Investigation of greenhouse gas emissions from 

the production of tin cans 
-  An evaluation of the EcoProIT method 

Master of Science Thesis in Production Engineering 
 

 

JOAKIM LARSSON 

MARINETTA TÖRNBERG 
 

 

Department of Product and Production Development 

Division of Production Systems 

CHALMERS UNIVERSITY OF TECHNOLOGY 

Göteborg, Sweden, 2013 



 

 

  



 

 

 

 

 

 

Investigation of greenhouse gas emissions from 

the production of tin cans 
 

-  An evaluation of the EcoProIT method 

 

JOAKIM LARSSON 

MARINETTA TÖRNBERG 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
Department of Product and Production Development 

CHALMERS UNIVERSITY OF TECHNOLOGY 

Göteborg, Sweden 2013 

  



 

 

Investigation of greenhouse gas eimissions from the production of tin cans  

-  An evaluation of the EcoProIT method 

 

JOAKIM LARSSON 

MARINETTA TÖRNBERG 
 

© JOAKIM LARSSON & MARINETTA TÖRNBERG, 2013 

 
 
 
 
Master of Science Thesis Work  
Department of Product and Production Development  
Chalmers University of Technology  
SE-412 96 Göteborg 
Sweden  
Telephone: + 46 (0)31-772 1000 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
Front cover:  
Superstudio 
Caption:  
EcoProIT Method 
Chalmers Reproservice  
Göteborg, Sweden, 2013   



 

 

Preface 
This master thesis was carried out during spring term 2012 at the institution for 

Product and Production Development at Chalmers University of Technology. The 

Master programme was Production Engineering and the Thesis covers 30 credits. 

 

We would like to thank Emballator Ulricehamns Bleck and EcoProIT for the 

opportunity to participate in this project. 

 

The supervisor was Jon Andersson and the examiner was Björn Johansson as a 

member of the EcoProIT project Anders Skoogh was also available for support during 

the master thesis. We would like to thank them for their support during this project. 

We would also like to thank everybody at Emballator Ulricehamns Bleck for 

answering our many questions with a smile.  

 

June 2013, Göteborg 

 

Joakim Larsson & Marinetta Törnberg  



 

 

  



 

 

Abstract 

The interest for the environment among industries has increased in recent years. One 

reason is the demands from legislations and restrictions getting harsher as well as 

higher demands from customers. Emballator Ulricehamn’s Bleck is a packaging 

company with an environmental conscious. They wish to know how much 

environmental impact their products have, expressed by a Global-warming Potential 

(GWP) value. EcoProIT (2010-2013) is a project working on combining the 

increasing use of Discrete-event Simulation (DES) models with Activity-based 

costing (ABC) to investigate this. Instead of measuring monetary values the ABC and 

the DES will measure time and consumption of energy to calculate a more dynamic 

GPW value.  

 

This master thesis has evaluated the EcoProIT model by investigating the GWP value 

of 2.5l – 6l (production line 180-1) cans at Emballator. The conclusions show that the 

GWP value varies from ~2.37 GWPs for a 2.5l can to ~4,35 for a 6l can. The model 

shows that the variation for one can vary, for example a 2,5 litres varies between 

~2,19 and ~2,88 GWPs. The most significant in house contribution to the GWP value 

is waste materials and processes and it is recommended to further analyse the waste in 

the production flow. 

 

The conclusions drawn after working with the EcoProIT method is that a pre-study 

should be performed before embarking on a similar product. The pre-study should 

focus on which parts of the production system that have the highest impact in terms 

on GWP value. The study will help when identifying important parameters and 

accuracy requirements that are needed to move forward in the project as well as 

deciding how and what parts of an existing DES-model built with production 

objectives in mind. DES-model need to be modified. This master thesis recommends 

that the EcoProIT method is evaluated on an existing DES model.  



 

 

  



 

 

Table of Content 

1. Introduction ...................................................................................................... 1 
1.1 Background ......................................................................................................................... 1 
1.2 Purpose and objective .......................................................................................................... 2 
1.3 Problem formulation ............................................................................................................ 2 
1.4 Delimitations ....................................................................................................................... 3 
1.6 Factory description .............................................................................................................. 4 
1.7 Disposition ........................................................................................................................... 4 

2. Theory .............................................................................................................. 5 
2.1 Global warming ................................................................................................................... 5 

2.1.1 Greenhouse gases ........................................................................................................ 5 
2.2 Life Cycle Assessment ........................................................................................................ 6 
2.3 Steel industry ....................................................................................................................... 7 
2.4 Ecolabeling .......................................................................................................................... 8 

2.4.1 Type I - Environmental labeling .................................................................................. 8 
2.4.2 Type II - Self declared environmental claims............................................................... 8 
2.4.3 Type III – Environmental declaration .......................................................................... 9 

2.5 Activity-based Costing ........................................................................................................ 9 
2.6 Discrete-event Simulation ................................................................................................... 9 

2.6.1 Demands on input data for a traditional DES project ............................................... 10 
2.6.2 Input data management method ................................................................................. 11 

3. Method ........................................................................................................... 15 
3.1 Description of the EcoProIT method ................................................................................. 15 
3.2 Method for evaluating the EcoProIT method .................................................................... 18 
3.3 Conduction of the EcoProIT method ................................................................................. 21 

3.3.1 Concept ...................................................................................................................... 21 
3.3.2 Data gathering ........................................................................................................... 22 
3.3.3 Modelling ................................................................................................................... 22 
3.3.4 Verify the model traditionally .................................................................................... 23 
3.3.5 Validate the model traditionally ................................................................................. 23 
3.3.6 Verify the calculations for environmental impact ...................................................... 24 
3.3.7 Validate the calculations for environmental impact .................................................. 24 

4. System description .......................................................................................... 25 
4.1 Information flow ................................................................................................................ 25 
4.2 Production flow ................................................................................................................. 25 

4.2.1 Surface treatment ....................................................................................................... 27 
4.2.2 Main production line .................................................................................................. 27 
4.2.3 Post production .......................................................................................................... 28 
4.2.4 Lid production ............................................................................................................ 28 
4.2.5 Bottom production ...................................................................................................... 29 
4.2.6 Ear production ........................................................................................................... 29 
4.2.7 Buffer and storage areas ............................................................................................ 29 

4.3 Energy flow ....................................................................................................................... 30 
4.4 Data gathering and data to information transformation ..................................................... 30 

4.4.1 Process data ............................................................................................................... 30 
4.4.2 Material data .............................................................................................................. 31 
4.4.3 Order, contracts, transportation and buffer data ...................................................... 32 
4.4.4 Energy consumption ................................................................................................... 33 
4.4.5 Distribution factors .................................................................................................... 34 

4.5 Model description .............................................................................................................. 34 
4.5.1 Enamel process .......................................................................................................... 35 
4.5.2 Printing press ............................................................................................................. 35 



 

 

4.5.3 Can production line ................................................................................................... 36 
4.5.4 Bottom and lid ............................................................................................................ 36 
4.5.5 Validation ................................................................................................................... 36 
4.5.6 Excel interface ............................................................................................................ 37 
4.5.7 Delimitations of the model ......................................................................................... 37 

5. Result ............................................................................................................. 39 
5.1 Generic Result from Field Notes ....................................................................................... 39 

5.1.1 Time Comparison ....................................................................................................... 39 
5.1.2 Result Accuracy .......................................................................................................... 40 
5.1.3 Possibility to update and maintain model .................................................................. 41 

5.2 The Emballator production system and EcoProIT method................................................ 41 
5.3 GWP result from the model ............................................................................................... 43 

5.3.1 In house contributions ................................................................................................ 44 
5.3.2 Material GWP contribution ....................................................................................... 45 
5.3.3 Waste Contribution .................................................................................................... 46 

6. Analysis .......................................................................................................... 49 
6.1 Problem identification of the EcoProIT method ................................................................ 49 

6.1.1 Suggested improvements ............................................................................................ 49 
6.2 Analysis of GWP result ..................................................................................................... 52 

6.2.1 Analysis of in house contributions ............................................................................. 53 
6.2.2 Analysis of material contribution ............................................................................... 53 

7. Discussion ....................................................................................................... 55 
7.1 Evaluation method discussion ........................................................................................... 55 
7.2 Evaluation result discussion .............................................................................................. 55 
7.3 GWP result discussion ....................................................................................................... 56 

8. Conclusions and recommendations ................................................................. 59 
8.1 Conclusions and recommendations regarding the EcoProIT method ................................ 59 
8.2 Conclusions and recommendations from model result ...................................................... 59 

9. References ...................................................................................................... 61 

Appendix I ..............................................................................................................  

Appendix II .............................................................................................................  
 

  



 

 

Definitions and abbreviations 

ABC – Activity-based Costing 

AGV - Automated Guided Vehicle 

DES – Discrete-event Simulation 

EcoProIT method – Method developed by the EcoProIT project (2010-2013) 

EcoProIT model – The model created using the EcoProIT method 

GWP - Global-warming Potential 

ISO – International Standard Organisation 

LCA - Life Cycle Assessment 

LCI - Life Cycle Inventory 

LPG - Liquefied Petroleum Gas 

MTTF - Mean time to failure 

MTTR - Mean time to repair 

SOU - Statens Offentliga Utredningar 

SMHI - Sveriges Metrologiska och Hydrologiska Institut 

  



 

 

  



1 

 

1. Introduction 

In recent years there has been an increased focus on environmental research and 

sustainable production. Government restrictions as well as EU regulations are issued 

regularly for a variety of industries, most prominent among the car industry 

controlling the CO2 emissions etc (Cuenot 2009). 

