWN....................................................................................................8
4.1 Optimisation................................................................................................................................8
4.2 Breakdown...................................................................................................................................9
5. RESULTS...........................................................................................................................................10
5.1 Task 1 – Optimisation................................................................................................................10
5.2 Task 2 – Breakdown...................................................................................................................15
5.2.1 Conveyor Belt Breakdown Stats..........................................................................................16
5.2.2 Conveyor Assembly Statistics..............................................................................................17
6. DISCUSSION.....................................................................................................................................19
7. CONCLUSIONS.................................................................................................................................19
8. LIST OF REFERENCES........................................................................................................................20

List of Tables

Table 1 Budget after optimization.......................................................................................................13

Table 2 Revenue after optimisation.....................................................................................................16
Table 3 Conveyor Belt statistics before breakdown............................................................................16
Table 4 Conveyor Belt statistics after breakdown...............................................................................17
Table 5 Assembly statistics before breakdown....................................................................................17
Table 6 Assembly statistics after breakdown.......................................................................................18
Table 7 Disassembly before breakdown..............................................................................................18
Table 8 Disassembly after breakdown.................................................................................................19

Student ID Page 2
EXECUTIVE SUMMARY

Manufacturing of gear box products in an automotive gear plant is considered to be a

challenging and complex task which involves combining lot of complicated components.
Simulation is widely used in the design and modeling of such complex production systems in
the gear plants. There are assembling and disassembling of these systems, in this case we use
the gear box pant to study the design and modeling. There is a great complexity in the
production of these gear box products due to their change in process, schedule changes,
dependency on the availability of the labour workers, skilled workers, and most importantly it
depends on the performance of the material handling equipments. In such complex systems,
simulation plays an important role and helps to design and model these complex systems.
These further allow us to optimize the performance of the entire plant. The ultimate aim of
this assignment is to create a model and optimise the revenue for the plant over ten days
operation after an appropriate warm-up period.

1. INTRODUCTION

Gear manufacturing plant is an intricate task of manufacturing and combining of

thousands of gearboxes and bearings. In automobile industry gearing system is very
important part for manufacturing automobiles as gearing system are vital piece of every
automobile. The major components that comprise the gearing systems are engine and
the transmission. Manufacturing good quality gear boxes are very important to
manufacture good quality automobiles. All kind of components of automobile or sub-
assemblies such as gears, engines and transmission are manufactured independently
and then put together to make a frame work for the final phase of an automobile. So the
main parts of the engine which are camshaft, crankshafts etc. are assembled according
to the sub-assemblies. And then these sub-assemblies are put together to form an
engine.

Various automotive manufacturing systems are designed and implemented by using a

regular tool called discrete event simulation. These automotive manufacturing systems
range from joining rod machining sub-system to automotive assembly system. Ulgen et

Student ID Page 3
al. (1994), Gunal et al. (1996), and Jeyebalan et al. (1992) provide some instances of
such booming applications. The aim of simulation differs from one system to another.
Some of the common objectives of simulation described by Jayaraman and Agarwal,
(1996) are system output determination, recognition of bottleneck, allocation and
maximization of operators, comparison of operating philosophies, designing and
analysis of logistics structure, material storage issue analysis, optimizing the patterns of
shift and Materials handling systems design. This study provides attention on the
application of simulation for gear plant production systems in automobiles. Throughout
the process till final assembly system, the advantage of simulation can be confirmed
that is from gear last test and repair and in transmission of gearboxes by conveyor belts.
Testing is the one and only operation that is performed in every assembling of gear and
transmission. Testing is done in every assembly on specific designed test stand. On an
average the time taken to process an assembly in the test stand is longer than the time
taken in other stations of the system. Even a breakdown operation depicts a normal
processing time because breakdown can happen in the gear assembly due to many
reasons. So this is the reason for having more test stands in the system. All the test
stands are linked to the left over assembly line by way of conveyors in a normal set up.
In order to save cost and to meet with the demand it is very essential to use the test
stations effectively and uniformly and this can be done by arranging them properly in
the set up. Simulation is often utilized to check the various options and to choose the
best pattern because of the dynamics of the testing part.

