The WeCycle Project - Carbon Calculator development for IT equipment

(1)

IN

DEGREE PROJECT ENVIRONMENTAL ENGINEERING, SECOND CYCLE, 30 CREDITS

STOCKHOLM SWEDEN 2018,

The WeCycle Project - Carbon

Calculator development for IT

equipment

KONSTANTINOS STOURIS

KTH ROYAL INSTITUTE OF TECHNOLOGY

(2)

The WeCycle Project – Carbon Calculator development for IT equipment WeCycle Project – Koldioxidkalkylator för IT-utrustning

TRITA-ABE-MBT-18426

Degree project course: Strategies for sustainable development, Second Cycle AL250X, 30 credits

Author: Konstantinos Stouris Supervisor: Anna Björklund Examiner: Göran Finnveden

Department of Sustainable Development, Environmental Science and Engineering School of Architecture and the Built Environment

KTH Royal Institute of Technology

(3)

K. Stouris / The WeCycle Project - Carbon Calculator development for IT equipment 1

© Konstantinos Stouris, 2018

Degree Project in Strategies for Sustainable Development, Second Cycle (AL250X) Royal Institute of Technology (KTH)

Stockholm, Sweden

(4)

Try to leave the Earth a better place than when you arrived.

Sidney Sheldon

(5)

Acknowledgments

A big thank you note to my thesis supervisor at KTH Anna Björklund, for the support and guidance during this project. She allowed this project to exist as an independent work, but was always there to steer it in the right direction when needed.

The same goes for WeCycle and Greener Scandinavia for providing the opportunity to get involved in such an interesting project, and contribute with my knowledge in a real case project, as well as to WeCycle’s supervisor Victor Ezeogo, for his constant support, excellent ideas, data, information related to the project, and the very interesting and constructive meetings throughout this project.

Acknowledgement is also needed to be expressed for Inrego, a company that acts in the IT reselling sector as well, for data and information provision regarding their practices and environmental calculations.

Lastly, a big thank you to everyone else who helped and supported the completion of this thesis project. It would not have been possible to complete this project without their help!

June 1st, 2018 Konstantinos Stouris

(6)

Abstract

With global emissions of human activities that drive climate change on the rise, global institutions and authorities are trying to introduce new regulations in the industry, in order to accomplish a significant reduction of carbon emissions. In order for companies to be more effective in reducing carbon emissions, not only from their products, but also along their value chains and product portfolios, it is of vital importance to understand and quantify them. Following that need, tools that can measure the carbon footprint of various corporate operations (carbon calculators) have risen in popularity in the latest years.

A sector in which companies can significantly improve their environmental impact is their IT equipment portfolio. WeCycle, as developed by Greener Scandinavia AB (partner of this project), is a platform that facilitates reselling of old IT equipment, while aiming to reduce its environmental impacts. This project then, in cooperation with WeCycle, aims to develop a software tool that calculates the environmental benefits (kg of CO2 eq. avoided) when reusing old IT equipment. This can help clients estimate this benefit, while also providing a CSR incentive.

The specific methodological steps needed in order to complete the project included literature review concerning the state of e-waste and initiatives to minimize its environmental impacts, guidelines and procedures related to LCA of IT equipment and various other carbon calculators, developing calculation model and assumptions in order to compile the database, interface design, and finally using and testing the software tool against a real case scenario - case study provided by WeCycle.

The results, and design process of the project, were enlightening in the matter of understanding potential benefits of reusing IT equipment, but also in identifying the “hotspot” stages of an electronic device’s lifecycle. Even though variations were noticed depending on the type of the device (e.g. smartphones vs desktop computers), it is evident that the emissions that occur during the production phase are considered of major importance (ranked either 1^stor 2^nd most important/emission heavy stage), and therefore the benefits of reusing are of a high relative magnitude.

All in all, this project resulted in a useful tool for WeCycle to measure the benefits of their practices, as well as for any user or company that would like to measure the carbon emissions that can be avoided when they give their old IT equipment up for resell. Hopefully, by easily quantifying these benefits, this tool can motivate both a behavioural change in the industry, as well as researchers to expand it in order to cover all sectors of the industry and everyday life.

Keywords: Carbon footprint, Carbon calculator, Circular Economy, Life cycle Assessment (LCA), GHG emissions, ICT industry, IT-equipment, E-waste, Laptops, Smartphones, Desktops, Printers, Monitors, Recycle, Reuse, Resell, Corporate Social Responsibility (CSR), WeCycle.

(7)

Sammanfattning

När globala utsläpp av mänskliga aktiviteter stiger, försöker globala institutioner och myndigheter att införa nya regler för att minska koldioxidutsläppen. För att företagen ska vara mer effektiva när det gäller minskade koldioxidutsläpp, inte bara från sina produkter men också med sina värdekedjor och produktportföljer, är det viktigt att förstå och kvantifiera dem. För att uppnå detta, har verktyg som kan mäta koldioxidavtrycket av olika företagsverksamheter (kolkalkylatorer) ökat i popularitet de senaste åren.

En sektor i vilken företag kan förbättra sin miljöpåverkan är deras IT-utrustning. WeCycle, ett projekt som utvecklats av Greener Scandinavia AB (partner för detta projekt), är en plattform som underlättar återförsäljning av gammal IT-utrustning medan den siktar på att minska miljöpåverkan. Projektet, i samarbete med WeCycle, syftar till att utveckla ett mjukvaruverktyg som beräknar miljöfördelar (kg CO2-ekv.) vid återanvändning av gammal IT-utrustning. Detta kan hjälpa kunder att uppskatta denna fördel, samtidigt som de ger ett CSR-incitament.

Projektets resultat var till hjälp för att förstå de potentiella fördelarna med att återanvända IT- utrustning, men också för att identifiera "hotspot" -stadierna i en elektronisk apparats livscykel.

Även om det märktes variationer beroende på enhetens typ (t.ex. smartphones jämfört med stationära datorer) är det uppenbart att utsläpp som uppstår under produktionsfasen är av stor betydelse (rankad antingen viktigaste eller näst viktigaste fasen) och därför ger återanvändning relativt stor miljönytta.

Förhoppningsvis, genom att kvantifiera dessa fördelar med ett lättanvänt verktyg, kan detta projekt motivera både en beteendemässig förändring i branschen och forskare att vidareutveckla verktyget till att omfatta alla industrisektorer och hushållens konsumtion.