EcoProIT is an on-going project at Chalmers and the main objective is to enable 

labelling of products and a tool for evaluating the environmental footprints during the 

lifecycle of a product. To then be able to connect several steps in the product lifecycle 

and get a final dynamic environmental impact. Different EcoProIT models should 

create a cluster or a chain as work together to support each other’s GWP calculations. 

A part of EcoProIT’s project aims to develop a new tool and methodology (the 

EcoProIT method) for environmental evaluations using DES. The thought behind the 

EcoProIT method is that is should be possible to implement on an existing DES 

model (Andersson et al 2011). 

1.1 Background 

Emballator is a market leading packaging manufacturer with strong environmental 

and quality focus. They provide metal and plastic packages to food, paint and 

technical chemistry industries. The packages range from bottles, bottle caps to cans 

and containers. Emballator Ulricehamn's Bleck, here after referred to as Emballator, 

produce cans ranging from 0,33l to 25l where the majority of the customers are in the 

paint industry. 

 

As a mean toward becoming an even more sustainable company, Emballator wish to 

label their products with environmental metrics. This would enhance their 

environmental profile among their customers as well as create and advantage against 

their competitors (Gallego-Álvarez et al. 2010). Emballator anticipates similar 

restrictions and regulations, as those in the car industry, within their own field, as well 

as their customers' field.  They view this master thesis as a proactive measure. 

Traditionally Discrete-event Simulation (DES) has been used to evaluate the 

relationship between monetary units, materials, time and other resources (Banks 2004, 

Banks 1999). With an Activity-based Costing calculation (ABC), activities in an 

organization are identified and indirect and direct costs are allocated to the product or 

services in accordance to their respective consumption of these activities (Skärvad & 

Olsson). Both are traditionally used separately, von Beck and Nowak (2000) showed 

that an ABC calculation can be implemented in a DES model.  

Emblemsvåg (2001) showed that ABC calculations suit well when performing an 

environmental impact analysis (Life Cycle Assessment, LCA). This is supported by a 

recent study by Andersson et al (2011) that conclude that ABC calculations in a DES 

model is a suitable combination to perform a LCA for a manufacturing system. This 

master thesis aims to implement this in a real world scenario by using the EcoProIT 

method that is under development. 

  



2 

 

1.2 Purpose and objective 

The purpose of this master thesis is to analyse green house gas emissions, expressed 

as Global-warming Potential (GWP) to provide Emballator with relevant 

environmental data that can be used as a basis for an ecolabel on their product or in a 

product catalogue. 

This master thesis will also provide researchers from the EcoProIT project with a test 

implementation of environmental metrics in DES for enabling the development of a 

new methodology and tool for environmental evaluations using DES here after 

referred to as the EcoProIT method. 

 

The objectives of this master thesis are: 

 To find the amount of greenhouse gases produced in the current state, 

measured per product. 

 To find all sources of greenhouse gases for the product and identify the most 

significant sources, contributing with at least 80% of the total GWP value. 

 To use and evaluate the EcoProIT method.  

1.3 Problem formulation 

This master thesis will answer the three main-questions below. To help Emballator 

improve their GWP value this master thesis will also answer sub question 1a. For 

Emballator's environmental commitment to reach out to their customers it is 

important that they (Emballator) know how they can and are allowed to use the GWP 

values. 

 

MQ1 How much greenhouse gases (expressed in GWP) are omitted to produce one 

product (can + lid)?   

 

SQ1 What can Emballator do to reduce their greenhouse gas emissions? 

 

There is no current method to follow when working with environmental factors and 

DES. This is one of the questions EcoProIT is trying to solve. The insecurities 

regarding the EcoProIT model is investigated by answering the sub-questions. 

 

MQ2 Is EcoProIT’s method suitable when investigating GWP impact? 

 

SQ2a What are the differences between the input data management for the 

EcoProIT method and the data for a LCA or a traditional DES?  

 

SQ2b What are the difficulties when verifying and validating a model created 

with the EcoProIT method in terms of environmental aspects? 

These questions will be investigated with the focus on the following areas: 

 Time consumption 

 Result accuracy   

 Possibility to maintain and update the model.  

 

  

  



3 

 

1.4 Delimitations 

The following points define the general delimitations for this master thesis. 

 This thesis will consider all activities inside the factory until the products 

leave Emballator (gate to gate). Environmental metrics from preproduction, 

scrap in production, usage and recycling and the end of the product life cycle 

will be taken into consideration but will not be investigated further.  

 Measurements for energy consumption on the machines/processes were done 

during the spring 2012 and are assumed to be valid for the whole year.  

 DES model will only analyze products produced at the 180-1 production line 

at Emballator. 

 The DES-model will be based on process and production data but will not be 

used to make production analysis or consider economic factors. 

 The EcoProIT method is still under development and the description available 

so far will be followed. Changes to the method during the time of this thesis 

will be taken into consideration if they are possible or applicable. 

 The EcoProIT method will be evaluated by execution, focusing on the 

questions in the problem formulation (SQ2a and SQ2b). 

 Comparisons between the EcoProIT method and similar or related methods 

will only regard SQ2a and SQ2b. 

 

Delimitations concerning the model will be further described in chapter 4.5.7 

Delimitations of the model. 

  



4 

 

1.6 Factory description 

Emballator's factory in Ulricehamn produces a wide range of metal cans in different 

sizes. The cans are slightly conical to improved decrease storage volume and stack 

ability. There are many ways to design the cans to fit customer’s needs. For example; 

handles or no handles, a range of different lids and the printing on the cans can be 

specific for each batch.  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Emballator manufacture several different can sizes. The size focused on in this master 

thesis has the top diameter 180mm and is a conical can. The line where the cans are 

produced is one of two manufacturing this size, 180-1 and 180-2. Depending on the 

height of the can different volumes are achieved. The 180-1 production line 

manufacture 2.5l, 3l, 4l, 5l and 6l cans. The cans are produced from metal sheets 

stored in a raw material buffer. The sheets are processed through an enamel paint 

machine and a printing press before stored in the printed sheet buffer. The sheets are 

stored in this buffer until Emballator receives an order for those cans. The sheets are 

moved to the 180-1 production line where they are cut and formed into cans. A 

bottom and an optional gripping wire are attached. The system will be described in 

further detail in chapter 4. System description. 

 

1.7 Disposition 

This master thesis will continue with a theoretical framework describing important 

aspects for the purpose of this project.  Thereafter the EcoProIT method will be 

described. This will be followed by a system description, which includes the 

information flow, the production flow, the energy flow, and a model description. 

After that the results from the EcoProIT model will be presented followed by an 

analysis of the result and evaluation of the EcoProIT method. The discussion will 

handle the above-mentioned chapters and which results in the conclusions and 

recommendations. 

 

  

Lid 

Bottom 

Gripping wire 

Ear Mantel area 

Plastic handle 

Figure 1 The different components of a can 

that will be discussed in the report 



5 

 

2. Theory 

To be able to answer the questions asked in the introduction a theoretical framework 

will be presented in this chapter. The theoretical framework consists of theory about 

global warming, life cycle assessment, steel industry, Activity-based Costing, 

Discrete-event Simulation and input data management 

2.1 Global warming 

The greenhouse effect is a process where the thermal radiation from the earth’s 

surface is absorbed by the atmospheric greenhouse gases and bounced back to the 

surface of the earth. Increases in concentration of greenhouse gases in the atmosphere 

have lead to what is referred to as global warming. Global greenhouse gases due to 

human activities have grown with 70 % between 1970 and 2004. Since the early 20
th

 

century the Earths average temperature has increased by 0.8°C and about two thirds 

of the increase has occurred after 1980 (Climate change synthesis report 2007). 

2.1.1 Greenhouse gases 

There are several gases that contribute to the greenhouse effect. The most significant 

are: 

 Carbon dioxide (CO2) 

 Methane (CH4) 

 Nitrous oxide (N2O) 

 

Each of these have different ability to reflect thermal radiation back to earth and also 

different residence times. Because of the different residence times it is impossible to 

find a measurement to compare greenhouse effect potential of the gases without 

defining the time horizon.  (Lashof, Ahuja 1990) The GWPs are therefore calculated 

with different time interval, commonly 20, 100 and 500 years. The residence time is 

not always known and therefore the values should not be considered exact. GWP is 

expressed as a factor of carbon dioxide, which has a GWP of one as per definition. 

See Table 1 below with GWP calculations with different time interval for some 

common greenhouse gases. (Climate change synthesis report 2007) 
 

Greenhouse gas GWP value 

Common name Chemical formula 20 years 100 years 500 years 

Carbon dioxide CO2 1 1 1 

Methane CH4 72 25 7,6 

Dinitrogen oxide N2O 289 298 153 

CFC-11 CCl3F 6730 4750 1620 

Carbon Tetrachloride CCl4 2700 1400 435 

Methyl Chloroform CH3CCl3 506 146 45 
Table 1 GWP values for some greenhouse gases (Forster et al, 2007). 

 

  



6 

 

2.2 Life Cycle Assessment 

Life Cycle Assessment (LCA) is a systematic tool to evaluate the impact on the 

environment of a product, process or activity. LCA takes the whole life cycle in to 

consideration, from cradle to grave (Curran 2004). It includes material extraction, 

manufacture, usage and end of life. An LCA does not only consider CO2 but all kinds 

of environmental emissions such as atmospheric emissions, waterbourne wastes, solid 

wastes etc. See Figure 2 below.  

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

The principles and framework for conducting an LCA is described by the 

International Standard Organisation (ISO), in ISO 14040. In the ISO framework there 

are four steps to conduct an LCA described in ISO 14041-14044. 