Discrete event simulation and continuous event simulation are different from each
other. As instant alterations in the system state can happen in discrete event simulation
but they cannot happen in other simulation. The illustrations of discrete event
simulation are counters in railway reservations, coffee shop queues, etc. (Mathew,
2013). Fujimoto (1990) describes that a discrete event simulation model imagines that
system that is simulated in the time of simulation changes its state at the discrete
position. The simulation model leaps from one position to another based on the
happening of the event. The software that is used for discrete event simulation is
WITNESS, Arena etc. This illustration used Witness software for simulation. To get a
complete understanding of the model, the plant that was discussed in the problem
description is being enlighten in this segment. The gear is produced outside the plant
and then it enters the plant and kept in the conveyor belt. The size of the gear is 80mm x
Student ID Page 4
80mm x 100 mm and it is supplied to the conveyor belt which is of 20m length. Then the
conveyor supplies the materials to the assembly unit. Here the gears are put together
with bearings and casted to make gearbox. The speed and the optimum capacity of the
conveyor are 5m/mm and 180 parts respectively. Bearings are obtained at the other
end of the assembly unit and then the casting to CNC.

The casting is bored by passing through CNC machine to machine. Loading and
unloading of the components to machines are taken cared by an operator. A uniform
cycle time is maintained in CNC machine for loading and unloading. The loading time
ranges from 0.5 minutes to 1.0 minutes and the maximum unloading time is 0.9
minutes. The operating time of CNC machine is 2.3 minutes. The 20th gearbox is
inspected while passing by CNC machine and the rest of the gear boxes are put into the
buffer called bin and maximum of 50 parts can be accommodated in the bin. The cycle
time is circulated triangularly in the inspection station and the time ranges from 0.8 to
2.2 minutes. Information about the plant is provided in the problem description and this
detail is utilized to create a design for the model to find out the profit of the plant.

2. WORKING METHODOLOGY

This segment discusses the background and types linked with modelling and the
system. It is clearly described in the assignment that gears are separately manufactured
arrive via a 20 m conveyor. In the assembly area, the gear boxes are assembled with 2
gears and 2 bearings.

Every 20th gearbox machined is subject to inspection and the remaining sent to the
buffer.

Cost of the CNC machine  £72000

Budget available  £15000

2.1 Petri Net

Student ID Page 5
A Petri net is a mathematical modeling language that is used for explaining distributed
system. Petri net is a source for modeling this plant that has stochastic system with
isolated events. According to Ling (2001), Petri net method was invented by Carl Adam
Petri in 1962. Petri net method is used to replica a distributed system by way of
concurrency, harmonization, distribution, parallel, nondeterministic or stochastic (Ling,
2001and Murata1989). This method can be used to any areas or system which can be
explained graphically using flow charts, block diagrams and other mathematical tools.
The thesis of Carl Adam Petri from which Petri net arrived provided an opportunity of
communication between both the scholars and the practitioners (Murata, 1989).
Murata (1989) in his paper said that the computer organisation group of MIT conducted
many vigorous studies on the Petri net method after its invention. The inbuilt
progressiveness and simplification feature of Petri net increased its range of
applications. Petri net consist of some difficult problems but still parallel and
concurrent activities can be represented by this. The help of computer aided means are
required to apply Petri net practically (Murata, 1989).

There are two types of node in Petri net. They are places and transitions. Places show
the condition of the system with using circles and transitions uses bars to depict
condition of the system. Arcs are used to create connection between transitions to
places and places to transitions. Both places and transitions are interconnected with an
arrow that depicts the direction of flow. Places are indicated by m and transitions are
indicated by n. the direction of flow from transition to places are represented as post
incidence matrix and the direction of flow from places to transition are represented as
pre incidence matrix (Fanti, n.d).

After having a look on the example in the Figure 2a &2b Yang (2014) says that the
buffers twisted from places and the processes comes from the transitions. According the
example transition has more time than place. A preset time is set to fire after the input is
completely marked in the system. This example explains that the components reaches at
every 6 minutes and the 10 minutes are taken for processing which clear from the Fig.
2b. This is just a raw plan of the Petri net with the distinct event reproduction.