Nyckelord: Koldioxidavtryck, Koldioxidkalkylator, Cirkulär ekonomi, Livscykelanalys (LCA), Växthusgasutsläpp, IKT-industri, IT-utrustning, E-avfall, Bärbara datorer, Smartphones, Stationär datorer, Skrivare, Bildskärm, Återvinning, Återanvändning, Återförsäljning, Företags samhällsansvar (CSR), WeCycle.

(8)

Table of Contents

Abstract 4

Sammanfattning 5

List of Abbreviations 9

1. Introduction 10

2. Goal of the Project 12

2.2 Research Objectives 12

2.4 Relevance 12

3. Background 14

3.1 Circular economy and carbon emissions 14

3.2 Corporate Social Responsibility 16

3.3 Life-Cycle Assessment of IT equipment 18

3.4 Similar projects - tools 20

4. Methods and Research Design 21

4.1 Methodology 21

4.2 Delimitations 23

5. Results 24

5.1 Assumptions and simplifications 24

5.2 Tool design 26

5.2.1 Calculation model 26

5.2.2 Database compilation 30

5.2.3 Tool implementation 34

5.3 E-LCA tool presentation 35

5.4 Case study 36

6. Discussion - Conclusion 41

6.1 Discussion on results 41

6.2 Data availability and validity 42

6.3 Uncertainty 43

7. Conclusions 44

8. Future studies - transferability 45

Notes 46

References 47

Appendix A: Calculation model 52

Appendix B: Database showcase 55

(9)

Appendix C: Detailed software tool screenshots 87

Appendix D: User Manual 91

Appendix E: Sensitivity analysis results 92

Appendix F: Assumptions per brand 95

(10)

List of Figures

(3.1) CSR Concept Model (4.1) Methods overview (5.1) Calculation Model (5.2) Device flow chart (5.3) Tool Interface

(5.4) Introduction Tab of the software tool

(5.5) The second introductory tab of the software tool

(5.6) Input and Results for the Laptop and Desktop categories (5.7) Overall Results tab.

List of Tables

(5.1) Assumptions and Simplifications

(5.2) Results of model calculations for iPhone 7 (32GB model) (5.4) Case study devices to be assessed

(11)

List of Abbreviations CO2 - Carbon dioxide

CSR - Corporate Social Responsibility EPA - Environmental Protection Agency

EPEAT - Electronic Product Environmental Assessment Tool EU- European Union

FU - Functional Unit

GEC - Green Electronics Council GHG - Greenhouse Gas

ICT - Information and Communication Technology Equipment ISO – international Organisation of Standards

IT - Information Technology Equipment LCA – Life cycle Assessment

LCI - Life Cycle inventory

PAIA - Product Attribute to Impact Algorithm PCF - Product Carbon Footprint

RO - Research Objective

WEEE - Waste Electrical and Electronic Equipment

(12)

1. Introduction

Human based carbon emissions that drive climate change are at a historic high, constantly growing, and threatening various life forms of the planet. According to UN reports, greenhouse gas emissions were higher than ever in 2017, rapidly increasing since the 1920’s (from around 4 billion tonnes of CO2) to a staggering amount of 35 billion tons of CO2 per year or 403.3 parts per million (number of molecules of CO2 per million molecules of dry air) globally (World Meteorological Organization 2017). According to various scientists, in order to minimize and stop global warming, along with all its disastrous and unprecedented impacts, carbon dioxide emissions must be lowered up to 85% below 2000 levels. As a result, there is an urgent need for anthropogenic carbon emissions to be reduced across all industry sectors, and corporate innovation and initiative is of vital importance to achieve this (World Meteorological Organization 2017; Intergovernmental Panel on Climate Change 2007).

As the climate change problem deepens, governmental institutions, are and can be expected to continue introducing and enforcing in the future strict regulations and legislations concerning carbon emissions in the industry, that corporations have to comply with. Moreover, according to the WRI, additional corporate action on the matter is considered good business and good proactive planning. Initiatives to reduce a company’s carbon footprint can be further motivated and facilitated by a detailed understanding and quantification of its GHG emissions or carbon footprint. Additionally, the industrial sector’s understanding of the need to calculate GHG emissions along their value chain, equipment portfolio, and products to effectively tackle climate change is rapidly increasing, along with the need to develop tools to calculate them. (World Resources Institute 2004)

A significant part of a company’s equipment portfolio, is electronic equipment, which due to the fact that companies regularly upgrade it, generates a big amount of waste. Electronic or e-waste is part of the “WEEE” or Waste Electrical and Electronic Equipment, a sector of waste heavily regulated by the EU, and targeted to be drastically reduced in coming years (Unger et al. 2017).

Worldwide, 44.7 million tons of WEEE is estimated to be produced per year, while the EU market is considered to be one of the main streams of WEEE generation globally, a number that has to be significantly reduced (Balde 2015). A big part of this is being treated by unofficial or informal recycling facilities in Asia, with shredding, burning and dismantling of them causing a significant amount of emissions with severe impacts to human health and natural environment. Chemicals are being released in air and water, polluting food sources and drinking water (Frazzoli et al.

2010)

In line with this, the WeCycle project, as developed by Greener Scandinavia AB aims to facilitate and promote reselling and reuse of used IT equipment, in order to reduce the environmental impacts of the IT equipment sector. The company’s vision is to give a “second life” to all reusable equipment, and to achieve that by establishing the reselling procedure as a hassle free activity, while reducing the environmental impact of IT equipment and promoting circular economy

(13)

(Wecycle 2017). This also complies with the EU’s vision of a circular economy, where the value of materials and products should remain in the economy for the longest possible time (EC 2015).

For this reason, the need rises to calculate the carbon footprint of avoided when reselling- reusing IT equipment. This can add a calculated CSR (Corporate Social Responsibility) incentive for all companies, to facilitate and promote reselling and reduce IT waste. Inrego, a company that also acts in the IT reselling sector in Sweden since the 90s, is making efforts to calculate the emissions avoided through their practices of reselling IT equipment and recently achieved results based on a simplified model, and average data for each device type (Inrego n.d.).

Life Cycle Assessment, as a growing in popularity method that assesses the potential environmental impacts of a product or process throughout all stages of his life, provides a solution to this need, as IT-equipment manufacturers use it to calculate the carbon footprint of their products throughout all stages of their lifetime, following internationally accepted protocols and guidelines (Life Cycle Assessment n.d.; Arvanitoyannis 2008; International Organization for Standardization 2006). By utilizing this data, it is possible to create a way to calculate the environmental benefit of reusing old IT-equipment.