 

ISO 14041 Goal and scope definitions – Define and describe the product, process 

or activity. Establish the boundaries and the context in which the 

assessment is to be made. Find which environmental effects will be 

reviewed. This is an important phase and will have a strong influence 

on the result.  

ISO 14042 Inventory analysis – Identify and quantify air emissions, solid waste 

disposal and waste water discharges. This stage is described in ISO 

14042. 

ISO 14043 Impact assessment – Assess the potential effects of the releases 

identified in ISO 14042.  

ISO 14044 Interpretation – Evaluate the results from the inventory analysis and 

the impact assessment. Consider the uncertainties and assumptions 

used to generate the results. (Rebitzer et al 2004).   

Raw material 

acqusistion 

Manufacturing 

Use/resuse/ 

maintanance 

Recycle/Waste 

management 

Raw material 

Energy 

Inputs Outputs 

Atmospheric emissions 

Waterbourne wastes 

Solid wastes 

Coproducts 

Other releases 

System boundary 

Figure 2 Input, output and 

system boundary in an LCA 



7 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 3 Life Cycle Assessment Framework - phases of LCA (Rebitzer et al 2004). 

According to Curran (2004) LCA faces three key barriers. 

1. Lack of awareness of the need to look holistically at the overall impacts of 

actions.  

2. Difficulties finding reliable and publicly available data.  

3. Lack of an agreed upon life-cycle impact assessment model.  

2.3 Steel industry 

Iron is a common element in our environment and essential for many living things. 

Most of the iron is compound in the crust of the earth as magnetite (Fe3O4) and 

hematite (Fe2O3). Iron takes part of a natural cycle where it transforms through 

different compounds. When iron is removed from natural steel cycle and processed to 

steel it enters into the technical steel cycle. The steel is then manufactured into a 

product and when that product is used up or in another way consumed it can be 

recycled and made into new products. There are two ways for the steel to exit the 

technical steel cycle. It could be too polluted with other materials to be used in 

existing industry processes or it can go back to the natural steel cycle through 

corrosion (Widman 2001). 

Today about 35-40 % of all steel is produced from recycled steel. To produce steel 

from recycled material is a an easier and less energy consuming process than using 

iron ore. To produce one ton of steel from iron ore consumes 23 GJ while only 7 GJ is 

consumed to make the same amount out of recycled steel (Widman 2001). 

It is difficult to find figures for the national recyclability for steel cans and even if it 

were possible to find these it would be impossible to separate Emballator’s product 

from the rest. However World Steel Association (2011) has calculated that about 

68 % of steel cans where recycled in 2007 worldwide. According to SOU, 62 % of all 

steel was recycled in 1999 (SOU 2001 – 102). The figures from SOU are calculated 

as recycled steel divided by the amount of steel products produced that year. 

Interpretation 

Goal and 

scope 

definition 

Inventory 

analysis 

Impact 

assessment 

Direct applications: 

 

 Product development and 

improvement 

 Strategic planning 

 Public policy making 

 Other 

 

Other aspects: 

 

 Technical 

 Economical 

 Market 

 Social 

 



8 

 

According to Jernkontoret's research on the steel cycle, a lot of the application and 

products using steel, such as bridges and cars, are used for a much longer time, than 

for instance a can, before they are recycled. If the median time until recycling, for 

steel products, is twenty years the amount of steel recycled that year should be 

compared to the production of steel products twenty years ago. Since the total 

production of steel today is higher than it was twenty years ago it is impossible to 

reach 100 % by using the same calculations as SOU (Ekerot 2003).  

2.4 Ecolabeling 

Ecolabeling is attracting more and more companies who see this as an essential factor 

to consider in their industrial and commercial strategies. In 1998 and 1999, ISO 

standardized ecolabeling practises and adopted the ISO 14020 series in which three 

different types of ecolabels are proposed (Lavalle & Plouff 2004). 

2.4.1 Type I - Environmental labeling 

Type I environmental label is standardized by ISO 14024 and the goal is to identify 

overall environmental performance of a product or service within a particular product 

or service category. This performance should be based on life cycle considerations.  

Many countries, including Sweden has adopted this environmental labelling type. Al 

type I labels include two steps. First a committee of the ecolabel program establishes 

a set of minimum requirements needed to obtain the label. The second step is that 

companies are given a certification to use the label on products that fulfil the 

requirements (Lavalle & Plouff 2004). 

2.4.2 Type II - Self declared environmental claims 

Type II environmental label is standardized by ISO 14021 and describes the 

environmental claim as an “environmental declaration made without certification 

from an independent third party, on the part of manufacturers, importers, distributers, 

retailers or any other entity able to gain benefit from this declaration”.  The goals for 

type II of environmental labels are to promote environmental performance, to reduce 

inaccurate claims, to decrease confusion, to facilitate international trades and to allow 

customers to make informed decisions (Lavalle & Plouff 2004). 

In the ISO standard a set of requirements are established which must be followed.  

These requirements say that the information must be accurate and not misleading. It 

also says that the information must be true for the finished product as well as it must 

also take the life cycle into account. This is to identify a potential increase in an 

environmental impact pursuant to the decrease in another (Lavalle & Plouff 2004). 

Frequently used terms in this kind of declaration are: 

 Compostable 

 Degradable 

 Designed for disassembly 

  



9 

 

2.4.3 Type III – Environmental declaration  

Type III product declaration is described in ISO 14025. The declaration consist of 

environmental information such as percentage of recycled material, information on 

toxic substances and other information about a product´s environmental impact on a 

simplified performance report card. This declaration does not usually contain 

comparative claims but the information in the declaration may be used for such a 

comparison. If the information should be used for comparative claims the ISO 

standard 14040 must be followed (Lavalle & Plouff 2004). 

The information on a type III environmental declaration must be based on procedures 

and results from a quantified life cycle assessment compliant with ISO 14040 

standards.  For many small and medium size businesses a complete life cycle 

assessment is too expensive and requires too much time investment (Lavalle & Plouff 

2004). 

2.5 Activity-based Costing 

ABC is a financial method to assign costs of activities or resources in an organization 

to all products or services based on the actual consumption by each. In ABC two 

types of costs are identified. These are: 

 Direct costs – Costs that can be traced directly to a product or service. 

Examples of these costs are material costs and machines that are only used for 

one product.  

 Indirect costs – Costs that cannot be traced directly to a product. These costs 

are also called overhead costs. Examples of these costs are resources used for 

more than one product or services such as forklifts and heating of facility.  

 

To get the true cost for each product or service the indirect costs has to be divided 

among the products or services manufactured. Cost drivers are identified to distribute 

the indirect costs. It is important to identify the correct cost driver and this is 

sometimes difficult since there may be several causes for one indirect cost. (Skärvad 

and Olsson 2008) 

2.6 Discrete-event Simulation 

Discrete-event Simulation (DES) is used to model the real world or a conceptual idea 

that is to be built or created in the real world. The models are often built in the image 

of production systems where the need for testing before implementing is needed or 

where system/process improvement is needed. DES can also be used to improve other 

areas such as healthcare, military or service sectors (for example hospitals and 

restaurants). The difficulties of predicting how a system of processes will behave and 

how they dynamically affect one another is one of the benefits of using DES (Banks 

1999). 

Traditionally Banks method is used when working with a simulation project, there are 

several steps included. Two of the steps are especially relevant for this project and are 

discussed in more detail in the two following subchapters. For further reading and 

information about the other steps and on Banks method please see Banks (1999).  

  



10 

 

2.6.1 Demands on input data for a traditional DES project 

There is variety of issues concerning input data for DES projects. One in particular is 

the time aspect. According to Skoogh and Johansson on average 31% of project time 

is spent on input data management and according to and Trybula (1994) between 10 

and 40%. The variations in time depend on the share of each category of data that 

needs to be collected. Following Robinson and Bhatia (1995) example one can divide 

the data into three different categories (Table 2).  

 

Category Type of Data 

A Available data 

B Not available but collectable data 

C Not available and not collectable data 
Table 1 (Robinson & Bhatia 1995) 

Category A represents already available data such as previous time studies or from 

automated logging systems. Category B represents data that needs to be collected for 

example through time studies. Category C is neither available nor collectable and 

needs to be estimated. The estimations have to be carefully done for the sake of the 

model quality. Depending on the data composition in terms of the three categories 

time spent working with the data differs. Only 7 % of companies have all data 

available for DES projects (Skoogh and Johansson 2007). Perera and Liyanage (2000) 

identified the major causes of inefficient data collection, these are:  

 Incorrect problem definition 

 Lack of clear objectives 

 High system complexity 

 Higher level of model details 

 Poor data availability 

 Difficulty in identifying available data sources 

 Limited data handling capacity  

For further reading on major causes of inefficient data collection see Perera and 

Liyanage (2000).  

According to Skoogh and Johansson (2007) the most time consuming activities are:  

 Data collection 

 Mapping of available data 

 Data analysis and preparation.  

In addition their study showed that only 20 % of DES projects finished their input 

data management according to set plans. The reason for this is that the company 

where the DES project is performed often overestimate the usability of the data or is 

unfamiliar with what type of or amount of data that is generally needed in DES 

projects. This leads to unforeseen work because of re-evaluations, calculations and 

additional measurements (Skoogh and Johansson 2007). 

The quality of the model is dependent on the quality of the data, which increases the 

need for efficient input data management. As the amount and complexity of the data 

grows so does the need for structure and order in the input data management (Skoogh 

and Johansson 2008).  