2.2 Modelling

Student ID Page 6
 Modelling of the system is done with help of WITNESS. WITNESS is distinct event
stochastic system reproduction software that is delivered by Lanner Group. Inc.
Modelling of system requires four major components namely parts, machines, buffer,
variables and function. Based on the need of the modelling, parts are used and taken to
the machines. Gears, castings and bearings are the parts that are used in modelling
which in turns become gearboxes. Buffer is a special function of the system and it
performs important function as bin, testing and other aided actions. Variables give the
path to perform the actions in the system. The example shows that 78% of the
components would be in the higher tolerance limit during the inspection station and the
CNC machine has to be dirt-free. Variable plays a vital role in the action of this function
in the system and the following program structure shows this;

IF Binomial ( 0.78,20 ) < N passed

CleanCNC =1

N passed =0

ELSE

N passed =N passed +1

ENDIF

The start and the end actions must be clearly explained in every part in the system.
Process of giving and taking with one another can be done by ‘PULL’ and ‘PUSH’.
Operators are appointed to the control the functions of the system. But with specialised
operators the need of large number of operators can be reduced. The revenue
administrator of the system can be guided by these functions and the operators.
Revenue of the system and shifts of the operators can be controlled by this. The
binomial function in the above program structure helps to repeat the actuality that
there is an opportunity of 78% parts to be near high tolerance limit for every 20%
components. Variable can be used to clean the CNC machine and after cleaning N passed
variable cab be utilized to make the next move in the system.

Since all the parts have to be accumulated to become gearboxes, the input principle of
the assembly must be done in sequence. The casting, gear and bearing are taken one

Student ID Page 7
after the other from their particular source to the system and they are operated in the
structure as below.

SEQUENCE/Wait Casting out of Bin ¿(1),

Gear out of Conveyor Belt at Front ¿ (2),

Bearingsout of WORLD ¿(2)

Once all the parts are assembled in the system to become gearboxes, the function called
CHANGE ALL to Gearbox is to be selected in the action on finish of the assembly. Gear
boxes have to be tested after finishing the assembly process. Before testing in the
testing machine these gearboxes have to be tested in the test buffer. While testing in the
test buffer there are probability that only 93% of the part passes the test and the rest of
7% fails. This evident from the earlier completed operations which are detailed in the
case. The unsuccessful gearboxes are disassembled in the disassembly station. The
disassembled bearings and castings are sent to the scrap place whereas the gear goes to
the conveyor belt once again as a part.

The shift of the operators can calculated in many ways. Proper shifts with or without off,
and duration of the shift can be selected for operator shift. A 10 days work report was
taken in this regards which stated the shift was selected with duration of 2 weeks with 2
days off in a week. This has to be given properly to imitate the actual system.

Witness can be used to determine the cost and profit of the system. The rate of the
machine, cost function of the plant, etc., can also e defined. Machine rate can be
identified by the time needed to retrieve the money paid and by the time needed to run
the machine to retrieve the money paid. The Cost function of the plant can also be
identified (Yang, 2014). Calculation has to be performed depending upon the direct and
indirect cost. For instance, the cost of a machine must be divided by some period of time
like 4 years and therefore the cost of machine is £12000. Similar to this the operator
cost can also be identified for an hour or so. Likewise all the cost has to be found out.
And with these results the real revenue of the system can be identified (Yang, 2014).

Student ID Page 8
3. REVENUE AND BUDGET

Equations can be used to calculate revenue of the system and the parts of the equations
are already discussed in the above segment. And so the revenue for the given system
can be recognized. According to Yang (2014) the revenue of the system should be
optimum.

RETURN NS h ip(Gearbox)∗80−¿

NQty( CNC)∗12+ NQty (Test Buffer )∗0.1+¿

NQty( Test Benc h)∗1.1+ NQty (Operator )∗16.35 ¿∗80

Several case studies of the organisation which had enhanced their revenue and
manufacture by using witness analysis have acknowledged in the website in
Lanner.com (2008). To enhance the system and to know the feature, it is vital to watch
over the revenus of the system.