(14)

2. Goal of the Project

The main aim of the project is to develop a software tool that includes Life Cycle Assessment carbon indicators (CO2) and industry data to enable the industry to assess the environmental profits of reusing IT equipment. The tool is aimed to be used by the WeCycle project of Greener Scandinavia AB to assess the environmental benefits of reusing old IT-equipment. As WeCycle is a platform that facilitates the resale process of old It-equipment, the tool will not only be used from WeCycle members, but also from prospective clients, who want to estimate the environmental benefits of giving up their old equipment for resale, in order to aid decision making.

2.2 Research Objectives

The specific Research Objectives (ROs) that shaped the project towards achieving the aforementioned aim, were:

RO 1: Identify and review environmental impacts, carbon emissions and life cycle indicators of IT equipment, as well as other carbon calculator projects that can be utilized when developing the tool.

RO 2: Research company specific data regarding carbon emissions and materials used in IT equipment. E.g. Apple, Dell, Lenovo, Samsung, Ricoh etc and compare their calculation methods.

RO 3: Establish a custom made database that includes carbon footprints of IT equipment in all life cycle stages, according to industry data and research - calculations.

RO 4: Develop and implement the carbon calculation tool.

RO 5: Test the tool by using a case study provided by Greener Scandinavia AB.

2.4 Relevance

As being involved with the development of a carbon calculator, this project is of great interest to the environmental assessment experts and engineers, who aim to developed standardized tools to assess environmental impacts of various human activities.

Moreover, the presented software tool could evolve and develop to a tool that involves all major companies in Sweden and worldwide as part of their Corporate Social Responsibility (CSR) actions, and their need to assess their emissions along their value chains and equipment portfolio. Motivating a reduction of emissions from IT equipment can help in achieving a

(15)

reduction on constantly increasing amount of 44.7 million tonnes of e-waste globally (Balde 2015).

Lastly, the creation of the software tool will serve the dire need of establishing a user friendly software tool, based on industry data, that can assess the benefits of reusing old IT equipment, by taking into account all stages of a product’s lifetime, and all specific characteristics of each model and brand.

(16)

3. Background

As this project is being involved with e-waste and aims to create a tool that calculates carbon emissions related to IT-equipment, it is important to set a theoretical background that covers these matters, in order to aid decision making during the tool creation (interface elements, design decisions, calculation model, database compilation etc), as well as establish a framework to assess and discuss the results. In accordance to this, the following chapter introduces the concept of circular economy (CE) in the e-waste sector, and discusses today’s state and challenges regarding e-waste and carbon emissions, and moves on to present Corporate Social Responsibility (CSR) practices, and ways that it can promote sustainable environmental behavior in the corporate world. This underlines the necessity of creating tools (carbon calculators) that can easily calculate emissions of various business actions, as does this project, but also lays the groundwork to later discuss how the results of this project and its accompanying tool can be reported by companies in their CSR or Environmental report.

Furthermore, this chapter continues by reviewing tools and guidelines, concerning LCA on IT- equipment, so that all tools, databases, models and information used during this project are explained and examined, to ensure compatibility. Lastly, similar tools and efforts that aided in the tool development, along with their contribution, are mentioned, to provide the “whole picture” to the reader.

3.1 Circular economy and carbon emissions

Even though researchers often argue whether increasing economic growth will always lead to increased pollution and carbon emissions, it is a common perception that until today, rapid economic growth has steered carbon emissions to higher than ever numbers (Lim 1997; Fodha

& Zaghdoud 2010; Kunnas & Myllyntaus 2007). Particularly, global carbon emissions are as high as ever, rapidly increasing since the 1920’s (from around 4 billion metric tonnes of CO2) to a staggering amount of 35 billion m.tons of CO2 per year (Global Emissions 2017; World Energy Outlook 2017).

Countering today’s economic model, various economic theories have tried to propose a solution towards a “green” and sustainable economic future, such as the steady state economy, or degrowth (Latouche 2011). Circular economy (CE), as the most popular theory aiming to overthrow today’s dominant economic model and counter environmental challenges, paves the road for a sustainable economic growth and development. Based on 3 prime principles (3 Rs) of Reducing, Reusing and Recycling, CE aims to transform a linear economic model to a circular, closed loop economy, where the stakes between economy, environment and society are balanced (Zhijun & Nailing 2007; Ren 2007). The first principle, named “Reducing” stands for reducing the need for energy and raw material usage by achieving an increased efficiency in human processes and products. The second one, “Reuse”, motivates the use of materials or products more than once for the same purpose, and finally, the “Recycling” principle stands for material recovery and reprocessing (EC 2008). CE is recently getting increased attention in the

(17)

industrial development debate, and the Ellen MacArthur Foundation, as well as the European Commission are the leaders in the field, by releasing guidelines for CE implementation and supporting research and development concerning CE (EC 2015)(EMAF 2012). There is a popular ambition amongst these institutions, that full and correct implementation of CE practices in the industrial sectors, will not only have environmental gains, but can also contribute to economic growth and can even result in gains of 1000 billion dollars annually, and can instate CE as the future economic paradigm (Mckinsey 2014; EC 2015; EMAF 2012).

The European Union’s vision of a proper Circular Economy (CE) implementation is described in its 2015 action plan, along with its guidelines for waste management (EC 2015). According to the EU directive materials and products should remain in the economy as long as possible, by means of reusing or reselling them. This can lead to reduced carbon emissions and higher efficiency of resources.

Adding to this, The Ellen MacArthur Foundation underlines the importance of creativity and innovation in closing the loop in the way that people produce, use and dispose consumer electronics. According to their vision, electronic products should be ”loved” for longer, and remain in the market for longer periods than expected, either by their initial owners, or by refurbishing / reselling.(EMAF 2018)

Balde underlines the growing problem of e-waste, and report that the mass of e-waste globally amounts to 44.7 million tonnes globally, with 435 tonnes of it being smartphones (Balde 2015).

To get things into perspective, this amounts to 1.2 times the mass of the Empire State building (Empire State Realty Trust n.d.).