11 

 

2.6.2 Input data management method 

Skoogh and Johansson (2008) developed a structured methodology for the input data 

management process. The methodology consists of 13 steps, see figure 4, and is a tool 

to decrease the time and secure the quality of the data management. The method is 

especially important for those with limited experience of DES projects (Skoogh and 

Johansson 2008).  

 

 
A. Identify and define relevant 

parameters.

B. Specify accuracy requirements.

C. Identify available data.

D. Choose methods for gathering 
of not available data. 

E. Will all 
specified data 

be found?

No

F. Create data 
sheet.

Yes

G1. Compile available data.
G2. Gather not available 

data. 

H. Prepare statistical or empirical 
representation.

I. Sufficient 
representation

?

No

J. Valid data representation 

Yes

K.Validated?

No

L. Finish final 
documentation.

Yes

 
Figure 4 Input data management (Skoogh & Johansson 2008) 



12 

 

 

A. Identify and define relevant parameters  

As the headline implies one should identify and define relevant parameters with 

regard to project objective. It is important to take level of detail into consideration as 

well as the system complexity to find a appropriate compromise. This should be 

accomplished by getting to know the system through experienced personnel with 

system knowledge or company production technicians, or both. With more system 

knowledge it is possible to define how each parameter should be measured to best be 

represented in the model. When measuring it is important to define, when or where, to 

start and stop to the measuring (Skoogh and Johansson 2008). 

B. Specify accuracy requirements  

The purpose and objective of the projects together with the knowledge of the system 

to be modeled will have to be guiding when deciding which accuracy level is needed 

on the different parameters. A parameter with less influence on the result can be less 

detailed than one important for the result. This can at a later stage be approved or 

disapproved by a sensitivity analysis. Skoogh and Johansson (2008) recommend this 

for all borderline cases. To get a good representation of a process a lot of data needs 

to be collected. The more variability a process has the more data is needed, in case of 

processes that are fairly constant, such as robot cycle time etc, only enough data to 

rule out any unexpected variability is needed (Skoogh and Johansson 2008).   

C. Identify available data 

Identify which category (A, B or C in 2.6.1 Demands on input data for a traditional 

DES project) the desired data belong to, to make the most of all available data. In 

industry a lot of processes are measured and data collected, but not intended for a 

DES project. Therefore many companies believe they have enough and sufficient data 

for a DES projects when this is not the case. It is important to make sure that it is 

possible to collect the data and that it is on the right form or possible to get the correct 

form through calculations (Skoogh and Johansson 2008). 

D. Choose methods for gathering not available data  

Some of the data in DES projects are almost always not available, either category B 

or C. In the case of category B data case time studies, video recordings and other 

types of data collection is necessary. When time studies are made with stopwatches 

and similar it is important to define where the process measured starts and ends as 

described in step A. Identify and define relevant parameters, this is especially 

important if several people are measuring. Category C data is most common in not yet 

existing processes or systems and needs to be estimated with care. Time studies and 

video recordings of existing similar processes can be used for these estimation 

(Skoogh and Johansson 2008).  

E. Will all specified data be found? 

“Found” in this step refers not only to finding data but to make sure the data satisfies 

the demands made in the previous steps B, C and D. If the demands, such as number 

of data points, data accuracy or data quality are not satisfied problems may occur 

further down the line. According to Skoogh and Johansson (2008) step I and K are 



13 

 

especially vulnerable to this and would result in less quality model. If the demands 

cannot be fulfilled, the need for future iterations is likely 

F. Create data sheet  

Skoogh and Johansson (2008) recommend that a data sheet is created where all raw 

data as well as all analyzed data is kept. It is discouraged to keep the analyzed data in 

the interface spreadsheet and the raw data in a temporary document. This less 

structured method is more time consuming since there is a possible risk of data loss 

(Skoogh and Johansson 2008). 

G1. Compile available data 

The category A data is collected as specified in step C and the amount of data 

specified in step B. Usually the data needs to be processed and additional efforts made 

to sort, filter, calculate and convert the data to a desirable form. This to prepare the 

data for the coming statistical or empirical representation in step H.  If the data is 

previously analyzed it need only be validated as described below in step K. (Skoogh 

and Johansson 2008). 

G2. Gather not available data 

In this step the focus is to make category B or C data category A (Robinson and 

Bhatia 1995).  

Category B data that is to be collected is a time consuming task (Skoogh and 

Johansson 2008), especially if they are gathered from a low frequency system or the 

product variability is high (high cycle time variability or long cycle time). Time 

studies on operators are controversial and this is usually solved with video cameras. 

However this means a doubling in real time measurements to collect data. 

To gather category C data is less time consuming. Category C data should to the 

furthest extent be based on assumptions and estimations made by a process expert or 

highly knowledge of the system. Gathering category C data can be time consuming if 

it is based on similar systems or processes or historical data (Skoogh and Johansson 

2008).  

When both step G1 and G2 are done it is possible to move onto step H. 

H. Prepare statistical or empirical representation  

For the data to be implemented in the simulation model the raw data prepared in G1 

and G2 need to be represented by statistical or empirical distribution, traces or 

bootstrapping (Robinson 2004).  

I.  Sufficient representation? 

In this step the distributions from step H. are evaluated. There are several ways to do 

this, for example with a goodness-of-fit test. Goodness-of-fit test can be difficult to 

pass for large number of samples. Skoogh and Johansson (2008) suggest graphical 

comparison of the representational and original data. This given that the accuracy 

requirements regarding level of significance are fulfilled (step B.). This can be 

ensured at a later stage with a sensitivity analysis for the parts where less satisfying 

results are present for the graphical comparison (Skoogh and Johansson 2008).  



14 

 

If the accuracy requirements are not fulfilled, iteration from the steps G1 and G2 

might be necessary. In the worst case scenario the iteration has to start at step B, 

which could have significant repercussions for the simulation model (Skoogh and 

Johansson 2008). 

J. Validate data representation 

According to Sargent (2010) lack of data validation is most commonly the reason 

model validation fail. It is also a costly and time consuming process. A validation of 

the data can save time and further iterations during the model validation (Skoogh and 

Johansson 2009). It is a difficult process to attain adequate validation quality data. 

The difficulties lies with the fact that the data for the conceptual model and project is 

based on is the same data it will be validated against (Sargent 2010). Both Sargent 

(2010) and Skoogh and Johansson (2008) recommend structure in the data collection 

process as the best guideline to avoid validation problems. 

Skoogh and Johansson (2008) recommend that to reach a good enough validity the 

data collection process should be performed alongside discussions with process 

experts. A final check is also advised towards the end where the data is reviewed in a 

structured way. 

K. Validated? 

When the data is validated it is ready to be incorporated in the model. The validated 

data can still cause problems in the model validation. If this problem occurs it is 

important to reevaluate the data to find the reason for the problem. Skoogh and 

Johansson (2008) conclude that a lot of the problems occur due to miscalculations in 

step H but that the problem might stem as far as choice of gathering method, step C 

and D. (Skoogh and Johansson 2008). 

L. Finish final documentation 

This entire method for data input management is focused on documentation, therefore 

most of the data should at this stage already be documented. To facilitate further 

studies and future projects as well as maintaining the validation some additional 

documentation is needed, such as (Skoogh and Johansson 2008): 

 Gathering methods 

 Sources of data 

 Validation results 

 Assumptions made 

 

This should be compiled in a data report and complete data sheet (Skoogh and 

Johansson 2008) 

  



15 

 

3. Method 

3.1 Description of the EcoProIT method 

EcoProIT 2010-2013 is an ongoing project in the Product and Production 

Development department at Chalmers University of Technology (Chalmers PPD).  

As a step towards this master thesis goal, the method used is the EcoProIT method. 

Parallels can be drawn between this method and Banks method traditionally used for 

DES projects but there are some important differences. For more information about 

Banks methodology, see Banks (2004). Among other things Banks model does not 

facilitate the verification and validation of environmental aspects (Banks 2004). The 

description below of the EcoProIT model is a summation from unpublished 

documents by the EcoProIT project.  



16 

 

2. Concept

3. Collect data 

4. Data to 
information 

transformation

5. Modelling

6. Verify the 
model 

Traditionally

7. Validate the 
model 

Traditionally

8. Verify the 
environmental 

calculations

9. Validate the 
environmental 

calcualtion

10. Analyse the 
model

12. Implement 
into 

organization

11. Report

Good and 
enough 

information?

Need more 
information?

Verified?

Validated?

Sensitivity 
analyse

Consumption 
validation, 

Reviews

Good 
enough?

Calculation and 
assumptions ok?

Concept model,
Pre-data collection
Logic definitions, 

detailes specification

1. Set up project
Set goals,

system bounderies, 
functional unit, etc.

 
Figure 5 The EcoProIT method 

1. Set up project 

The goal definitions are a crucial part of a project, when these are set satisfactory the 

goals will help decide the level of detail for the project. They will be useful when 

defining the system boundaries for the environmental analysis. As a result of the 

system boundaries and the goals it is possible to define the limitations of a project. 

2. Concept  

A conceptual model is a schematic showing the system logic and flow. To develop 

and verify this, close collaboration with the people in day to day contact with the real 

system is needed, discuss all assumptions made. The conceptual model develops 

throughout the project to coincide with the level of detail needed to fulfil the previous 

set goals. The system boundaries are clarified. A part of creating a conceptual model 



17 

 

and moving forward in the project is to define where and what type of data that has to 

be collected. The concept phase and data gathering is closely linked.  

3. Data gathering 

With the conceptual model as a base it is possible to identify the data needed for the 

model. The data can be categorized as process data, material data, energy 

consumption data and energy content data. 