As discussed earlier in the plant description, the budget for setting up the plant is
allocated for an amount of fifteen thousand pounds not including the CNC. There are
various direct and indirect costs come with set up of any plants (Yand, 2014). The
budget should include all the cost of the system. Budget calculation should also include
the various machines after comparing with the revenue of the system to obtain the best
for the plant. The equation (Eq.2) is being improved in the budget function is shown
below.

Return NQty( Assembly )∗1800+ NQty ( Bin)∗200+¿

Nqty (Test Benc h)∗2200+ NQty (Operator )∗1500

4. OPTIMISATION AND BREAKDOWN

4.1 Optimisation

Student ID Page 9
A case study conducted by the Industrial Engineering department of Slovakia at
Volkswagen utilized WITNESS analysis to maximise the new manufacture and logistics
system in the company and which resulted in obtaining good output in the firm
(Lanner.com, 2003). Many other firms also used this analysis to develop their
functionalities same like Volkswagen. A study of the business process behaviour can be
performed by an important method of optimization in the reproduction. Mathematical
and statistical statement can be obtained by the simulation optimization which can aid
to identify the most favourable output with running input in the system (Chramcov,
2013).

According to Chramcov (2013), a favourable solution can be obtained by coordinating

the sequence of the simulation system with the help of optimization. Optimization mode
is given by the lanner group for testing various mixtures in the model so as to identify
the best between them (Markt and Mayer, 1997). Chramcov (2013) describes that
optimization can be done on the bases of Hill climb algorithm. The statistics and the
outcome of an instance with optimization are also available in the work. Low cost of
production, high revenue or upper output level are the outcomes of an optimization.
The optimization means increasing or decreasing a specific function by building the
inputs associated to it. In order to get the best result, this model is also related to
optimization so that these results can be used by the company to enhance the main
objective of the firm and to develop their revenues. Hill climb is one of mathematical
algorithms used for optimization in this work. Apart from this, there are many other
mathematical algorithms available for optimization. Juels and Wattenberg (1994)
clarifies Hill climbing by considering optimization as a landscape in which the tallness of
each position and the result for each ‘s’ relates to the strength of the solution (f(s)).With
higher strength Hill climb rush forward to a peak by shifting to close by state. It can help
to find the solution space. There are several steps in the Hill climbing. These steps start
with the selection of a random position in the search space. The next step is to consider
the neighbouring of the present state and then selecting the neighbour with the most
excellent quality and shifting to that place. This step is repeated till all the other
neighbouring state turn into lower quality (Kendall, n. d.). Hill climb uses this algorithm
as bases for optimization. And this algorithm can also be utilized for optimization of the
system.

Student ID Page 10
4.2 Breakdown

Waller (2010) defines the term breakdown as a main factor that brings unevenness in
producing systems by breakdown of equipment. He says that the science of modelling
breakdown is weak. And so it is clear that every system has a chance of breakdown and
it can be visible in the witness analysis. There are various reasons for disruption in the
machine at various breakdowns that can be understood in WITNESS analysis. In order
to avoid these breakdowns the division has be selected properly. The actions of the
distribution curve of the real life can be different from the action of the distribution
system curve and this is arrived from the few system failures (Waller, 2010).

When breakdown occurs in a system, witness provides two alternatives of modeller.

One option of witness is to sustain the component in the cycle and again using the same
part to continue the cycle after repairing the part. The second option is to crumb the
breakdown part and to repair the cycle with the new part and to use this new cycle to
start working (Tague, 2010). Breakdown in the system should be taken into account
while using this model. Almost 90 minutes of time must be considered as the
breakdown time in the system. According to the name breakdown time is described as
the time needed to repair the part of the machine together with breaking down (Lanner,
2008). The numbers of breakdowns that occur in the system are counted and showed
using breakdown counter. The breakdown time for the conveyor belt is 10000 minutes
and for the CNC machine is 50000 minutes. The breakdown in the system can counted
using the following function.

