Environmental Impact of Computer Information
Technology in an Institutional Setting:
A Case Study at the University of Guelph
Melanie Adamson, Robert Hamilton, Kathryn Hutchison, Kaitlin
Kazmierowski, Joming Lau, Deigh Madejski, and Nicole MacDonald
University of Guelph
April 2005
1
Abstract
Computer use at the University of Guelph is an important aspect of campus life, however, its
environmental impacts are often not realized or considered. These impacts are expressed
throughout the manufacturing, use and disposal of on-campus computers, and thus require
monitoring and an understanding of each stage of a computer’s lifecycle. The computers located
in the various laboratories, libraries and faculty/graduate student offices at the University of
Guelph consume various quantities of energy, but as a whole are not operating at optimal
efficiency. In addition, the disposal of on-campus computers does not occur in the most
environmentally sound manner possible, thus resulting in various departments either diverting
unwanted units to landfills or storing them for extended periods of time. Both the inefficient use
of energy and the manufacturing and disposal of computer systems leads to the generation and
release of toxic compounds into the environment. This report identifies the need for the
implementation of campus-wide green procurement strategies with respect to computer
acquisition, use and disposal, and offers recommendations regarding improvements of the
University of Guelph’s current systems. The implementation of these recommendations will aid
the University in serving as an example for other institutions, saving money in the long run, and
decreasing its overall environmental impacts.
2
Acknowledgements
We would like to thank the following people for their support and aid during the completion of
this project: Dr. Joseph Ackerman, Avin Duggal, Gillian Maurice, David Fallow, Leon Loo,
Shelley Dano and the IT managers that responded to our survey. We would also like to thank the
Faculty of Environmental Science at the University of Guelph for their financial support.
3
Table of Contents
Abstract
2
Acknowledgements
3
1.0 Introduction
7
2.0 Background
2.1 Computers and Associated Environmental Problems
2.1.1 Energy Consumption
2.1.2 Physical Components and Toxins
2.1.3 Computer Manufacturing
2.1.4 Social and Political Implications
2.2 Green Procurement
2.2.1 The Acquisition of Green Computers
2.2.2 Power saving Techniques and Ecolabeling
2.2.3 End of Life Management
2.3 Case Study: Green Procurement Guidelines for the
University of Manitoba
13
13
14
14
17
20
21
23
24
26
31
3.0 Materials and Methods
3.1 Survey Design
3.1.1 Quantifying Computer Energy Use in Libraries
and Laboratories
3.1.2 Quantifying Computer Energy Use by Faculty
and Graduate Students
3.2 Statistical Analysis
3.2.1 z-Test Analysis
3.2.2 S-Plus Analysis
3.3 Environmental Impacts and Green Procurement Strategies
33
33
33
4.0 Results
4.1 Statistical Results
4.1.1 z-Test Statistical Results
4.1.2 S-Plus Statistical Results
4.2 Energy Consumption
4.2.1 Faculty and Graduate Students
4.2.2 Computer Laboratories
4.2.3 Libraries
39
39
39
41
43
44
44
45
5.0 Discussion
5.1 Statistical Analysis
5.1.1 Discussion for z-Test
48
48
49
34
37
37
37
38
4
5.1.2 Discussion for S-Plus
5.2 Energy Consumption
5.2.1 Faculty and Graduate Students
5.2.2 Computer Laboratories
5.2.3 Libraries
5.3 Computer Equipment Purchasing Guidelines
5.4 Computer Equipment Disposal Guidelines
50
50
50
50
51
51
53
6.0 Recommendations
6.1 Purchasing Computer Equipment
6.2 Energy Saving Strategies
6.2.1 Computer Laboratories and Libraries
6.2.2 Faculty and Graduate Students
6.3 Computer Equipment Disposal Methods
56
56
57
58
59
60
7.0 Conclusion
64
References
67
Appendix 1: Composition of a Personal Desktop Computer
71
Appendix 2: Ecolabeling Comparison
73
Appendix 3: Electronics Recyclers Pledge of True Stewardship
76
Appendix 4: Computer and Electronics Recycling
78
Appendix 5: Sampling Sites
80
Appendix 6: S-Plus Statistical Summaries for all Monitoring
Times in Richards Building
81
Appendix 7: Energy Consumption Raw Data
82
5
List of Figures
Figure 1: ENERGY STAR Symbol
24
Figure 2: Popular ecolabels under the Global Ecolabeling Network
25
Figure 3: Current Energy Consumption
46
Figure 4: Current Energy Consumption per Computer
47
Figure 5: Energy Savings for Conservation plans and New Computer
Equipment
47
List of Tables
Table 1: Components of CRT panel and funnel glass
15
Table 2: Basic steps in computer chip fabrication
19
Table 3: Resource Use in production of various computer components
20
Table 4: University of Guelph's Current Energy Consumption in Comparison
with the Worst-Case and Best-Case Scenarios
45
Table 5: Energy Consumption in University of Guelph Buildings
45
Table 6: Energy Consumption per Computer in University of Guelph Buildings
46
6
1.0 Introduction
Recently, topics such as global warming and climate change have drawn a lot of attention in the
media and general public. The production and use of various forms of energy is a large
contributor to greenhouse gas (GHG) emissions and climate change (BSD Global, 2002).