 

Process data: Logged data, interviews and measured data, much like in a 

traditional DES project. 

 

Material data: What materials are used, material histories, solvents paints etc. 

 

Energy consumption data: How much electricity, Liquefied petroleum gas 

(LPG), oil, compressed air is consumed etc.? 

 

Life Cycle Inventory (LCI)-data: Emissions for the electricity used, LPG 

content etc. 

It is possible to gather data and start building the structure of the model 

simultaneously although, it is necessary to gather some amount of production data 

before it is possible to start with the model. This is a step towards a complete 

conceptual model and important not to make unnecessary mistakes (or waste time). 

4. Transformation of data to information  

Usually data straight from a log or measured cycle time need to be filtered, sorted, 

calculated converted etc. before it can be used in the simulation model. It is useful to 

represent information, energy and production flows etc. schematic as a complement to 

the conceptual model. This step includes calculating the different distributions 

necessary to create a dynamic model. 

Is the Information enough to implement the model? This question should be possible 

to answer with the help of the goals and the conceptual model. If the answer is no, 

more data need to be collected. 

5. Modelling 

The model can be built with software specially developed for DES models or with 

ordinary languages such as C++. It is important that the code is structured and built in 

layers with a traditional DES as a foundation and the environmental aspects built as 

the outer layer. This will facilitate both verification and validation at a later stage as 

well as make the simulation model more flexible (Banks 1999). 

6. Verify the model traditionally 

The verification ensures that the model is working as the conceptual model, following 

its logic laws etc. It is also done to ensure that the code is running correctly, no 

infinity loops etc. The verification of the code is closely connected to the modelling 

since the verification needs to be done continuously through the modelling. 

7. Validate the model traditionally 

To validate a model is essentially to make sure that it corresponds to reality. It is also 

essential to be able to trust the models final result enough to use it for an analysis. 



18 

 

There are a number of steps that can be taken to ensure this. Through historical data 

validation, does the model perform according to realty with historical input data 

compared to historical output data? Does the model perform according to reality 

during extreme conditions such as breakdowns, overload or under load of the system? 

A Turing test can be performed, output data from reality and output data from the 

model is compared by someone familiar to the system.  

8. Verify the Calculations for Environmental Impact 

It is important to make sure the model level of detail corresponds to the goals set. This 

is done by a sensitivity analysis. A sensitivity analysis makes sure that the data, from 

the sources that have the most impact on the products, is detailed enough and 

modelled in corresponding level of detail. An analysis should be performed on the 

sources that together stand for 80% of the GWP impact. 

9. Validate the Calculation for Environmental Impact 

To validate the environmental part of the DES let a certified reviewer review/validate 

the analysis. It is possible to compare the result to similar products with similar 

analysis, keep in mind the different methods (dynamic static and so on) and 

differences. Another possible way to validate is to compare a cell’s or process’ used 

consumables to the same in the model. It is important to define the limitations for the 

analysis and document these together with the analysis. Otherwise the environmental 

impact analysis becomes invalid (changes etc.). 

10. Analyse and use the Model 

The model is analysed so that the current state is known. This is important when 

evaluating the different improvement suggestions modelled. 

11. Communicate the Results 

When all experiments and analysis are done and documented the information 

regarding and answering the projects goal should be compiled. The correct 

information should be communicated to correct stakeholders. 

12. Implementation in organisation 

The EcoProIT method is still under development and has not yet defined guidelines 

for how to implement in organisation. The thought is to be able to use both the 

production analysis and production changes to also see how the GWPs are affected. 

The case study in this master thesis is a part of the development of the EcoProIT 

method. 

3.2 Method for evaluating the EcoProIT method 

Evaluation of the EcoProIT method has been done continuously throughout the 

master thesis work. To have an organized workflow of the evaluation an affinity 

diagram has been made. This is also called the KJ-method after its originator Jiro 

Kawakita. Below is the method described as by Bergman and Klevsjö (2003). 

 

 The first step is that all kinds of ideas and thoughts are written down on small 

cards or “post it” notes. Preferably this is done in brainstorming sessions. It is 

important that everyone is allowed to have his or her say.  

 



19 

 

 All notes are gone through to make sure that everybody agrees on what the 

text says. If necessary new notes are written that better represent the essence.  

 

 The notes are then placed in different groups. There may be notes that do not 

fit together with any other notes and these are then called “lone wolves” or 

“vagrants”.  

 

 A headline is decided for each group. The headline should in some sense sum 

up the group.  

 

 Arrows are drawn between the groups to illustrate the relationship between the 

groups. The arrows are moved around until the group members agree that they 

describe the connections correct.  

 

Some adjustments had to be made to the method to fit the purpose of this master 

thesis. Since the ideas and thoughts have come up during the whole project time the 

affinity diagram has been improved and reworked continuously throughout the 

project. Discussions have also been held with members of the team working with the 

EcoProIT project.  



20 

 

2. Concept

3. Collect 
data 

4. Data to 
information 

transformation

5. Modelling

6. Verify the 
model 

Traditionally

7. Validate the 
model 

Traditionally

8. Verify the 
environmental 

calculations

9. Validate the 
environmental 

calcualtion

10. Analyse 
the model

12. Implement 
into 

organization

11. Report

Good and 
enough 

information?

Need more 
information?

Verified?

Validated?

Sensitivity 
analyse

Consumption 
validation, 

Reviews

Good 
enough?

Calculation and 
assumptions 

ok?

Concept model,
Pre-data collection
Logic definitions, 

detailes specification

1. Set up 
project

Set goals,
system bounderies, 
functional unit, etc.

Analysis of 
input data 

managment

Comparison 
of time 

consumption

Analysis of 
traditional 
verification 

and validation

Identify 
possible errors 
and difficulties 

with 
environmental 

verification 
and validation 

A.

C.

B.

D.

Identify 
problems 
with the 
EcoProIT 
method

Possible 
solutions

EcoProIT method

 
Figure 6 Method for evaluating the EcoProIT method 

 

  



21 

 

The focuses of the evaluation were, in accordance with the objective, input data 

management and verification and validation. These two areas are closely linked and 

one affects the result of the other which is represented in Figure 6. The evaluation 

have at times spilled into other areas as well, these have been presented if they are 

relevant for the EcoProIT method or connected to the focus areas mentioned in the 

problem formulation. 

When step three was reached in the execution of the EcoProIT method the time 

consumption of the environmental part of the input data management were 

documented for future reference. This step is represented by arrow A in Figure 6. 

There are several projects performed evaluating for example time consumption 

(Skoogh and Johansson 2009) and input data management (Skoogh and Johansson 

2008) on DES projects. In combination with the theory gathered and previous 

experience of DES projects a comparison was made. It was possible to compare the 

time spent collecting and processing data in this project with the evaluations 

previously done. 

The analysis of the input data management depends in part on the data validation and 

the general input data management method used in the EcoProIT method, represented 

by arrow A in Figure 6. It also depends on the result from the verification and 

validation analysis (and error identification and difficulties). Therefore the analysis 

cannot be completed without feedback from this part. This is represented by arrow D 

in Figure 6. 

In turn the analysis of the verification and validation depend on the result from the 

Input data management analysis, represented by arrow B in Figure 6. 

When the analyses were done, their combined result highlights problems within or 

relating to the focus areas. Solutions where sought for these problems. 

The result of this evaluation will be a time chart and identification of possible errors. 

The analyses will be presented in chapter 6. Analysis together with suggested 

improvements to the EcoProIT method.  

3.3 Conduction of the EcoProIT method 
In this chapter a description of how the EcoProIT method has been implemented is 

presented. The result, the EcoProIT model, refers to both a DES model and an excel 

interface for GWP calculations.   

3.3.1 Concept 

It is important to have a full understanding of the system before embarking on the 

simulation part of a project. In this master thesis the production flow is important, in 

addition the energy flow and the material flow is as relevant. Since the GWP is 

derived from all parts of the system each flow and their logic is necessary to get an 

understanding of the system. To visualize a conceptual model several softwares are 

available, in this master thesis Microsoft Office Vision 2007 (Visio) was used. In 

Visio it is possible to create several different types of schematics both for business 

and administrative flows and maps, as well as engineering flows. 

To get a good overview of the system the production leaders explain how and where 

the products and its components were transported and processed throughout the 

factory. Blueprints of the factory were also used to get a better understanding of the 



22 

 

flow. For the understanding of all the energy sources and resources used, the facility 

manager gave a thorough description of the energy flow in the factory. The 

description also included information about some of the processes such as the enamel 

and printing press ovens and the effect of solvents on the temperature in these. 

When the conceptual model of the production and material flows was created the 

production leaders could verify it. The energy flow created was verified by the facility 

manager. As a step towards a complete conceptual model the system boundaries are 

set in accordance with the master thesis goals and clarified together with the 

schematic flows of the factory. 

In the real system large stock are kept as buffers between the departments. Because of 

this the ordering system had to be studied carefully to be able to interpret this into the 

model. The enamel and printing press processes where the ones thought to cause the 

most impact and was focused on for collecting data. The reason for this is the use of 

liquefied petroleum gas (LPG) and the burning of the enamel in the ovens. Therefore 

all the above aspects (material, energy consumption, production process) were taken 

into consideration. 

3.3.2 Data gathering 

Expert fit was used to find the statistical distribution for a collection of data points, at 

times empirical data was used.  