Breakdown counter=Breakdown_counter+1

5. RESULTS

This assignment focuses on addressing two different problems that has been mentioned
earlier chapters. The primary aim is to optimise the system and determining the best
possible revenue for the system within the provided budget. The secondary aim of this

Student ID Page 11
assignment is to break the system into different points in order to determine the nature
of the system.

5.1 Task 1 – Optimisation

WITNESS has the ability to provide a critical and detailed experiment method to
determine the optimisation of a system. This optimisation is used to experiment the
system which would give nearly equal real time result. These results are helpful to
identify the issues within the systems and provide an opportunity to address these
problems by modifying the existing system. By optimising the plant, we can find the
revenue and budget of the plant.

This plant has various machineries, elements, parts, and variables. These elements are
all linked through an optimisation tool, hill climb algorithm. Some of the components
that we use to optimise the plant includes but not limited to assembly, bin, and
Conveyor belt, Operator, Test Bench and Test Bench. These elements are optimized by
using the hill climb algorithm.

Minimum Quantity  1

Maximum Quantity  3

Hill climb algorithm utilises various combinations of these components to maximise the
optimisation. Thus the best revenue can be found which comes within the budget.

The revenue and budget for the best is obtained from the combination 40, which has 3
assemblies, 2 Bin, 1 Conveyor belts, 2 Test benches, 3 Test buffers and 3 operators. This
combination resulted in the best budget and revenue for the system. The budget of the
system will be £14700. When you check with the other combination 42, which has 3
assemblies, 2 Bin, 1 Conveyor belts, 3 Test benches, 3 Test buffers and 2 operators, the
budget exceeds £15400.

Conveyo Test Test

Student ID Page 12
 r benches Buffer
Assembl Bin Belt Quantity Quantity Operato
y Quantity Quantity r shift
Quantity
1 1 3 3 2 1 2 9800
2 1 2 2 2 2 2 9600
3 1 1 2 2 2 2 9400
4 2 2 2 2 2 2 11400
5 1 2 3 1 1 1 5900
6 2 2 1 1 1 2 9200
7 3 2 2 2 2 2 13200
8 1 2 1 2 1 2 9600
9 3 2 3 3 2 2 15400
10 1 2 1 2 1 2 9600
11 3 2 1 3 1 2 15400
12 1 2 1 2 1 1 8100
13 2 2 1 1 1 2 9200
14 3 2 2 2 1 2 13200
15 2 2 2 2 2 2 11400
16 1 2 2 3 1 2 11800
17 1 2 3 3 1 2 11800
18 2 2 3 3 1 2 13600
19 3 2 3 3 1 3 16900
20 1 3 3 3 1 3 13500
21 1 2 3 3 2 3 13300
22 1 2 2 3 1 3 13300
23 2 2 3 3 1 3 15100
24 3 2 3 3 2 3 16900
25 2 2 3 2 2 3 12900
26 3 1 3 3 2 3 16700
27 3 2 1 2 2 1 11700
28 3 2 3 3 2 2 15400
29 3 2 1 3 2 1 13900
30 3 2 1 2 2 1 11700
31 3 2 2 3 1 1 13900
32 3 2 1 2 2 2 13200
33 3 2 2 3 2 2 15400
34 2 2 1 3 2 2 13600
35 2 1 2 2 3 8200
36 2 2 2 3 3 2 13600
37 2 2 2 2 3 2 11400
38 2 2 1 3 3 2 13600
39 3 2 1 2 2 1 11700
40 3 2 1 2 3 3 14700
41 3 2 3 3 3 1 13900