Institutions and organizations worldwide have begun to take measures to reduce energy
consumption and increase energy efficiency in an attempt to lessen their environmental impact.
Computers and office equipment play an increasingly large role in energy consumption. Desktop
computers, scanners and other electronic technology account for the fastest growing source of
energy consumption in Canada (NRCan, 2002). Although energy consumption is rising, there are
various methods that can be employed to increase energy efficiency. Many organizations and
institutions have implemented green procurement policies that promote the purchasing of energy
efficient products and the adoption of energy saving practices. These energy saving practices do
not reduce the performance of the computers, they simply reduce their power consumption when
not in use (Nordman et al., 1997). Most energy savings are derived from low power or 'sleep'
modes that occur when the computer is idle. Green procurement policies also require an
assessment of the environmental impacts of the products through all stages of its lifecycle
(cradle-to-cradle). An important element of this assessment is determining the end-of-life
disposal techniques available for various forms of office equipment, especially computer
monitors containing lead bearing cathode ray tubes (CRTs).
As the student population and computer usage increases at the University of Guelph, an
information technology (IT) strategy needs to be developed to address issues of energy
7
consumption by computers and the procurement and disposal of IT equipment. The University of
Guelph is facing a significant budget deficit (University of Guelph, 2005), and energy saving
techniques for computer technology could be applied to help reduce costs attributed to inefficient
energy practices. This project is especially significant due to the lack of similar studies at
educational institutions across Canada. As Canadian universities are becoming more dependent
on computer resources, they have the potential to save a significant amount of financial and
environmental wealth by using efficient and environmentally sound equipment.
Although general computer usage of computers at the University of Guelph is increasing, actual
values for energy consumption are unknown, as there currently is an unidentified number of
computers on campus that are left active for indeterminate lengths of time. The University of
Guelph has no large-scale energy conservation or cradle-to-cradle environmental efficiency
strategies. An appropriate strategy would include guidelines that integrate the acquisition of
energy efficient and environmentally responsible products, as well as environmentally sound
disposal methods for older computers and CRT monitors. While there is a abundance of
information regarding the recycling of older computer systems and CRT monitors, only a few
examples have been found regarding such strategies in the context of post secondary institutions.
This project aims to incorporate knowledge from previous case studies and implement strategies
with an on-campus perspective that consider the various demands associated with post secondary
institutions. Also, this project aims to provide the University of Guelph with recommendations to
reduce the energy consumption of on-campus computers, to purchase energy efficient computer
products, and to properly dispose of old computer equipment in an ecologically sound manner.
8
In order to achieve this goal, the objectives that we will address are as follows:
1. Quantify, to the best of our ability, the approximate energy use in University of Guelph
computer laboratories having greater than 20 computers, the libraries, and personal
computers used by faculty and graduate students
2. Compare current energy use to better-case scenarios according to the null hypotheses
3. Investigate potential end-of-life disposal and recycling techniques as well as, options to
dispose of toxic materials
4. Research the purchasing potential of energy efficient and environmentally responsible
computer equipment
5. Explore energy conservation measures that reduce power consumption in computer
laboratories and personal computers across campus
To achieve these objectives, this project was undertaken using several important assumptions.
Firstly, laptop computers available for student usage in the library were not taken into
consideration for our study. It was beyond the scope of this study to obtain an accurate estimate
of energy consumption, as much of the power requirements for laptops are met through battery
power. Personal computers in residence were also not included in our study as their energy
consumption varies year to year, and energy saving techniques would be difficult to implement.
Secondly, computer use varies at different times during any given semester, and throughout the
academic year. Student workloads and computer usage are subject to variability. This is an
important point to consider, as the results found in this study correspond with weeks nine and ten
of the winter semester and may not be representative of computer use at other times throughout
the academic year. Finally, and perhaps most importantly, all computers surveyed in this study
9
are assumed to follow the same ratio of new liquid crystal display (LCD) monitors to old CRT
monitors as identified in the MacLaughlin Library. Due to the constraints of this project, the
MacLaughlin Library was used as a sample to quantify the usage of energy efficient LCD
monitors throughout the University of Guelph campus. Other computer laboratories, faculty and
graduate students are also assumed to follow this pattern.