Europe, is responsible for a big amount of it, as 27,5% of all e-waste is being generated in Europe. This number amounts to 12.3 million tonnes, or an average of 16,6 kg of e-waste generated per capita. In Sweden, Scandinavia and the northern part of Europe this number is even higher, rising up to around 20-25 kg per inhabitant (Balde 2015)

Environmental impacts of e-waste can be severe, mainly due to the fact that a lot of the processing (mainly in developing countries) is happening in informal facilities, leaking environmentally hazardous materials into water and food streams worldwide, affecting animal and human health (Frazzoli et al. 2010) Specifically, emissions and material discharge from desoldering circuit boards, as well as lead, tin barium and mercury releases to the atmosphere and water can prove harmful to animals and humans. Emissions from dioxides, heavy metals and hydrocarbons that are emitted during stripping shredding and burning of parts add to this problem, alongside with materials that are disposed in landfills (Wath et al. 2011). According to Baker et al, it is safe to assume that actually only 50% of a laptop’s materials can be recycled.

The rest is composed by non-recyclable materials (Wath et al. 2011; Baker & United Nations Environment Programme. Division of Environmental Conventions 2004). It is important to note here that this varies greatly from country to country, as there are different treatment methods in each country (Umair et al. 2015).

(18)

These negative effects on environment, food and water contribute to severe negative social impacts due to informal recycling processes around the world, mainly concerning the health of workers and the community that outweigh the positive impacts of creating new job opportunities in developing countries (Umair et al. 2015).

The Ellen MacArthur Foundation presents the circular economy system and strategies to reduce e-waste, and singles out reusing and prolonged life of devices as the most effective strategy. A useful example to follow is the market of modems routers, that in average, are being reused to serve up to five households during their lifetime, as companies are renting them out to multiple clients (Boddington & Hubert 2016).

3.2 Corporate Social Responsibility

CSR (Corporate Social Responsibility) is considered to be a set of global self-regulation strategies and transnational mandatory legislations (Sheehy 2014). Initially, CSR was considered to consist only of non mandatory corporate-internal self-regulatory techniques and initiatives that help promote community development, social justice and sustainability. However, during later years CSR has been transformed from a system of internal corporate policies to compulsory national and global legislation schemes (Sheehy 2012). International laws have been enforced, pushing corporations to engage into "actions that appear to further some social good, beyond the interests of the firm and that which is required by law"(McWilliams et al. 2006).

A business may strategically choose to engage in CSR due to legal compliance requirements, ethical standards, or to reduce legal risk. CSR can also lead to increased shareholders trust and even economic benefits in the long term. Furthermore, CSR encourages businesses to engage in positive actions regarding the environment, stakeholders and the society (Farrington et al.

2017). CSR acts as an aid to organize a company’s mission and also provide guidelines to accomplish it.

In Sweden, CSR is divided in 3 major categories. Financial, Environmental and Social responsibility (Grankvist 2009). CSR Sweden, which fully complies with EU practices, describes economic responsibility as ways of behaving in the market to ensure firm profit. If the firm cannot ensure profit, cannot ensure its financial sustainability, and therefore cannot assist societal improvement. This applies both to internal and external practices of a firm (CSR Sweden n.d.).

The same goes for social responsibility, where internal includes healthy working environment and good employee relationships, and external ensures fair treatment of different customer groups. These can be achieved by establishing a balance between different sexes, nationalities, ages, and religions in the workplace, eliminating any form of discrimination, and donations (CSR Sweden n.d.).

Environmental responsibility, according to CSR Sweden, introduces actions that promote sustainability and increase awareness of the environmental impacts of a firm’s practices. Actions to achieve this include careful selection of materials and production methods, and reducing

(19)

direct and indirect impacts of a company’s actions, like environmental footprint of products, materials shipment etc (CSR Sweden n.d.). An overview of the concept model of Corporate Social Responsibility (CSR) and its subcategories can be found under Fig 3.1 CSR Concept Model.

Fig (3.1): CSR Concept Model

Calculating the emissions of a company’s actions is an integral part of corporate environmental responsibility. In order for a company to develop an effective corporate climate change and sustainability strategy, it is required to understand a firm’s GreenHouse Gas (GHG) impact.

However, until recently, corporations focused mainly on calculating the emissions of their own business operations and did not take into account emissions that were generated along their value chains and product portfolios (World Resources Institute 2004). Nowadays, as the industry demands methods and tools that facilitate the inclusion of GHG emissions along companies’

values chain, initiatives and protocols are emerging to facilitate this.

The GreenHouse Gas (GHG) protocol is such an initiative that aims to develop standardized methods and tools to calculate greenhouse gas (GHG) emissions from private and public sector operations, and propose mitigation actions (Greenhouse Gas Protocol n.d.). Alongside this, GHG protocol releases accounting and reporting standards and tools that calculate GHG

(20)

emissions to enable more effective GHG emissions management (World Resources Institute 2004).

Even though the GHG protocol toolbox is very rich, it does not include a tool that can help calculate GHG emissions generated by e-waste, and that highlights the need of one that can help companies calculate GHG emissions along their value chains.

3.3 Life-Cycle Assessment of IT equipment

Life cycle assessment (LCA) , is a method that assesses the potential environmental impacts of a product or process throughout all stages of his life, starting from extracting raw material and material processing, and continuing with manufacturing, transportation and distribution, use stage, maintenance and disposal (Life Cycle Assessment n.d.). To aid in achieving this, LCA guidelines follow a set of procedures that include assembling an input-output inventory, evaluation of potential significant environmental impacts related to inputs and outputs, interpretation of the inventory results, and lastly assessment of the results in relation to the study goals (Arvanitoyannis 2008).

Life cycle Assessment (LCA) popularity is constantly growing and it is establishing itself as a common environmental management tool for products, services and various technologies (Farrell & Sperling 2007). Evidence on this are both the increased use in state and national policies in the US of life-cycle greenhouse gas reduction, as well as the increased supply (and demand) for life-cycle carbon footprint labels globally (MacGregor 2010). However, as LCA’s popularity rises, various studies have resulted in contradicting results, increasing uncertainty in the field and undermining its reliability (Lenzen & Munksgaard 2002; Farrell 2006).