Data gathering should be as objective as possible and should observe the process 

without affecting it. Methods often used are filming, interviews, observing, time 

studies etc., see chapter 2.6.2 Input data management method. In this master thesis the 

existing production data was used as far as possible. At Emballator the production 

process can be uneven i.e. planned stops on the chosen line, hardly any setup on day 

to several setups and failures the next, depending on the customer demand and 

material. This made time studies diverse and data points difficult to collect. The 

production data was therefore complemented and validated with interviews and 

observations, see chapter 2.6.2 Input data management method. 

Several types of data that was gathered, therefore the approach differed. Some data 

was easily accessible in the production lines process systems, some had to be filtered, 

sorted, calculated converted etc, and others were measured. To make sure all parts of 

the facility where included data was gathered mainly for the year 2011, some of the 

earlier data does not include new additions to the factory and building (latest added 

2009). 

3.3.3 Modelling 

The model was built with a software license that was available to the master thesis, 

AutoMod 12.3.1, which is based on a general purpose simulation language. A special 

purpose simulation language will generally have fewer errors than that of a general 

purpose simulation language. On the other hand the opposite is valid for flexibility 

(Sargent 2010). The EcoProIT method is new and unfamiliar and therefore the choice 

of software was decided so that not to cause unnecessary work or disruptions to the 

schedule. Because of its flexibility AutoMod is well suited for modelling a factory as 

well as the phantom parts of the model keeping track and calculating the 

environmental aspects. 

The production part of the model was built first and divided into sub processes 

according to the conceptual model.  Each part was modelled separately and thereafter 



23 

 

added to the final model. This minimizes the time spent correcting errors in the entire 

code as well as simplifies the verification and validation process (Sargent 2010). By 

using load specific attributes in the model the loads created could be specified 

according to gathered data described in chapter 3.3.2 Data gathering. These 

specifications could be number of cans in one order, size of can, number of enamel 

layers, type of bottom etc. With the help of the attributes the loads move through the 

system as the real steel sheets, these are divided into cans at the same part in the 

system that the real cans are created. 

For the environmental part, attributes where also used to calculate the time each load 

spent in the different processes, keep track of number of plastic layers etc. This 

together with the more process specific attributes would then act as a basis for further 

calculations on GWP.  

The GWP calculations were done using excel where sheets with relevant process data, 

material data, LCI data and distributing calculated the GWP values for materials and 

processes using different excel algorithms.  

3.3.4 Verify the model traditionally 

Verification is necessary to make sure the model is behaving according to the logic of 

the conceptual model and that no logic choices are missing in the model. Verification 

of the model is also to make sure that the simulation language is used in the correct 

way. This is possible to accomplish is several ways for example by structured 

walkthroughs, trace, input output relation and using the software debugging option 

(Sargent 2010) 

One should keep in mind that errors that occur can be caused by the data, conceptual 

model or other factors regarding the program or software (Sargent 2010). Verification 

of the model was continuously done while modelling. This to make sure not to build 

on something that is already faulty and minimize time spent searching for errors. By 

tracking entities throughout the system errors where made visible and possible to 

correct. Tracking was in this case done by printing messages in different parts of the 

process. The AutoMod debugger is also helpful when tracking or in general verifying 

the program. By sending single loads or very high number of loads through the 

system errors could also be detected and corrected.  

When printing the attributes (cycle time etc) to excel, errors or unreasonable times 

were a good indication of errors in the code.  

3.3.5 Validate the model traditionally 

Validation is important to make sure that the model behaves according to reality or 

reflects that part of reality that is modelled to a satisfactory extent. In context of the 

conceptual model it is making sure the conceptual model corresponds to reality 

(Sargent 2010). 

When the data collection were made some of the data were saved, others were not 

applicable. Several of these could be of use for the system validation. For example 

number of cans produced per shift and utilization of the processes. The time spent in 

the different processes are important from a GWP stand point, therefore these were 

viewed as especially important to validate. The excel sheet were the process times etc 

were printed were useful when validating. The cycle times that were printed served as 

a good validation point and data for several orders could be compared to the real 



24 

 

system. The loads could be traced through the system and distributions and empirical 

representation could be validated (Sargent 2010). 

Where insecurities in the data or the model where discovered a sensitivity analysis 

where performed. This could help validate which factors are important from a GWP 

perspective. 

3.3.6 Verify the calculations for environmental impact 

The EcoProIt method is still under development and there is limited literature or 

instructions for verification of the environmental aspects. The LCI data were therefore 

collected and chosen together with master thesis supervisor, Jon Andersson, and its 

examiner, Björn Johansson. The material requirements were filled to as large extent 

as possible and where LCI data were missing substitutes were found that satisfied the 

requirements as far as possible. 

When collecting environmental data from the production system it is handled with the 

same procedure as the ordinary production data. Therefore also going through the 

same validation process. The data collected and measured by the electrician is 

assumed to be correct. 

To verify that the calculations are correct the GWP excel sheet where thoroughly 

inspected. This was done continuously as it grew and simultaneously as the excel 

sheet with the AutoMod result was tested and verified, to make sure the units 

corresponded. The data was also compared to see if the values were reasonable in 

relation to each other. 

As described in 3.1 Description of the EcoProIT method a sensitivity analysis should 

be performed on the sources that together stand for 80% of the GWP impact 

3.3.7 Validate the calculations for environmental impact 

The results from the model should be compared to similar products keeping different 

methods in mind. No LCA data has been found, therefore this is not possible. The 

result has been internally reviewed by three people. 

 

  



25 

 

4. System description 
In this chapter a description of the production system is presented. The description is 

complemented with an account of the information and energy flow. These three parts 

together represent the conceptual model that has been the base during this master 

thesis. 

4.1 Information flow 

The sales office establishes a contract between Emballator and a customer. These 

contracts handle questions about size, appearance and number of cans. The contracts 

can also regulate other commitments such as an obligation for Emballator to keep a 

safety stock of finished cans. After a contract is established one or several orders for 

cans will arrive to Emballator in accordance with that contract. There are two kinds of 

contracts. 

 Short term contract. Usually the customer is obligated to buy the specified 

quantity within a year. When the quantity has been shipped the contract have 

to be renewed.  

 Long term contract.  The contract specifies a yearly quantity.  

The first order of a contract is delivered within 15 days, but the following parts have a 

five days delivery guarantee. This means that large quantities of printed sheets are 

stored in the printed sheet buffer awaiting customer orders. Emballator have recently 

started contacting customers if their printed sheets have been in the buffer for more 

than a year. The customers are then obliged to either order cans or hire the storage 

area. 

The contracts are sent to the production planning office. A rough planning for the 

enamel and printing process is done by the planning office for the needed quantity. 

Depending on the size of the contract the production planning office may divide the 

contract into two or more printing operations. This is done in order to keep the stock 

of printed sheets down.  This planning is made in SAP and will trigger an in house 

order to the printing and enamel office. Adjustments of the rough planning from the 

planning office are made in the printing and enamel office to keep the set up times to 

a minimum. For example, batches with the same size, colour or enamel is done after 

each other to reduce the set up times.  

Orders for cans arrive to the production planning office. When planning the sequence 

of the order set up times as well as priority is taken into consideration. This means 

that the operators are not allowed to change the sequence of the orders without first 

asking the production leader. This is because the operators do not know if the can 

order is make to stock order or if the customer is already waiting. 

4.2 Production flow 

The cans produced at Emballator are produced in eight different lines. Each 

production line produces cans with a certain upper and lower diameter. The height of 

each batch of cans can be varied meaning that the volume of the cans can be changed 

at each line. The production line focused on in this master thesis is called 180-1 which 

produce cans with an upper diameter of 180mm and a bottom diameter of 168 mm. In 

this line five different volumes is produced; 2,5l, 3 l, 4l, 5l, and 6l. The production 

flow can be seen in figure seven below. Descriptions of the different processes will 

follow, as well as a description of the important buffer areas used within the facilities. 

 



26 

 

 

 
Figure 7 Flow for the 180-1 cans. 

  

Enamel Paint 
Machine

Printing Press
Raw Material 

Storage

1. Surface Treatment1. Surface Treatment

Enamel Paint Buffer Printed Sheet Buffer

Punch

Deep 
Drawing

3. Ears

Punch

Deep 
Drawing

4. Handles

Flow 180-1

Cut

Shape and 
weld

Seal and 
temper

Tapering

Role top and 
bottom edge

Attach 
Bottom

Attach ears

Attach 
handle

Seal

Pressure test

Temper

Cooling Buffer

Robot 
palletizing

Cooling Buffer

2A. Main Production line2A. Main Production line

Punch

Role

2B. Bottom Production

Gumming

Temper

Punch

Role

2C. Lid Production

Gumming

Temper

Storage for finished 
goods



27 

 

4.2.1 Surface treatment 

The surface treatment is illustrated in step 1 in figure 7. 

1. Metal sheets in different sizes, depending on which size of can will be 

produced, are transported to Emballator and stored until needed.   

2. The metal sheets are transported by forklift to the enamel paint machine. 

There are approximately 22 different types of enamels that can be used in the 

enamel process. The sheets go through the enamel paint machine at least one 

time. Depending on whether the sheets should have enamel paint on one or 

two sides and which enamel paint that should be applied, it may be necessary 

to run the sheets through the enamel paint machine two or three times. Every 

cycle includes: 

o One coating operation 

o Heating in oven to harden the enamel 

o Cooling transport 

3. After the enamel paint machine the sheets are stored in a buffer next to the 

enamel paint machine. 

4. Depending on the customers' needs the sheets can be printed in a printing 

press. The printing operation consist three stages. 

o Two color printing operation 

o Heating in oven to harden the color 

o Cooling transport 

It is possible to apply two colors each run through the printing process. For 

some prints it is necessary to run through the printing process up to eight 

different times. On average there are 4.2 colors in each print. After the printing 

process the sheets are transported by forklift and stored in a buffer until 

needed.  