This study seeks to rank the University of Guelph’s current computer efficiency on a scale
between a ‘worst-case scenario’ and a ‘best-case scenario’. For the purpose of our study, we
have defined the worst-case scenario as all computer systems at the University of Guelph using
old CRT monitors, old central processing units (CPUs), not making use of power saving
strategies such as ‘sleep’ and ‘standby’ mode, and are active 24 hours per day, 7 days per week.
This worst-case scenario also lacks of provisions for acquiring energy efficient products and for
environmentally sound disposal methods of computer equipment. We have defined the best-case
scenario as all computers on campus having LCD energy saving monitors, new CPUs being
ENERGY STAR certified, using energy saving techniques, and being active 8 hours per day, 5
days per week. ENERGY STAR certified technology allows computers to automatically switch
to standby mode when inactive for a certain amount of time, and thus allowing for energy
savings. The best-case scenario also includes provisions for acquiring energy efficient products
and disposing of computer equipment in an environmentally sound manner. By stating these
scenarios, this study is able to make comparisons between the University of Guelph’s current
computer energy consumption with the potential energy consumed within the best and worst-case
scenarios. The fundamental premise behind these comparisons is that the University of Guelph is
not running at optimal energy efficiency, and that through increased power management
10
techniques, the purchasing of energy efficient products and the usage of proper disposal
techniques, the University of Guelph can improve its current practices. Explicitly stated, our null
hypothesis is the following:
The University of Guelph's current practices will be the same as the best-case
scenario for energy consumption and cradle-to-cradle environmental efficiency.
This null hypothesis is the basis for this report; however, several other comparisons will be made,
with two sub-null hypotheses being identified:
1. Conservation plans alone cannot reduce the energy required to power computer usage at the
University of Guelph.
This sub-null hypothesis compares the worst-case scenario with a scenario using CRT monitors
and old CPUs, but utilizing energy saving techniques such as shutting the computers down at
night.
2. New computer equipment alone cannot reduce the energy required to power computers at the
University of Guelph.
This sub-null hypothesis compares the worst-case scenario with a scenario where all computers
on campus use LCD monitors and Energy Star certified CPUs, but are left active for 24 hours a
day, 7 days a week.
11
The re-evaluation of the University of Guelph’s energy conservation strategies and computer
disposal methods is significant. Not only can it save the University money, but it will also
perpetuate its excellent reputation as an environmentally and ecologically conscientious
institution. Such measures will allow the University of Guelph to act as an example of an
institution demonstrating cost-effective green procurement strategies.
12
2.0 Background
While the environmental issues involved in computer manufacturing, use, and disposal employ
large quantities of fossil fuels and hazardous wastes, a new push towards the “greening” of the
various components of the computer industry provides hope and practical strategies for the
future.
2.1 Computers and Associated Environmental Problems
The environmental problems associated with computers are two-fold. High energy consumption
and highly toxic component materials are currently inherent characteristics of computers, thus
making their production, use and disposal ecologically unsound (Lee et al., 2004).
Unfortunately, due to their sheer global quantities and current product life of roughly two years,
the problems associated with such characteristics become greatly enhanced at an alarming rate
(Brennan et al., 2002). Zhang and Forssberg (1999) projected that by 2005, roughly 150 million
personal computers (PCs) and workstations will be disposed in landfills in the US alone. By this
same year, Gungor and Gupta (1999) predicted that every family in the US will own a computer,
and given the aforementioned product life of these systems, it appears that computers are being
disposed of as quickly as they are being produced.
Unfortunately, disposal in landfills is only the first step in a dangerous sequence of events
involving the breakdown and leaching of computer material components. Examples include lead,
barium, chromium and other endocrine and central nervous system disruptors (Baul, 2002).
Aside from hazardous wastes, the production and use of computers consumes vast amounts of
13
energy, thus further depleting fossil fuel reserves and playing an increasingly significant role in
climate change and global warming (Gungor and Gupta, 1999).
2.1.1 Energy Consumption
Globally speaking, the issue of energy consumption is one that involves all sectors and industries.
According to Norfold (1990) and Kawamoto (2002), electronic office equipment such as desktop
computers use significant amounts of electric power. A typical CPU uses 120 Watts (W = 1
joule/second) of electricity, while a CRT monitor consumes an added 150 W (United States
Department of Energy, 2005). This implies that a standard office computer which is left on 8
hours per day, for 5 days a week can consume up to 561.6 kW of fossil fuel derived energy.
However, this figure more than triples if such a computer is left on throughout the night or during
the entire week.