On the other hand, protocols, organizations (EU) and university institutions have issued detailed methods, tools and guidelines that aim to standardize the LCA process across the industry. ISO 14044:2006 and 14040:2006 provide researchers with specific prerequisites and directions for Life Cycle Assessment (LCA) and shape the assessment procedure by setting the definition, goal and scope of the LCA, Life Cycle Inventory (LCI) phase, assessment, interpretation, reporting and review of LCA (International Organization for Standardization 2006; Arvanitoyannis 2008). Most tools, that are involved with LCA actions, follow ISO 14044:2006 and 14040:2006, as do all that are involved in this project.

The Massachusetts Institute of Technology (MIT), along with the participation of members of the ICT industry, researchers and NGOs, has developed a tool that facilitates LCA of IT equipment by applying a simple, scientific process to improve the carbon footprinting process. The Product Attribute to Impact Algorithm - PAIA approach aims to streamline and reduce time and resources needed to calculate the carbon footprint of IT products, and create a baseline level of comparability between Product Carbon Footprints (PCF) (Lenovo n.d.).

(21)

The PAIA methodology, developed from scratch specifically for IT products, identifies the crucial stages of a product’s lifecycle, along with specific components that have a significant contribution to its carbon emissions or is of high importance to stakeholders. The components (e.g. screen, hard drive, etc) are then associated to manufacturing processes, transport and other activities. It takes into account all the materials extracted in order to produce each component, and by using existing LCI data, it calculates the impact related to it and its associated processes. After compiling the models for each part and process of the product, they are imported into an algorithm that covers the product as a whole. The algorithm then, provides the users with a reasonable estimate of its carbon and environmental impacts (MIT Materials Systems Laboratory n.d.).

The LCI (U.S. Life Cycle Inventory) database, that the tool is based on, is developed by the National Renewable Energy Laboratory (NREL), and provides information regarding energy, material flows and emissions that are produced when manufacturing materials, components or assemblies, from cradle to grave. The LCI’s aim is to provide support to LCA efforts worldwide, either in tool development, Life Cycle Assessment (LCA) reports etc (National Renewable Energy Laboratory 2012). Its most popular application is through the GaBi LCA software, one of the most commonly used LCA tools, that aims to support LCA, Life Cycle Reporting (LCR), LCC (Life Cycle Costing) and LCWE (Life Cycle Working Environment), by modelling every aspect of a product or system, from a life cycle perspective, and calculate its environmental impacts across various impact categories. Its applications vary from EPDs and Product Carbon Footprints to applications in the automotive industry, building certifications and LCA, or Ecodesign (GaBi Software n.d.).

The Electronic Product Environmental Assessment Tool (EPEAT) is a similar tool, developed by the United States Environmental Protection Agency (EPA) and the Green Electronics Council (GEC). The EPEAT, in an effort to promote sustainable practices in electronic devices manufacturing, is a tool that assesses electronic products based on specific components and attributes of the devices, in a similar way to PAIA and in accordance to ISO 14044:2006 and 14040:2006. (EPA 2014)

Concerning the ICT manufacturing industry, most big global manufacturers have engaged in recent years in LCA labeling techniques, and by following the guidelines of ISO 14044:2006 and 14040:2006 have released environmental reports for a variety of their products that present their estimated greenhouse gas emissions over their life cycle.

Moreover, most companies (Apple, Lenovo, HP etc) calculate the GHG emissions of their products throughout their lifecycle by using the PAIA tool of MIT (or EPEAT), and the information stored on GaBi database. Assumptions concerning the LCA process like the functional unit, lifecycle of the product, transportation etc are similar throughout the industry, rendering results highly compatible and comparable. Their carbon footprint calculations consider the GHG emissions’ contribution to global warming in kg of CO2 equivalents (kg CO2eq).

(22)

3.4 Similar projects - tools

During the undertaking of this project, various similar projects, or carbon calculators were researched and studied. Some of them offered useful insight in the interface implementation, others on the database compilation etc. (see chapter 4. Methodology and 5. Results for more) The most important assistance though, was provided by a Swedish company, Inrego, that is involved in similar actions as WeCycle and Greener Scandinavia AB.

As a company that also acts in the field of reselling IT equipment, Inrego aims to minimize environmental impacts of IT equipment and shape a new future of IT equipment handling in the society. Inrego, a company that started its operations in 1995 in Lund, is now considered a market leader in the IT reselling industry, being responsible for 260,000 reused devices per year and 2,800 tonnes of carbon emissions avoided. (Inrego n.d.)

In order to achieve their aim of strengthening sustainability efforts in Sweden and worldwide, Inrego made an effort to calculate the carbon emissions avoided when reusing / reselling IT equipment. In accordance with the study done in 2005 by the University of Berlin and their Energie- und CO2-Bilanz von PCs - Relevanz für ReUse-Strategien research, Inrego has compiled a simple calculation model that is based on the average carbon footprint of 10 selective devices (Apple iPhone 3, 4 and 5, Sony Xperia Arc, Sony Ericsson W890, HTC One, Samsung Galaxy Note 2 and Blackberry Torch 9810) and returns one value as the carbon saved when reusing a random smartphone/laptop/computer. Inrego considers that the model, even though general and average-based, represents a sufficient basis for general use (Inrego 2014). Inrego’s contribution to the project was useful not only during the calculation model compilation and assumptions (as explained in chapter 5: Results), but also in establishing a baseline project for how IT-equipment should be environmentally assessed. However, there is no common ground for comparing the two efforts, as Inrego’s simpler calculation model is based on average numbers of few selected devices for own use, instead of creating a software tool, usable by everyone, with a database of over than 500 entries that aims to ensure accuracy in its calculations.

(23)

4. Methods and Research Design

The following chapter presents the scope of the project, along with any assumptions and limitations that were set in order to complete the project and carry out software development.

Moreover, it includes a precise description and illustration of each methodological step and tool used during this thesis project.

4.1 Methodology

The methods used in order to complete the proposed project, included (but not limited to) literature review, algorithm development and testing through the use of Microsoft Excel, and cooperation with Greener Scandinavia AB and the WeCycle project for consultation. The step by step methodology used for achieving the project’s Research Objectives (ROs) and its overall aim are described below, along with the relations between the ROs of the project and how they collaboratively aid in achieving the aim of the project, illustrated under Figure (4.1): Methods overview.