4.2.2 Main production line 

The fully automated main production line is illustrated in step 2A in figure 7. 

5. The sheets are transported with forklift from the buffer to the can machine. 

The steps in the can machine are: 

o The sheets are cut to desired size depending on the height of the can. 

Each sheet is now corresponding to one can.  

o The sheets are welded to a cylindrical shape. To make sure the weld 

damage does not compromise the seal of the can, enamel paint is 

applied at the seam. The cans are thereafter tempered and transported 

on a conveyor that works as a cooling buffer. 

o The cylinders are made conical by applying pressure to the upper part 

of the cylinder. 

o The cans are tapered and the top rim is gradually folded to get the 

desired height and profile of the rim. The can is then turned upside 

down, to achieve the desired profile on the bottom rim, also by 

gradually folding the edge. 

o The bottom is attached by folding the rim. 



28 

 

o The ears are attached the cans by spot-welding them to the sides of the 

cans.  

o  The handles are thereafter attached to the ears. The handles are 

unrolled and shaped at a sub flow to the main line, here the plastic 

grips are attached to the handle of some orders of cans. 

o Enamel paint is applied to the welding points of the ears. 

o A pressure test is performed to make sure the cans performance is 

satisfactory. 

o The cans go through an oven to harden the enamel paint at the welding 

points. 

o After the oven the cans are cooled on a conveyor belt and then packed 

by a robot to a pallet.  

4.2.3 Post production 

6. The pallets are picked up by an automated guided vehicle (AGV) and 

transported to a station where the pallets are covered with plastics.  

7. After step 7. the pallets are transported with a second AGV to third AGV that 

transport them to the storage area for finished goods. Here they are picked up 

by a forklift and transported to the right place.  

8. Soon before the truck comes to pick up the pallets they are moved from their 

place in the storage of finished goods to an area at the pick up place. 

4.2.4 Lid production 

The lid production line is illustrated in step 2C in figure 7. 

9. The raw material for lids is metal sheets and these are surface treated with 

enamel in the same manner as step one, two and three above. On rare 

occasions they also go through the printing operation in step four. There are 

several kinds of lids to meet all kinds of customer needs. Emballator uses 

several different fully automatic lid machines and three of these are used for 

production line 180-1.  

o The metal sheets are transported with a forklift to the lid production 

machine.  

o Round pieces are punched out of the metal sheet in specific pattern to 

reduce waste.  

o Depending on what kind of lid that is produced the lids are rolled in 

different steps to the finished shape.  

o A rubber is placed on the side that will be in contact with the can to 

make sure that the lid will be able to retain he substance in the can. The 

rubber is liquefied when applied to the lid but will soon harder to a 

rubber.  

o When the lids are finished they are they are packaged and transported 

to the storage area for finished goods. 



29 

 

4.2.5 Bottom production 

The bottom production line is illustrated in step 2B in figure 7. 

 

10. The bottoms are produced in the same manner as lids. As for the lids there are 

different types of bottoms to meet the customers' needs. The only differences 

in the production flow between the bottom production and he lid production is 

that when the bottoms are finished they are stored until needed at the main 

production line.  

4.2.6 Ear production 

The ear production line is illustrated in step 3 in figure 7. 

11. There are two sizes of ears used at Emballator depending on the size of the 

cans. Both kinds are produced in the same way in similar machines. The cans 

produced in the 180-1 production line uses the smaller size of ears.  

o The steel for the ears arrives to Emballator on rolls and stored in the 

Raw material warehouse. 

o When the rolls of steel are needed a forklift transfer the ear production 

line transports them.  

o A fully automatic machine produces ears in the following steps: 

o Round pieces are punched out 

o Every piece get a hole in the middle that the gripping wires later can be 

attached to.  

o The pieces are deep-drawn to a small cup. 

o The machine spits out the finished ear to a container. 

o The container is picked up by a forklift and stored until needed at the 

line. 

4.2.7 Buffer and storage areas  

According to Emballator's purchaser the raw material storage that contains about 

9000 tons of steel (as of February 2012) and the maximum capacity (to not take up 

space intended for other purposes) is between 6000-6500 ton. Emballator's desire is to 

lower the inventory level in this area. The high raw material inventory is partly due to 

long lead times and insecurities in deliveries, since the raw material is purchased from 

Asia. 

In comparison to the raw material buffer there is a small buffer area for the sheets 

processed in the enamel machine. These are processed regardless of order and works 

similar to a safety stock so that the printing machine always have available sheets. 

In the printed sheet buffer every contract has its own space, and every customer can 

have several contracts. There are also buffers for ears, bottoms, lids etc. Because of 

the buffers one process does not have to wait for products from another process 

earlier in the production flow. 

  



30 

 

4.3 Energy flow 

Emballator uses electrical energy, LPG and solvent in the enamel as energy sources.  

Emballator uses LPG for heating the facility. Since the ovens at the printing process, 

enamel process and after the welding operations at the can production line generates 

heat to the facility Emballator only need to heat the facility during the winter. When 

the average outside temperature for day and night is above four degrees the 

production at Emballator is enough to keep room temperature inside the facility. 

Because of the Swedish climate no cooling of the facility is needed.  

 

Finished Can

Transportation

Enamel paint process Printing press process

Forklift

AGV

Oven

LPG Solvents

Electricity

Electricity

Enamel paint machine

Enamel paintElectricity

Compressed air

LPG Solvents

Printing press

PaintElectricity

Compressed air

Oven

Electricity Electricity

Main production line

Electricity

Compressed air

Oven 1

LPG Electricity

Oven 2

LPG Electricity

Enamel paint

Cooling water

Lid

Bottom

Ears

Electricity

Compressed air

LPG

Electricity

Electricity

Compressed air

LPG

Electricity

Electricity

Electricity

Electricity
Electricity

 
Figure 8 Energy flow at Emballator 

4.4 Data gathering and data to information transformation 

Much of the process data were available through the process computer systems used 

by the production leaders and operators. It was necessary to evaluate the data from a 

quality perspective. Discrepancies where dealt with by discussing with production 

leaders and assumptions were documented for the future model. There were also 

unstructured interviews with operators responsible for the processes.  

4.4.1 Process data 

The enamel machine and the printing press are to a large extent used by products 

bound for other production lines than that of 180-1. The data not regarding 180-1 was 

used to find distributions for order sizes, number of incoming raw material, number of 

printing rounds and number of enamel layers etc for non 180-1 lines. The different 

lines where not differentiated but treated as one type of sheet and can. 

Data for the enamel paint machine was retrieved from the machines stop log data for 

2011. This included stops, failures and set up times. In the collected data all stops 

were marked with a cause, i.e. set up time, kind of failure or service action. During 

the interviews the operators explained that they did not always label the stops with the 

correct description (i.e. setup marked as failure and vice versa), due to inadequate 

information. Therefore to a large uncertainty was present to separate the setup time 



31 

 

from the failures. Because of this the distribution for mean time to repair  (MTTR) 

includes set up times and the distribution for the mean time to failure (MTTF) 

represent the time to failure or set up.  

The stop times in the enamel process and the printing process are considered as 

indirect costs since the stops are assumed to be related to the machine and not because 

of the nature of the order. To distribute the GWP contributions due to stops, the total 

time of all stops for each process was summed up and divided with the total number 

of sheets that had gone through the process during the same time frame. Each sheet is 

then given a GWP contribution that is divided among number of cans, bottom or lids 

that sheet will later becomes. Since the set up times in the enamel process is not 

possible to separate from the stop times these GWP contributions were also 

distributed as overhead costs.  

The stop log for the printing press for 2011 was collected. This include order number 

for each order, number of times that order had gone through the printing operation, 

start and end time for each order, start and end time for set up and start and end time 

of stops because of failure or service. Stops for adjustments or failures in the middle 

of an order are not recorded. From the collected data the distributions for MTTR and 

MTTF for the longer stops between orders and set up times could be calculated. To 

find all short stops, such as minor adjustments, not recorded in the collected data, the 

speed in sheets per minutes where calculated for each order and then distribution was 

found. 

The system to record set up times and stop times, in the 180-1 production line, is 

considered too complicated by Emballator's employees and is not in use. Therefore it 

was not possible to find data points for MTTR or MTTF. Instead the time recorded 

for quality checks for every thousand can were studied. From this data it was possible 

to find the time between the first quality check on one order, to the first quality check 

for the next order. From the times between these time points a rate of cans per minute 

was calculated. In this way the set up and failure times are included in the cans per 

minute data. Distributions for batch sizes up to 1000 cans, 1001-2000, 2001-3000, 

3001-4000 and above 4000 cans were calculated. This separation of batch sizes was 

done in order to make the set up times more proportional to the number of cans.  

The data for the lid, bottom, ear and plastic wrapping machine were not considered as 

important as for the enamel paint machine, printing press and production line. This 

assumption would have to be validated in the sensitivity analysis. Data for the lid and 

bottom machines were collected by estimations by operators.  

The ear machine and plastic wrapping machine were observed for a couple of minutes 

for estimations of the cycle times.  

4.4.2 Material data 

Both the mass and specification of content for each material used in production were 

needed. Some materials, such as plastic handles, are optional and for these materials 

the share of orders with and without these options were also needed.  

The amount of each material was collected from SAP.  Plastic handles and wrapping 

plastic are not specified in weight but only pieces per can and length. For these 

materials a physical example was collected and weighed.  