2.1.2 Physical Components and Toxins
Desktop computers generally consist of three major units: the main processing machine (CPU
consisting of power supplier, fan, IC boards, DVD drive, CD drive, hard disk, soft disk and shell
casing), the monitor and the keyboard (Lee et al., 2004). However, as demonstrated in Appendix
1, these major units are composed of various materials, which, in turn consist of a wide range of
chemicals, elements and heavy metals. Some of these materials, such as platinum, have a high
recovery and recycling efficiency (95%), while others cannot be recycled at all (e.g. mercury,
arsenic and barium). There are, however, two desktop components that represent the largest
environmental hazards with respect to bioavailability, monitors containing CRTs and flame
retardant plastics (Lee et al., 2004).
14
Cathode Ray Tubes
Since the 1950s, CRTs have been used in television and computer screens. Historically, their
production has grown in step with computer demand (Williams, 2003). In 2001, the global CRT
monitor industry was valued at US $19.5 billion, producing 108 million units. This figure is
expected to fall due to the increasing popularity of LCD monitors (Williams, 2003).
The CRT of a typical monitor accounts for approximately 50% of the monitor’s weight, and
contains a veritable cocktail of elements (Table 1) of which lead is considered the most important
due to its high content (up to 20%) in the funnel glass component of a CRT (Lee et al., 2004).
Table 1: Components of CRT panel and funnel glass (reconstructed from Lee et al., 2004)
Type of Glass
Major Elements (>5% wt)
Minor Elements (<5% wt)
Panel
Silicon, oxygen, potassium,
Titanium, sodium, cesium,
barium, and aluminium
lead, zinc, yttrium, and
sulphur
Funnel
Silicon, oxygen, iron, and
Potassium, sodium, barium,
lead
caesium, and carbon
In most basic terms, a CRT creates the visual image displayed by the monitor, by employing the
interaction between an electron tube and a phosphor coated screen (Anonymous, 2003). In order
to avoid radiation exposure to the viewer, the funnel glass of the CRT contains high
concentrations of lead-oxide (Lee et al., 2004). According to the US Environmental Protection
Agency’s (EPA) toxicity characteristic leaching procedure (TCLP), the lead found in funnel glass
is considered a hazardous waste because it far exceeds the TCLP threshold of 5 mg/L leached,
with values ranging from 10-20 mg/L leached per monitor (Lee et al., 2004). Williams (2003)
15
also found that CRT monitors exceeded TCLP limits for zinc leachate, thus classifying it as a
hazardous waste. The hazard truly occurs when monitors are permitted to weather in landfills,
releasing these toxic chemicals into soil, and subsequent water systems.
Lead is especially an issue in waste disposal because it becomes bioavailable in soils with
increasing pH, and becomes available to animals and humans through the food chain and soil
dust inhalation (Martinez-Villegas et al., 2004). Once in the body, it can attack proteins and
DNA (Bechara, 2004) as well as interfere with the functions of the central and peripheral nervous
systems (Needleman, 2004). At high enough doses, it can result in brain edema and haemorrhage
(Needleman, 2004).
Liquid Crystal Display
The global shipment of LCDs, also known as “Flat Screen” monitors, is projected to surpass that
of CRT monitors by 2007. In 2001, the global market for LCDs was valued at US $9 billion and
totalled 12 million units (Williams, 2003). While LCDs are preferred for their efficient use of
space, thus allowing more to be shipped at once, they also contain significant amounts of
mercury (4-12 mg/unit), which can be leached from improperly discarded systems. Mercury is
already a problematic substance in US landfills since in 2000, it was estimated that 172 tonnes
were accumulating in locations across the country (Williams, 2003). Additionally, the
production of an LCD monitor requires 266 kg of fossil fuels, a figure that surpasses that required
for the production of CRT monitors (Williams, 2003).
16
The liquid crystals within an LCD monitor are a mixture of polycyclic or halogenated aromatic
hydrocarbons, and contain 588 various compounds. However, of these, only 26 possess the
potential for acute toxicity in humans (Williams, 2003). While no tests for the carcinogenicity of
these compounds have been conducted on animals, tests using bacteria showed no trace of
mutagenic effects (Williams, 2003).
Plastics and Casings
Most electronic equipment contains plastic casings that serve as the protective shell and structure
for various products including computers (Brennan et al., 2002). These casings often contain
plastics such as polybrominated diphenyl ethers (PBDEs); part of a wider group of materials
known as brominated flame retardants (BFRs) (Domingo, 2004). While BFRs are considered a
safety precaution, they are difficult to recycle and separate from other plastics, and due to their
high bromine content, will be banned from the European Union as of July 1, 2006 (Osako et al.,
2004). Very little is known about the effect that BFRs exert on human health, however, due to
their long half-lives (2-10 years) and structural similarities with polychlorinated biphenyls
(PCBs) and dichloro-diphenyl-trichloroethane (DDT), they are considered environmentally
persistent and are known to biomagnify (Domingo, 2004). BFRs have caused
neurodevelopmental toxicity in lab rats, and have been found in increasing quantities in human
blood, adipose and liver tissues, and in breast milk (Domingo, 2004).