In order to identify the specifics and setup today’s state of WEEE and IT waste, along with respective strategies and innovations that put the present project into context, an extensive literature review concerning IT waste was compiled, alongside with a study regarding the practices of companies involved in IT waste, Greener Scandinavia AB (WeCycle) and Inrego AB, or companies / institutions that innovate on carbon calculation projects (Cleanfox, GHG protocol). Inrego AB, provided detailed data and information regarding their practices in Sweden, and information regarding other initiatives was found online. Literature concerning e-waste was obtained by browsing databases (e.g. KTH Library) for keywords like: IT equipment, e-waste, WEEE, Laptop GHG emissions, smartphone carbon footprint etc.

The next step included identifying and reviewing current guidelines, procedures, protocols and tools used for LCAs in IT equipment, EU laws (GHG protocol, Scope 3 etc.) and Environmental Product Declarations (EPDs). This was complemented by reading about other software tools (GHG protocol tools) that can help measure environmental benefits of various corporate practices, circular economy, environmental impacts of IT equipment, recycling and reuse of IT equipment, previous practices of LCA of IT equipment. The GHG protocol, as mentioned in Chapter 3: Background, presents in its website a big selection of tools - carbon calculators used for various other applications along the industry, e.g electricity, but provided the project with useful insights regarding useful LCA indicators and uncertainty involved when developing LCA based carbon calculators.

From the aforementioned steps, information concerning the environmental impacts, emissions and indicators of IT equipment were derived, that were later used for design decisions, software development and setting the framework for the project (RO1).

(24)

Afterwards, in order to achieve RO2, extensive research was done on industry data, specific to each device. Most IT equipment manufacturing companies are releasing data regarding the environmental impacts of their products, that proved extremely useful (RO2). However, calculation methods and assumptions needed to be studied and compared in order to ensure compatibility of the industry data and establish a common ground for comparison. Results of this study can be found under Chapter 5: Results. These first two ROs created the foundation for RO3 and RO4 (as can be shown in Fig (4.1)), and their results are not presented separately, but alongside these of RO3 and RO4, in chapter 5: Results.

Fig (4.1): Methods overview

By establishing the necessary common ground, the project moved on with the compilation of the database that includes the carbon footprints of all products along with their respective carbon footprint for each life cycle stage (RO3) as an Excel spreadsheet. Specifically, data used for compiling the database was used from the following product manufacturers / institutions: Apple, Lenovo, HP, Dell, Samsung, Lexmark, Xerox, Canon, CFD Japan. In order to deal with data gaps, statistical averages were derived not only for each device type and brand, but also for similar models of devices, to increase accuracy.

Moreover, by studying the literature alongside with previous, similar studies, a calculation model was developed, that takes into account all life stages of a product and compares two scenarios.

A resell and non resell scenario. The model consists the foundation of the whole project, as all following calculations of the software tool are based on this model. The model and its results which can be found in Chapter 5: Results.

(25)

In order to develop and implement the software tool (RO4), other carbon calculators and similar tools were studied and useful elements were derived to use in the E-LCA tool (The GHG protocol inventory includes various similar tools for other industrial sectors, e.g. electricity generation). Some examples are the Purchased Electricity tool, created by the World Resources Institute, and the Philadelphia Green Business Carbon Calculator (World Resources Institute 2015) (O’Brien & Gere (OBG) 2010). Afterwards, the project continues with algorithm development, which were input in the Excel spreadsheet, by utilizing the created database and specific functions to calculate the carbon footprint of reusing IT equipment and visualizing the results.

Lastly, for running and testing the carbon calculator (RO5), Greener Scandinavia AB provided the project with a real life case study. The case study consisted of a list of IT equipment that were available for resale from a specific client / user. By using this data, the software tool was tested, assessed and any bugs and shortcomings fixed. The tool was continuously shared and with the WeCycle project, so that useful feedback and consultation regarding the software, as well as information concerning the refurbishing and re-selling procedures could be shared throughout the process of the project.

4.2 Delimitations

Even though the project was carried out in Sweden, it has a universal application, as it currently covers (and includes in the software database) products that are used worldwide, considering that all reselling / refurbishment transportation is happening at a national level. The database focuses on delivering accurate industry data from big manufacturers like Apple, Dell, HP etc, and calculates an average number, to provide as an estimation for the rest of the products, based on the material / brand or other characteristic of each product. These brands and products are popular globally, and that establishes the program as a tool usable worldwide in order to assess environmental benefits of reusing old IT equipment.

Moreover, calculations and results, while taking into account LCA indicators, focus on impacts of products only on the climate change impact category. All results are presented in amounts of GHG emissions avoided when reusing old equipment or kg of CO2 equivalent, as is the standard in the ICT industry.

(26)

5. Results

This chapter includes a presentation of project results, starting with a description of all assumptions and limitations encountered when developing the calculation model, the database and the tool, as well as during the case scenario testing. As explained in 4.1 Methodology, RO1 and 2 are used in order to achieve RO3 and RO4, so the following chapter offers results on RO3: Establish a custom made database according to industry data, RO4: Develop and implement the tool and RO5: Test the tool, use a case study provided by Greener Scandinavia AB.

5.1 Assumptions and simplifications

In order to overcome calculation difficulties and lack of data, certain assumptions needed to be made. Embodied energy of some materials/procedures and products were missing from global indexes, and assumptions had to be made concerning transport methods and specifications.

In order to deal with data gaps, statistical averages were calculated not only per type of device, but also per brand (due to similarity of manufacturing methods, manufacturing locations), and similar devices (due to similar materials and specifications). In order to minimize the uncertainty of this step and statistical calculations, a sensitivity analysis was conducted, when compiling the database, as explained under Chapter 5.3.2 Database compilation.

Moreover, different companies / manufacturers do not always use the exact same methods when calculating the carbon footprint of their products. During this project though, an effort was made to compile a database based on company data that can be comparable (all companies follow ISO 14044:2006 and ISO 14040:2006, as well as either the EPEAT or PAIA method, which have similar assumptions and techniques) and make required adjustments when needed.

Still, it is fair to assume that 100% comparability is not obtainable, and a specific level of uncertainty must be considered when dealing with the results. Under Chapter 6.3: Uncertainty, a detailed description regarding the uncertainty involved in this project can be found.

Moreover, in order to take into account all stages of a product’s life cycle, assumptions needed to be made regarding the emissions during refurbishment process. This phase included transportation to the refurbishment center, electricity needed for the refurbishment process, and lastly transporting the product to the new user, considering that all are happening at a national level. Trying to add these as inputs to the end program was considered to overcomplicate things for the user, while not significantly affecting the end result. For that reason, emissions for the whole refurbishment phase involved assumptions regarding average distances and electricity needs, based on literature. According to Inrego and WeCycle, for refurbishment and redistribution processes emissions can rise up to 1 kg of CO2 equivalent (Inrego 2014).