Student ID Page 13
 42 3 2 1 3 3 2 15400
43 1 1 1 2 3 3 10900
44 3 2 3 3 1 1 13900
45 1 2 3 2 3 1 8100
46 2 1 3 3 1 2 13400
47 1 2 3 2 1 3 11100
48 1 1 3 3 3 3 13100
49 1 2 3 2 1 2 9600
50 2 1 3 3 1 1 11900
51 2 2 3 2 1 3 12900
52 2 1 2 3 2 2 13400
53 1 1 2 2 3 3 10900
54 1 2 3 2 1 1 8100
55 1 2 2 3 2 2 11800
56 1 2 3 2 3 1 8100
57 1 2 2 2 3 3 11100
58 1 2 2 2 2 2 9600
59 2 2 2 2 1 3 12900
60 1 2 3 2 3 2 9600

Table 1 Budget after optimization

Conveyor Test Test

benche Buffer
s
Assembl Bin Belt Quantit Quantity Operator Revenu
y Quantit Quantity y shift e
Quantity y
1 1 3 3 2 1 2 13700
2 1 2 2 2 2 2 13500
3 1 1 2 2 2 2 13100
4 2 2 2 2 2 2 14500
5 1 2 3 1 1 1 9000
6 2 2 1 1 1 2 12900
7 3 2 2 2 2 2 16500
8 1 2 1 2 1 2 13500
9 3 2 3 3 2 2 18500
10 1 2 1 2 1 2 13500
11 3 2 1 3 1 2 18500
12 1 2 1 2 1 1 12100
13 2 2 1 1 1 2 12900
14 3 2 2 2 1 2 16500
15 2 2 2 2 2 2 14500
16 1 2 2 3 1 2 14900

Student ID Page 14
 17 1 2 3 3 1 2 14900
18 2 2 3 3 1 2 17000
19 3 2 3 3 1 3 19500
20 1 3 3 3 1 3 16700
21 1 2 3 3 2 3 16300
22 1 2 2 3 1 3 16300
23 2 2 3 3 1 3 18200
24 3 2 3 3 2 3 19500
25 2 2 3 2 2 3 15400
26 3 1 3 3 2 3 19200
27 3 2 1 2 2 1 14800
28 3 2 3 3 2 2 18400
29 3 2 1 3 2 1 16900
30 3 2 1 2 2 1 14800
31 3 2 2 3 1 1 16900
32 3 2 1 2 2 2 16500
33 3 2 2 3 2 2 18500
34 2 2 1 3 2 2 16500
35 2 1 2 2 3 11900
36 2 2 2 3 3 2 16700
37 2 2 2 2 3 2 14700
38 2 2 1 3 3 2 16900
39 3 2 1 2 2 1 14900
40 3 2 1 2 3 3 17500
41 3 2 3 3 3 2 17100
42 3 2 1 3 3 2 18500
43 1 1 1 2 3 3 12500
44 3 2 3 3 1 1 16900
45 1 2 3 2 3 1 11800
46 2 1 3 3 1 2 16700
47 1 2 3 2 1 3 14500
48 1 1 3 3 3 3 16500
49 1 2 3 2 1 2 12900
50 2 1 3 3 1 1 15100
51 2 2 3 2 1 3 15900
52 2 1 2 3 2 2 16500
53 1 1 2 2 3 3 12500
54 1 2 3 2 1 1 11800
55 1 2 2 3 2 2 15100
56 1 2 3 2 3 1 11800
57 1 2 2 2 3 3 14500
58 1 2 2 2 2 2 12900
59 2 2 2 2 1 3 15900
60 1 2 3 2 3 2 12900

Student ID Page 15
 Table 2 Revenue after optimisation

5.2 Task 2 – Breakdown

The second major task is to determine the nature and character of the plant after it is
totally broken down. It is a well known fact that breakdown is unavoidable in any plant.
It is very important to forecast the breakdown; this forecast helps to manage the
breakdown and it also helps to reduce and avoid these breakdowns. Conveyor Belt and
CNC are breaking down for 90 minutes in the experiment on the daily production rate.
And the time between each failure given for the Conveyor Belt is 10000 minutes,
whereas for CNC machine is 5000 minutes.

5.2.1 Conveyor Belt Breakdown Stats

The below table shows the key stats of the conveyor belt before the breakdown; since
the Conveyor is not broken down the broken down value shows 0 whereas 31.55% is
moving and 68.45% parts were blocked on the conveyor. The average time for the
Conveyor belt is 22.56.