2.1.3 Computer Manufacturing
In order to obtain an accurate measure of the environmental impact of computer technology, the
production process of computer technology must be examined. From the extraction of raw
17
materials, to the production of computer parts and constituent materials, and the final assembly of
whole computer units, a myriad of environmental problems arise (Fava et al., 1993).
The impact of computer technology largely depends on the scope used in analysis, and can
involve an assessment of a system’s entire life cycle, including the environmental impacts of
preliminary manufacturing activities. These include the mining and smelting of raw materials,
and the refining of petroleum to provide the vast amounts of energy required to produce and use
computers (Curran, 1996). However, it is the unique manufacturing processes from which
computers themselves are derived, that exerts further impacts and thus requires an in-depth
understanding.
While the environmental impacts associated with the production and disposal of CRTs, LCDs
and plastics are highly significant; the production of other computer components must also be
explained in order to fully grasp the extent of the environmental impacts imposed by their
fabrication.
Microchip Fabrication
CPU function is based on microchips, and it is the fabrication of these components that cause the
largest environmental impacts related to manufacturing (Geiser, 2001). Over 400 individual
processing steps exist in the production of semiconductor microchips, but the basic process
involves a sequence of layering, oxidation and patterning processes and is explained in Table 2.
18
Table 2: Basic steps in computer chip fabrication (reconstructed from Williams, 2003)
Process
Description
Layering
Application of a thin layer of desired material, usually
silicon or aluminum
Oxidation
Changes a semi-conducting silicon layer into a
insulating silicon dioxide layer
Patterning
Carving of a dense, maze-like set of furrows into a
layer
Etching
Use of solvents or particle bombardment to alter the
layer patterns
Between each step, microchips are processed with large amounts of ultra-pure water. Microchips
are then bathed in a wide range of chemical solvents in order to ensure their purity, as any small
defect on its surface can hinder its function (Geiser, 2001). Chemical solvents used include:
hydrochloric acid, hydrofluoric acid, arsenic, benzene and hexavalent chromium; many of which
are known to cause deleterious environmental and human health effects (Williams, 2003).
Circuit Board Fabrication
Printed circuit boards are responsible for connecting microchips and other components of the
computer. The physical base of the circuit board is made of an insulating material sandwiched
between thin copper layers (Williams, 2003). The fabrication process of printed circuit boards is
quite similar to that of microchips. Metals such as copper, lead, silver, tin and chromium, as well
as PBDEs, act as flame retardants incorporated into the manufacturing process, as well as, in the
unit itself. These various materials are bioaccumulative and along with neurodevelopmental
problems, can cause thyroid disruption (Darnerud et al., 2001).
19
Resource Use in Manufacturing
In many manufacturing processes, environmental impacts will arise through the intensive use of
chemicals, energy, or water resources at each step of production. Table 3 shows the general
amounts of resources used in the manufacturing of various computer components (Williams,
2003). In many cases, this intensive resource use is necessary to ensure that high-grade
chemicals and components are employed in the manufacturing process. Chemicals used in
computer fabrication are of purity typically in the range of 99.995-99.9999%, compared with
industrial grade purities of 90-99% (Williams, 2003).
Table 3: Resource Use in production of various computer components (Reconstructed from: Williams, 2003)
Component
Fossil Fuels (kg)
Chemicals (kg)
Water (kg)
Computer Chips
94
7.1
310
Printed Circuit Boards
14
14
780
CRT monitors
31.5
0.49
450
LCD monitors
226
3.7
1290
2.1.4 Social and Political Implications
The 1980’s proved to be an era of tightening environmental regulations with respect to the
transboundary disposal of hazardous wastes (Secretariat of the Basel Convention, 2005).
Unfortunately, as prices for hazardous waste disposal rose, so did the occurrence of “toxic
traders”; the practice of shipping electronic wastes to developing countries where they are
cheaply and manually sorted (Secretariat of the Basel Convention, 2005). On May 5, 1992, the
Basel Convention, under the United Nations Environment Program, came into force in order to
control the movement of hazardous wastes across international borders (Secretariat of the Basel
Convention, 2005). In 1995, the Basal Action Network (BAN) Amendment was drafted in order
20
to prohibit the export of hazardous waste from the European Union, Liechtenstein and what are
known as Organization for Economic Cooperation and Development countries (The OECD
countries include Australia, Austria, Belgium, Canada, Czech Republic, Denmark, Finland,
France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Japan, Republic of Korea,
Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Poland, Spain, Sweden,
Switzerland, Turkey, the UK and the US) to all other parties of the convention (Secretariat of the
Basel Convention, 2005). Adherence to the convention is voluntary, and of the 164 parties, only
Afghanistan, Haiti and the US have not yet ratified it (Secretariat to the Basel Convention, 2005).