32 

 

It is difficult to find out the exact content for each material. Most of the chemicals are 

only labelled with the suppliers name and the material safety data sheet only mentions 

components that are toxic or flammable. Since it would not be possible to find LCI 

data for supplier specific products, estimations based on the material safety data sheet 

were considered good enough.  

Blueprints of the metal sheets were studied to find the exact amount of steel used for 

all cans. Also the wastes due to the cutting of metal sheets were retrieved from these 

blueprints.  

The raw material sheets used for the 6l can are used for other products as well. There 

are no data available regarding how the 6l sheet is used among other can sizes. The 

blueprints show a large amount of scrap for the 6l cans. In reality the production use 

the scrap for other cans as well as using the sheet for other cans when there is a lack 

of the correct raw sheet.  

4.4.3 Order, contracts, transportation and buffer data 

To find a distribution for batch sizes, data from can production line with reported cans 

per shift were studied. 

 

The contracts divisions into suborders depend on the size of the contract and the 

customer. The larger the contract the more sub-deliveries, this is usually the case there 

are of course exceptions. Some contracts stay in the system for a long time sometimes 

up to a year or more. Because of the contract system it was not possible to find 

sufficient historical data to support this representation. The data could not be 

validated and a simplification for the order system had to be made. This simplification 

is based on the call offs of produced cans statistic. A sensitivity analysis will have to 

ensure if this is a good enough representation. The sensitivity analysis will also have 

to rule if the differences in time in buffer because of this simplification have a 

significant impact on the final model. The theory behind this problem is 

acknowledged in the theory chapter 2.6.2 Input data management method. 

 

To find the distances for transportation of materials to Emballator the purchaser at 

Emballator named the cities from where the materials are shipped. Google maps and 

Eniro were used to find the distances from each supplier. Materials shipped from Asia 

are assumed to be shipped by boat to Göteborg and then shipped by truck to 

Emballator in Ulricehamn. Shipments from Europe (not including Sweden) are 

assumed to be shipped by train to Göteborg and then shipped by truck to Ulricehamn. 

All shipments from Swedish suppliers are assumed to be shipped by truck.  

 

Efforts were done trying to find the dynamic of the storage areas. Data from 2011 for 

the raw sheet buffer, buffer for sheets with enamel, buffer for printed sheets, buffer 

for finished cans and buffer for bottoms were collected and analysed.  

 

The raw sheet buffer data contained information about incoming and outgoing parts 

for each sheet size. Since the batch sizes are quite large only a few points of data for 

each sheet size were available.   

 

The buffer for sheets with enamel contained information about incoming and 

outgoing amounts for each kind of sheet with each kind of enamel layers. At most, 

twelve data points were available for each kind of sheet.  



33 

 

 

Data for the buffer for printed sheets were collected using SAP. Since the sheets in 

the buffer for printed sheets are custom made for each kind of can the data contained 

information for incoming and outgoing sheets for each kind of can. Due to the high 

variety of products it was not possible to draw any useful conclusions.  

 

The data collected for the buffer of finished goods contained information of incoming 

and outgoing amount of cans for every month. Due to the high variety of products it 

was not possible to draw any useful conclusions.  

 

The data collected for the buffer of bottoms contained information about the incoming 

and outgoing amount of each bottom type for each month. At most, twelve data points 

were available for each kind bottom.  

4.4.4 Energy consumption 

The energy sources at Emballator are electrical energy, Liquefied Petroleum Gas 

(LPG) and compressed air. The compressed air is produced by compressors using 

electrical energy.  

The electricity consumption for the entire factory is available at the owner of 

Ulricehamn's power distribution grid. It is measured by the hour.  To complement this 

data Emballator's electrician was able to help with consumption measurements 

throughout the flow, as well as the forklift batteries/chargers and the compressors 

producing he compressed air.  

It was not possible to measure the consumption of compressed air at each place where 

it is consumed. Instead the facility manager looked at the efficiency of the compressor 

at a given time when he knew which machines that where active. From this data he 

could estimate the share of the consumption.  

The consumption of LPG for heating the ovens at the enamel paint machine and the 

printing press machine could be collected by observing sensors near the ovens. 

Attempts were done trying to find the stoichiometry for the combustion of solvents in 

the enamel and LPG in the enamel process and the LPG in the printing press process.  

During the autumn of 2011 Emballator exchanged their utility unit, and therefore also 

their energy source for heating the facilities, from oil to LPG. The consumption of 

LPG is very much associated to the outside temperature. About twenty data points of 

LPG consumption could be found and together with temperature data from Sveriges 

Metrologiska och Hydrologiska Institut (SMHI) an equation could be calculated for 

the LPG consumption depending on the outside temperature. The equation shows that 

when the average outside temperature for one day and night rises above 4°C no LPG 

is consumed for heating the facility. This is because the ovens at the paint machine 

and enamel paint machine produces so much heat to the rest of the facility.  

  



34 

 

4.4.5 Distribution factors 

Distribution factors for distributing overhead GWP contributions was calculated by 

analysing blueprints of the sheets together with number of cans produced for each can 

size during 2011 and number of cans per pallet for each can size. The distribution 

factors were calculated both for each can size and for each can production line.  

 Share of pallets produced 

 Share of cans produced 

 Share of metal sheets consumed 

 Share of weight consumed 

 Share of cutting waste 

Also, the blueprint of the facility was studied to find the share of the total area for 

storage of finished goods, can production line, raw material buffer and so on.  

The working hours during 2011 for enamel paint machine, printing press and can 

production line were calculated by analysing data for reported cans or metal sheets. 

By looking at times the number of shifts at each resource was counted. If there had 

been any activity during one shift it was assumed that the resource was running 

during the entire shift. After counting all shifts it was multiplied with the number of 

minutes for each shift. From the total number of minutes the share of an entire year 

was calculated for each resource. 

LCI data for steel, steel wire, enamel paint, alkyd, boat transport, truck transport, train 

transport, polyethane, polystyrene, production of LPG, combustion of LPG, water, 

electricity, where collected with support from Björn Johansson and Jon Andersson 

using SimaPro 7.3.  

GWP data were collected from Intergovernmental Panel on Climate Change (IPCC) 

Report of the Intergovernmental Panel on Climate Change 2007. For additional 

reading about the LCI and GWP data collected please see Appendix.  

4.5 Model description 

The functional unit is the can and its lid, since the system starts with the raw material 

sheets so does the model. The large inventories on hand throughout the real system 

acts as buffers between the production processes. The effects the processes have on 

each other are decreased and in turn the dynamic between them. This is mirrored in 

the model and supported by the production technicians. The buffers reduces the 

dynamic in the real world as well as in the model. Each process is only dependent on 

itself and its breakdowns and setups.  

 

The order data is based on the call offs, made when delivering cans, as mentioned in 

4.1 Information flow. The order numbers for the cans correspond to the bottoms and 

lids. The order gets a specification in the beginning of the code which decides what 

type of bottom and lid. The specification also give number of enamel layers, printing 

press rounds, speed in 180-1, plastic handle or not, number of plastic packaging layers 

etc. These factors decide how and how fast the loads move through the processes. 

  



35 

 

4.5.1 Enamel process 

There are four steel sheets processed for 180-1 production line, these in turn have 

between one and three enamel layers. All 22 type of enamels or their thicknesses have 

not been modeled, instead an approximation of one, two or three layers in 

combination with the four different sheet types have given 12 different types of 

enameled steel sheets. 

 

Nr of Enamel 
Layer 

Type of steel sheet 

 1 2 3 4 
1 1 2 3 4 
2 5 6 7 8 

3 9 10 11 12 
Table 3 The different combinations of sheets and enamel. 

The enamel machine and the printing press are to a large extent used by products 

bound for other production lines than that of 180-1. The data compiled about non 180-

1 was used to model loads to simulate not available time in the enamel machine and 

printing press. By this the printing press and enamel machine would be dynamically 

occupied by non 180-1 sheets. These loads where discarded after the printing press 

since there is no processes where the 180-1 and other lines share processes after the 

printing press. The printed sheet buffer accommodates sheets to all lines, but to 

simulate Emballator's entire product variation in the model is not within the scope of 

the project. 

4.5.2 Printing press 

Each printing round gets a cycle time and a setup time, a distribution decides whether 

a failure or maintenance time is appointed to that printing round. Since the print on 

the can is customer specific a similar simplification as in the enamel process has been 

done for the sheets processed in the two color printing press. The sheets processed in 

the printing press intended for the 180-1 line are printed between one and seven 

rounds. This results in 84 different combinations of sheets. Since the printing press 

can handle two colors the one and seven rounds correspond to between 1 and 14 

colors, there is no historical or other data supporting any assumptions on the number 

of colors. 

 

Nr of printing 
rounds 

The different enamelled sheets 

 1 2 3 4 5 6 7 8 9 10 11 12 
1 1 2 3 4 5 6 7 8 9 10 11 12 
2 13 14 15 16 17 18 19 20 21 22 23 24 
3 25 26 27 28 29 30 31 32 33 34 35 36 
4 37 38 39 40 41 42 43 44 45 46 47 48 
5 49 50 51 52 53 54 55 56 57 58 59 60 
6 61 62 63 64 65 66 67 68 69 70 71 72 

7 73 74 75 76 77 78 79 80 81 82 83 84 
Table 4 Different combinations of enamelled sheet and printing press rounds 



36 

 

4.5.3 Can production line 

In the 180-1 line the setup and failure time is included in the cycle time, by that logic 

a larger order has a higher speed per can than a small order. This is specified in the 

beginning according to order size. 

 

The raw material sheets are cut into the metal sheet creating the body of the can.  This 

is represented in the model by cloning the load into the same number of pieces.