% Empty 0
% Move 31.55
% Blocked 68.45
% Queue 0
% Broken 0
Down
Now on 1
Total On 1027
Avg Size 2.15
Avg Time 22.56

Table 3 Conveyor Belt statistics before breakdown.

% Empty 0
% Move 24.25
% Blocked 75.75
% Queue 0

Student ID Page 16
 % Broken 0.79
Down
Now on 1
Total On 1007
Avg Size 2.15
Avg Time 22.56

Table 4 Conveyor Belt statistics after breakdown.

5.2.2 Conveyor Assembly Statistics

The average size of the assembly used in the test is 1.91%. The % idle and % busy of the
system is 97.65 and 0.65 respectively. The machine is 1.43% blocked which is before
the breakdown. The same for the machine after the break down is also identified.

Name Assembly
% Idle 58.5
% Busy 47.5
% Filing 0
% Emptying 0
% Blocked 0
% Cycle Wait 6.8
Labour
Broken Down 0

Table 5 Assembly statistics before breakdown

Name Assembly
% Idle 58.78
% Busy 47.58
% Filing 0
% Emptying 0
% Blocked 0
% Cycle Wait 6.54

Student ID Page 17
 Labour
Broken Down 0

Table 6 Assembly statistics after breakdown

The idle percentage of assembling before breakdown of the conveyor was 58.5 where as
the idle percentage after breakdown was 58.78. This clearly shows that there is small
increase in this idle percentage. The percentage of busy status before breakdown was
47.5 and this has been increased to 47.58 after breakdown. The filing, emptying, and
blocked percentage of the assembly remained same before and after the breakdown of
the plant. We can also see that the percentage of cycle wait labour has shown a small
decrease in the percentage from 6.8 to 6.54.

Name DisAssembly
% Idle 59.68
% Busy 28.65
% Filing 0
% Emptying 0
% Blocked 0.19
% Cycle Wait Labour 3.95
Broken Down 0
% Setup 0

Table 7 Disassembly before breakdown

The disassembly of the conveyor belt before the breakdown has shown the above
statistics such as the idle percentage was 59.68, the busy percentage was 28.65, the
filing and emptying was zero. The percentage blocked before the breakdown was 0.19
and the percentage of cycle wait labour was 3.95. The above statistics clearly shows that
the broken down and percentage of setup before the breakdown of the assembly was
zero.

Name DisAssembly
% Idle 60.15
% Busy 27.25

Student ID Page 18
 % Filing 0
% Emptying 0
% Blocked 0.25
% Cycle Wait Labour 4.15
Broken Down 0
% Setup 0

Table 8 Disassembly after breakdown

The disassembly after breakdown shows some variations in the parameters such as
little increase in the idle percentage to 60.15, and decrease in the busy percentage from
28.65 to 27.25. The filing and emptying percentage was same before and after the
breakdown.

6. DISCUSSION

This system was simulated using the software WITNESS. The results obtained from the
simulation have given an idea about the plant and what and how it is going to work. The
revenue and budget for the best is obtained from the combination 40, which has 3
assemblies, 2 Bin, 1 Conveyor belts, 2 Test benches, 3 Test buffers and 3 operators. This
combination resulted in the best budget and revenue for the system. The budget of the
system will be £14700. The revenue generated with this combination is £17,500.The
breakdown statistics clearly shows how each machines have impact before and after the
breakdown of the plant. There is some significant change in the assembly parameters
after the breakdown.

7. CONCLUSIONS

Witness software was used to model and simulate the plant and it has been optimised.
This model provides clear optimum values for the plant that includes the highest
revenue within the provided budget for the plant. Within the budget of £14700, the
revenue generated is £17,500. It is also concluded that there is significant change in the
nature and character of the plant after the breakdown. The impact of the breakdown of

Student ID Page 19
each of the CNC workstation and the conveyor for 90 minutes has on the daily
production rate are also addressed in this report.