Unfortunately, many of the stipulations of the Convention can be interpreted in various ways,
thus paving the way for “toxic trading” to continue (Secretariat of the Basel Convention, 2005).
Baul (2002) estimates that even with the Basel Convention in place, roughly 200 tonnes of
obsolete computer waste are shipped to South East Asia each year, of which Canada is a
contributor. The reported recycling practices in such countries include manually smashing
monitors and the melting of circuit boards over open flames; practices detrimental to human and
environmental health (Baul, 2002).
2.2 Green Procurement
In order to effectively manage the vast quantities of computer waste being generated annually, at
a global scale, the implementation of proactive and design-based measures has begun in several
countries. The creation of environmentally friendly products and waste recovery techniques has
become increasingly important aspects of computer production, use, and disposal. This is due to
a decrease in the number of available landfill sites, society becoming more environmentally
21
aware, and increasing scarcity of non-renewable energy sources (Gungor and Gupta, 1999). One
term which incorporates all these factors is “green procurement”. Green procurement is the
environmentally responsible selection of products and services with consideration of the
consequences of such a product throughout the various stages of its life cycle (BSD Global,
2002). This implies that the various ecological costs of securing raw materials, manufacturing,
transporting, storing, handling, using and disposing of a product, must not only be considered,
but become inherent parts of that product’s design. Examples include: designing computers
which can easily be broken down for recycling and are less hazardous to recycle due to lower
levels of toxic components (Lee et al., 2004), or designing efficient systems which effectively
separate and recover recyclable components from obsolete systems (Zhang and Forssberg, 1999).
Equipping computer systems with internal energy regulation devices (SVTC, 2005) and ensuring
that human and environmental health are not compromised during recovery and end-of-life
management, through policy and public education, (Nagel and Meyer, 1999), are all significant
components of green procurement.
Green procurement strategies have already been demonstrated in countries such as Germany and
Taiwan, where producers are responsible for the recycling of obsolete computers, thus creating an
incentive to produce products which can be easily recycled (Lee et al., 2004). The various
“ecolabels”, such as ENERGY STAR, ensure that computers are running under energy
conservation measures (SVTC, 2005), and even Canadian post secondary institutions such as the
University of Manitoba have recently employed green procurement strategies in an on-campus
setting (Searcy, 2001). Due to its broad definition, green procurement can be applied to all facets
22
of the computer industry, thus making it a model strategy not only for the management, but the
elimination of computer wastes.
2.2.1 The Acquisition of 'Green Computers'
The acquisition of computers using green procurement strategies can potentially reduce their
environmental burden. By supporting manufacturers who utilize “green” production processes,
these environmentally sound means of production can be further propagated and perhaps
eventually become the industry standard (Saied and Velasquez, 2003).
Many organizations today are beginning to develop and adopt green procurement policies, as
they can result in significant cost savings from more energy efficient technology. Also, growing
public environmental awareness requires environmental and social accountability, thus requiring
producers to take responsibility for their products (Gungor and Gupta, 1999).
There are several manufacturing options currently available which employ fewer toxic chemicals
in their various processes. Methods of manufacturing and purifying computer chips have been
developed to fit this strategy, and the rise in PBDE-free plastic casings reflects the move towards
alternative processing as well (Shigekazu, 2004). Computer components that contain lead-free
soldering are also being produced as a means of reducing this toxic and harmful substance
(Griese et al, 2000).
Manufacturing is only one component in the successful acquisition of ecologically sound
computer equipment. Another important factor is the need for extended producer responsibility.
This concept recognizes that manufacturers and brand-owners are responsible for their equipment
23
throughout all stages of its lifecycle, and should therefore strive to produce computers in an
environmentally conscientious manner in order to avoid future negative environmental impacts
for which they will be responsible (Gungor and Gupta, 1999). This involves producer
understanding of the product’s environmental impacts throughout all stages of its lifecycle, as
well as the implementation of environmental product design in order to minimize these impacts at
the beginning of the lifecycle (Gungor and Gupta, 1999). The acquisition of environmentally
sound computer products should be based upon these various practices, thus creating a demand
for such products in the market and eventually leading to the implementation of these practices
into all computer production processes (Saied and Velasquez, 2003).
2.2.2 Power Saving Techniques and Ecolabeling
The ENERGY STAR symbol (Figure 1) is an increasingly common sight on the
outer surfaces of various computer systems. The ENERGY STAR program was established in
1992 by the U.S. EPA in order to reduce the amount of energy used by computer systems
(USEPA, 2005). ENERGY STAR power management features
Figure 1: ENERGY STAR
are now standard in Windows and Macintosh operating systems
Symbol
and all the buyer must do is ensure that they are enabled. These
power management features place inactive computers and
monitors into a “sleep mode”. While in sleep mode, the computer consumes 15% less power
than when in regular mode (USEPA, 2005). The computer or monitor can be “woken up” simply
by touching the keyboard or moving the mouse.