(27)

Additional simplifications that were done in order to streamline the process and implement the tool, regarded the use phase of the devices. Firstly, during model calculations, the use stage of the refurbished model was considered the same as a new one (3 years), a practice that can be realistic, as cloud computing is becoming increasingly popular, transferring computing power need from standalone devices to remote servers, and considering that refurbished devices are sold to users with lower performance needs than their initial owner (EMAF 2018). Secondly, in order to build a model, the difference in power efficiency between a 3 years old refurbished device and a new one that serves the same purpose is not considered during the model calculations. This is a basic assumption - simplification that the model is based upon, as without it, it would not be possible to predict the new device’s efficiency. However, as can be observed under Appendix C: Database showcase, this assumption is often a reality, as device power efficiency may not increase significantly in 3 years (especially for devices produced after 2012).

Such an example can be Apple’s iPhone, where the emissions of the use stage of iPhone 5s (9,12 kg of CO2 eq.) is approximately the same with iPhone 8 (9.1 kg of CO2 eq.). Even on cases that this was not a reality, it was evident, that this difference was not capable of altering significantly the end result.

Lastly, a matter that this project does not deal with, is the allocation of potential benefits by reusing old IT equipment, between the initial owner and the user that buys the refurbished equipment. The result of the software tool presents the overall benefit - emission reduction by reusing old IT equipment and does not decide on who will get the credit for it. A summary of simplifications and assumption follows under Table (5.1): Assumptions and Simplifications.

Table (5.1): Assumptions and Simplifications

Assumptions and simplifications

Dealing with data gaps Statistical averages per type of device, brand, model family and other specs

Production phase

Extraction and transportation of raw materials, manufacturing, transportation and assembly of end products and their respective packaging.

Information derived from industry data

Transportation

transportation and packaging, either by air or sea, from each manufacturing factory to the local distribution hubs

Use phase 3 years

Energy during use phase As long as reselling of devices happens in a national level, this does not interfere with the end result.

Refurbishment and Redistribution 1 kg of CO2 eq, assuming it happens on a national level

(28)

Second life stage duration 3 years, same as the first

Disposal

Includes transportation and energy demand for mechanical separation and shredding. Derived from industry data. Material recovery is being taken into account in production, not here.

Allocation of benefits between previous and current user

Not taken into account, just calculated the total benefit

5.2 Tool design

The tool implementation techniques will be documented in this chapter, by presenting results concerning the scope, calculation model, data collection, and final implementation of the tool, covering RO4 (Establish a custom made database according to industry data) and RO5 (Develop and implement the tool).

5.2.1 Calculation model

Data obtained from manufacturers regarding the carbon footprint of their products gives information on their emissions during production, transportation, use (3 years) and disposal of them. In order to calculate the emissions avoided when re-using a device, and compile the database, a calculation model needed to be developed.

Figure (5.1): Calculation model illustrates the model that these calculations were based on when creating the database. This can be observed in the last column of the database table, under Appendix B: Database showcase. In the model then, two different life cycle scenarios are compared. In the first one, two users-companies A and B fulfill their equipment needs by buying new equipment and disposing it after 3 years (use phase), whereas in the second scenario, User B’s needs for IT equipment are fulfilled by buying used equipment that User A has not use of anymore. The effort to build a straightforward model, with the least possible complications, was inspired by Berlin’s Technical University’s Reuse model, a model that is widely used on the reuse industry (Inrego 2014; Technischke Universität Berlin 2005). A basic difference between Berlin’s university model and the present one, lies in this project’s the lack of assumptions on how user A fulfills his IT equipment needs after year 3 (he may choose to purchase a new device or no new device at all), as this assumption does not contribute to the end result (cancels itself out, as it is identical in both scenarios), and can usefully be omitted.

(29)

Fig (5.1): Calculation Model

Specifically the stages that are considered in the life cycle of each device are the following, as also illustrated in Figure (5.2): Device flow chart:

1. Production phase. Where extraction and transportation of raw materials is included, as well as the process of manufacturing, transportation and assembly of end products and their respective packaging. Information for this stage was derived from industry data, specific for each device. Industry data assumes that all benefits from using recycled material is taken into account in this stage, instead of the End of Life stage, and considers a varying percentage of materials being recycled per device, according to model characteristics and specifics (Lenovo n.d.; Dell n.d.)

2. Transport. Where transportation of the end product, along with its packaging, either by air or sea is considered, from each manufacturing factory to the local distribution hubs, and then to end customers based on regional geography. Information for this stage was derived from industry data, specific for each device and company.

3. Use phase (3 years). A three-year period is considered as the length of each use phase of a product, based on customer use data. 3 years is considered the standard (mainstream) lifetime of an IT product (Apple 2016; Inrego 2014). Emissions that occur during this phase are mainly due to electricity generation to power / charge devices, but not adjusted for EU power generation emissions (US based). As long as reselling of devices happens in a national level, this does not

(30)

interfere with the end result. Industry calculates this according to the Energy Star® Typical Energy Consumption (TEC) method (with US based electricity emissions), which assumes different on – off times for different types of devices. E.g. desktops are considered to be off for 55% of the time, 5% on sleep mode, and 40% active, whereas laptops are calculated for 60%

off, 10% on sleep mode, and 30% on. This number is then multiplied by each device’s energy efficiency in order to calculate energy consumption (in Kwh) during the device’s use stage and then carbon emissions. (Energy Star 2010). By taking this into account, it would be possible in the future to achieve a EU adjustment, by using coefficients of energy grid emissions in different countries.

4. Refurbishment / Redistribution. This process consists of 3 phases. First, the transportation of a product from the seller to a local refurbishment center. Then the refurbishment process being carried out at the refurbishment center, and lastly the transportation from the refurbishment center to the new user - buyer. Information regarding this step’s carbon emissions was derived from Greener Scandinavia AB and Inrego AB data (Inrego 2014). For this step, a literature-based assumption was made regarding the average redistribution distances in Sweden, as also mentioned under 5.1 Assumptions and simplifications.