APPENDICES

Gear statistics

Name Gear before Gear After

breakdown breakdown
No. Entered 1113 1087
No. Shipped 186 182
No. 0 0
Scrapped
No. 922 900
Assembled
No. 65827 65847
Rejected
WIP 5 5
Average 6.67 7.23
WIP
Average 77.62 86.21
Time
Sigma 6 6
Rating

Gear Stats before and after the breakdown

Student ID Page 20
 1200

1000

800

600

Gear before breakdown

400 Gear After breakdown

200

0
d d ed ed IP IP e g
re pe pp bl W W im tin
te ip a m ge
T a
.E
n
.S
h cr se ra ag
e aR
No No .S As e er gm
N o . Av Av Si
No

Gear Stats before and after the breakdown

Appendix 2: Bearing Stats before and after the breakdown

Name Bearing bearing

before after
breakdown breakdown
No. Entered 1034 1014
No. Shipped 186 182
No. Scrapped 0 0
No. Assembled 844 828
WIP 4 4
Average WIP 4.53 5.03
Average Time 58.72 64.28
Sigma Rating 6 6

Student ID Page 21
 1200
1034
1014
1000
844
828
800

600

400 Bearing before breakdown

bearing after breakdown
186
182
200
64.28
58.72
0 0 4 4 5.03
4.53 6 6
0
ed d d ed IP IP e g
ter ppe ppe bl W W Tim atin
i ra em e
.E
n
.S
h
Sc ss ag ag
e aR
No o o. . A v er v er igm
N N A S
No A

Bearing Stats before and after breakdown

Appendix 3: Casting stats before and after breakdown

Name Casting Casting

before after
breakdown breakdown
No. Entered 683 670
No. Shipped 0 0
No. Scrapped 118 112
No. Assembled 513 506
WIP 52 52
Average WIP 49.27 49.38
Average Time 934.82 955.26
Sigma Rating 2.39 2.41

Student ID Page 22
 1200
1000
800
600
400
Axis Title
200 Casting before breakdown
0 Casting after breakdown

Axis Title

Casting stats before and after the breakdown

8. LIST OF REFERENCES
 Chen, C.H. 2010. Stochastic Simulation Optimization: An Optimal Computing Budget
Allocation, London: World Scientific Publication, pp. 1-7.
 Donald, L., Rizzo, D.L., Caird, J., and Lee, J. (2011) Handbook of Driving Simulation for
Engineering, Medicine, and Psychology. California: CRC Press
 Doumeingts, G., and Browne, J. (1997) Modelling Techniques for Business Process Re-
engineering and Benchmarking. London: Springer Publishers
 Fanti, M. P. n.d modelling by Petri net, Dipartimento di electtrotecnica Ed Elettronica,
Polytechnic of Bari, Italy.
 Fujimoto, R.M. 1990. Parallel Discrete Event Simulation, Communication of the ACM,
33(10), pp. 30-53.
 Matthew, J. 2013. Discrete event simulation: Resonance: Journal of Science Education.
[online] <http://web.a.ebscohost.com/ehost/detail?sid=d5d4e5c1-a1df-49be-a267-
fda491beeffa
%40sessionmgr4002&vid=1&hid=4114&bdata=JnNpdGU9ZWhvc3QtbGl2ZQ%3d
%3d#db=ehh&AN=85283434> [Accessed on 19 April 2015]
 Robinson, S. 2004. Simulation: The Practice of Model Development and Use, Chichester:
John Wiley and Sons, p.40.
 Thomas, P. (1999) Simulation of Industrial Processes for Control Engineers [online]
available from < https://books.google.co.in/books?id=-jdauwTehZ4C&dq=> [Accessed
on 20 April 2015

Student ID Page 23
  Waller, T. 2010. Simulation Blog-Breaking Down Breakdowns, [online]
<http://www.lanner.com/en/blog-entry.cfm?theFqID=698845A0-15C5-F4C0-
99764B68F0DEFBDF> [Accessed on 25 April 2015]
 Yang, X.S. 2014. Discrete-Event Simulation: Introduction to Witness, PDE4905
Engineering Simulation. Middlesex University. London

Student ID Page 24