Some advantages of using ENERGY STAR qualified computers include a 70% reduction in
electricity, and equipment runs at cooler temperatures and therefore, lasts longer because the
24
systems are not continually running at full power (USEPA, 2005). However, some
misconceptions and concerns have arisen from the use of ENERGY STAR management
practices. While users may believe that repeatedly shutting down and starting up a computer will
shorten its lifespan, this is not the case. Modern computers are designed to be able to withstand
40,000 on-off cycles without technical problems (USEPA, 2005). Another common
misconception is that screen savers save energy. Screen savers are designed to maintain
computer screens, not conserve energy (USEPA, 2005). With current technological
developments, screen savers are more a form of entertainment than a necessity. Another
common concern is that computers and monitors that are ENERGY STAR qualified will cost
more money upon purchasing. This is also untrue, as there is no increased cost for having these
systems integrated into a personal computer.
Ecolabels, however, can represent more than simply power saving techniques, and can be applied
to many products including household paints and paper products (SVTC, 2005). Many countries
employ various ecolabels for desktops and laptops alike. Figure 2 shows examples of some of
the most popular ecolabels found under the Global Ecolabling Network: German “Blue Angel”,
European Union, Nordic Swan and Swedish Confederation of Professional Employees (TCO).
Figure 2: Popular ecolabels under the Global Ecolabeling Network (From SVTC, 2005).
25
Appendix 2 contains full descriptions of these ecolabels and the environmental guarantees they
entail.
2.2.3 End of Life Management
The disposal of computers is a unique issue due to the fact that most computers are often
disposed of before they truly become useless. In fact, the main reason for purchasing a new
computer is not to replace a non-functioning system, but to keep up with rapidly changing
technologies (Williams and Sasaki, 2003). One key term which is important for industry, the
government, and the public, with respect to computer disposal, is “upstream management”; the
various methods employed to reduce the amount of in-coming computer wastes before they are
disposed of for good (Williams and Sasaki, 2003). These methods embody the concept of
Reduce, Reuse, Recycle, and have proven to yield many benefits, both environmental and
socioeconomic.
Reduction and Reuse
Reducing the amount of computer waste relies heavily upon the reuse of systems that may be out
of date, but fully functional. Reusing old computers can manifest itself in two main ways; by the
selling or donation of old systems, or by up-grading existing systems (Williams and Sasaki,
2003). The key concept with respect to reuse is to meet the user’s needs with existing machines,
while extending that machine’s lifespan.
There are a number of organizations around the world, which focus on the redistribution of old
computers. In the 1990’s, Williams and Sasaki (2003) found that schools and small businesses
were generating the greatest demand for used computers. The Canadian federal government
26
initiative, “Computers for Schools”, was founded in 1993, and since then has collected, repaired,
refurbished and redistributed over 500 000 computers across Canada (CFS, 2005). Currently,
Computers for Schools delivers 80 000 additional computers each year to schools, public libraries
and non-profit organizations, for a cost of CDN $85 per computer (CFS, 2005). The National
Cristina Foundation (NCF) in partnership with Dell computers, carries out both computer redistribution and recycling, and also donates systems, as well as provides, IT training to disabled
and disadvantaged members of the community (Dell, 2005).
Reselling old computers is another way to extend a working PC’s lifespan, however, many re-sell
agents only accept the newest generation of equipment (Williams and Sasaki, 2003). Computer
Renaissance is the largest chain of re-sell shops in North America (110 stores), and even original
equipment manufacturers such as IBM, Dell and Hewlett Packard/Compaq sell their own
refurbished systems on-line (Williams and Sasaki, 2003). In 2002, eBay, the largest on-line
auctioneer sold US $2 billion worth of computer equipment, of which 40% was new, 14% was
refurbished and 46% was used (Keafe, 2002).
Upgrading a computer in order to suit current technology is a choice often made only by
computer specialists or hobbyists, due to the fact that user knowledge is required and full
upgrades can be costly. While upgrading individual components of a computer, such as the hard
drive or processor, may cost less than a new PC; upgrades which involve the addition of a
Universal Serial Bus (USB) port result in the “complete upgrade” and often cost more than a new
system (Williams and Sasaki, 2003).