5. End of life - Disposal. Including transportation and energy demand for mechanical separation and shredding. Information for this stage was derived from industry data, specific for each device. As mentioned previously, industry data assumes that all benefits from using recycled material is taken into account in the Production stage, instead of here. It is assumed that a designated percentage of material is being recycled for each device, and the rest goes to landfill (Lenovo n.d.; Dell n.d.). This number varies between 50-75% depending on the device, and is mostly in accordance with scientific research that claims that usually around a 50% of a laptop’s materials can be recycled (Apple 2016; Baker & United Nations Environment Programme. Division of Environmental Conventions 2004). However, as different countries have different regulations regarding EoL this assumption can vary a little depending on device, even though most companies are making an effort to treat themselves their devices. E.g. Apple is claiming that in 99% of the countries that their products are sold, devices are mainly treated by Apple through a variety of take-back and recycle programs (Apple 2016).

(31)

Fig (5.2): Device flow chart

By using the above model then, it is possible to calculate the avoided emissions when reusing old IT equipment, as it is shown in Table (5.2): Results of model calculations for iPhone 7. More detailed calculations on this can be found under Appendix A: Model Calculations, where the final results show, that the total emissions avoided when reusing one old electronic device can be considered to be the solution to the following equation:

Total emissions avoided = Production + Transportation + Disposal - Refurbishment and Redistribution

During the model calculations, the functional unit (FU) is considered to be 2 use cycles of a specified device (1 laptop / smartphone / desktop / printer / monitor). Either in the form of two new, separate devices (Scenario 1), or one new device cycle and one reused (Scenario 2). In the following example (Table 5.2) 2 use cycles of one iPhone 7 (32GB model) used by each user for a 3 year lifespan is set as a FU.

Table (5.2): Results of model calculations for iPhone 7 (32GB model) Iphone 7 (32 GB)

Scenario 1 Scenario 2

Company A Company B Company A Company B

(32)

Production device 1 43.68 43.68

Transport device 1 1.68 1.68

Use device 1 (3y) 10.08 10.08 10.08

End-life / Dispose device 1 0.56 0.56

Refurbish / Redistribute

device 1 1

Production device 2 43.68

Transport device 2 1.68

Use device 2 (3y) 10.08

End-life / Dispose device 2 0.56

Total emissions (kg CO2) 112 67.08

Difference 44.92

Difference between Scenario 1 and Scenario 2: 44.92 kg of CO2 (Production + Transportation + Disposal - Refurbishment and Redistribution)

All Tables and calculations needed for the model can also be found under Appendix A: Model calculations.

5.2.2 Database compilation

A major part of the software tool development was the compilation of the database. The database is based on industry data and includes information regarding the carbon emissions of smartphones, laptops, desktops, printers and monitors throughout every life cycle stage of each device. The life-cycle stages included are described above, in 5.2.1 Calculation model. The database was input in Microsoft Excel in order to be connected with the software tool. Devices are divided by category (smartphones, laptops, desktops, printers and monitors) and brand name (Apple, Lenovo, HP, Samsung, Dell, Fujitsu, Lexmark, Canon, Xerox, Other). This information is stored in the first two rows of the database sheet, with the name and code of the device following. Right beside them, information is stored regarding the carbon footprint of the device’s production, transportation, use and disposal stage. Lastly, on the final row, the total carbon emissions avoided by reusing the device is displayed, based on the calculation model, as explained above in 5.2.1. The database sheet along with all the aforementioned information can be found under Appendix B: Database showcase.

In order to compile the database though, information was gathered from product manufacturers / institutions. Specifically, data used for the compilation was derived from the following product manufacturers / institutions:

(33)

Apple. The US-based company, apart from its yearly environmental report, releases product specific environmental reports for all its devices called Product Environmental Reports. The Product Environmental Reports calculates the device’s performance across three environmental priorities. Climate change, resources and materials (Apple n.d.). The climate change category describes how each stage of the product’s life cycle contributes to its carbon emissions (carbon footprint). Apple utilizes the Electronic Product Environmental Assessment Tool (EPEAT), and in accordance with ISO 14040 and ISO 14044 guidelines calculates the emissions (in CO2

equivalency factors (CO2e) of each life cycle stage as described in 5.2.1. All model assumptions (during production, transportation, use phase and disposal of the device) coincide with the assumptions of the project’s model (Apple 2016). This can also be observed under Table (5.3) Assumptions per brand.

Lenovo. Lenovo releases Eco-Declarations and product specific environmental reports, called Product Carbon Footprint Information Sheets (PCF) for a variety of its products, including laptops, desktops, tablets, workstations, servers and monitors (Lenovo n.d.). The Lenovo PCF is generated by using the Product Attribute to Impact Algorithm (PAIA) model, and reports the carbon emissions (in CO2 equivalency factors (CO2e) of each life cycle stage as described in 5.1.1. Again, all model assumptions are the same as this project’s model, and the rest of the manufacturers of the industry (e.g. Apple), establishing them as directly comparable to each other (Lenovo 2018).

HP. In an effort to understand and reduce the environmental impacts of its products, HP calculates and documents the carbon footprint of its devices at all stages of a product’s life cycle in their Product Carbon Footprint (PCF) Reports (HP n.d.). HP also uses the PAIA model to calculate the GHG emissions by its products throughout different stages during its lifetime, and again, all model assumptions are the same as this project’s model, as well as the rest of the manufacturers of the industry (HP 2018).

Dell. In 2010, Dell started to calculate and publish carbon footprints of various products, in order to understand and minimize the emissions of their devices. By using the EPEAT tool and the same assumptions as other major companies in the industry, starting back in 2010, Dell releases their carbon footprint reports in order to measure the total environmental impact of their products, throughout their lifecycle (Stutz 2010; Dell n.d.).

Other companies in the industry, such as Fujitsu, Lexmark, Canon and Xerox follow the same pattern, as they release their Product Carbon Footprint Reports and mainly follow the same methods and assumptions (FUJITSU 2011; JEMAI CFP Program n.d.; Lexmark n.d.). In some cases, a 5 year use stage was considered by the manufacturer, instead of 3 years. Even though the use stage does not impact the final model calculations of this project, adjustments were made and all models were rebased at 3 years use stage, in order to establish a complete and accurate database for future use. Table (5.3) Assumptions per brand (Also in Appendix F), includes a useful comparison across all brand calculations used during this project and their assumptions, as well the established model assumptions, as explained above in 5.2.1 Calculation model.