27
Recycling and Reduction
As rates of computer disposal appear to keep in-step with the rate at which technology changes,
the disposal of computers in landfills is no longer an option for end of life management (Lee et
al., 2004). With the introduction of increased producer take-back legislation, or “extended
producer responsibility” in such countries as Germany and Taiwan, making use of efficient and
effective recycling strategies is now more economically and ecologically important than ever
(Zhang and Forssberg, 1999). Recycling is always more economically beneficial if it gives
priority to whole-product recycling (i.e. re-use), however, this cannot always be the case;
therefore efficient material recycling techniques are needed (Williams and Sasaki, 2003).
One of the main problems encountered during the computer recycling process is the effective
physical separation of components. The concept of “Intelligent Liberation” is presented by
Zhang and Forssberg (1999) as a means of accurate mechanical separation of computer scrap
components, such as printed circuit boards, for further recycling. Materials used in computer
equipment are held together via weak interfacial bonds such as welding, fastening and wrapping,
and therefore intelligent liberation of these materials can occur via low energy methods. One
method utilizes a ring shredder which is capable of producing particles of desired shapes and
sizes which can be further separated and recycled (Zhang and Forssberg, 1999). Once these
“liberated” materials are obtained, they can be further separated through eddy current separation;
a method in which an eddy current is created through the spinning of magnets at over 3000 rpm.
This current reacts differently with different metals, thus allowing for separation to take place at a
95% recovery rate for particles as small as 2 mm in diameter (Anonymous, 2003). Other
28
mechanisms of separation include sink-float density-based separation and screen-aspirator
systems (Zhang and Forssberg, 1999).
CRT monitors are another component that require careful recycling measures due to their high
lead content (Lee et al., 2004). Since panel and funnel glass contain different amounts of lead,
they should be separated by either the electric-wire heating method or the gravitational fall
method (Lee et al., 2004). Both methods may be effectively employed, however, using a super
heated wire to separate the glass, often achieves cleaner separation (Lee et al., 2004).
Once all separation and grinding has occurred, individual metals are separated by electrolysis and
recycled accordingly (e.g. ferrous materials are reprocessed in traditional steel-waste processing
plants or smelters), while cleaned CRT glass is sent to ceramic plants where it is re-used as raw
material for future production (Lee et al., 2004). While these processes are effective, some tend
to generate waste waters and residuals which are difficult to treat and dispose of (Klatt, 2003).
Effective Recycling and Reduction Options
Many countries have ratified the Basel Convention and have even initiated legislation forcing
producers to take back old systems, however, the problem of toxic traders is still very real
(Kuehr, 2003). For this reason, in 2003, the “Electronic Recycler’s Pledge of True Stewardship”
was first launched by the Basel Convention (BAN, 2005). Those who take the pledge must agree
to meet rigorous criteria in order to recycle electronic waste in the most sustainable and socially
just manner possible. The pledge itself can be found in Appendix 3. Maxus Tech Inc. and Retro
29
Systems in Alberta, and Logic Box Distribution Inc. in Mississauga, Ontario, are the only three
Canadian recycling companies have taken the pledge thus far (BAN, 2005).
Logic Box accepts any type of computer and peripheral devices free of charge (Chezzi, 2004).
The goal at Logic Box is to dismantle, refurbish and repair computers for re-sale, or individual
parts within the tenets of the pledge (Chezzi, 2004). When electronic devices reach the stage
where recycling is the only option for end of life management, they are shipped to the
Mississauga based Electronic Product Recovery Services (EPRS). Here, a specially designed
shredder sorts metals within electronic devices as well as associated packaging products, such as
cardboard and shrink wrap (Chezzi, 2004). Even the dust produced from this system is captured,
compressed and re-processed as to achieve a no-waste standard. Not only are these companies
developing their own environmentally sound technologies, but they are also local, thus providing
potentially viable options for the future.
No matter how many environmentally sound re-use and recycling schemes are implemented,
computer wastes will continue to be a serious environmental and social problem, unless the
public becomes educated and informed regarding these processes and the reasons behind them.
In turn, the choices of environmentally informed consumers will play a huge role in the
“greening” of the computer industry (Kuehr, 2003). As previously mentioned, ecolabels provide
environmental symbols which can be easily recognized by consumers. In fact, many labels,
including Norway’s “Nordic Swan” and Sweden’s “TCO”, provide user instructions regarding
available take-back procedures, upgrading options and energy saving techniques (SVTC, 2005).
30
One original idea regarding public education is called the “Green Port Identification Unit” (Nagel
and Meyer, 1999). This theoretical system would allow consumers to “plug and recycle”;
meaning that when an electronic product has reached its end of life, it could be hooked up to
software that would enable the customer to read all information about its structural and chemical
components, as well as its life time (hours of operation, physical or temperature shocks). The
consumer would then be provided with information regarding the appropriate recycling stream
for the product; thus making an environmentally informed decision (Nagel and Meyer, 1999).
Public education is a key component of green procurement because it can exert a strong influence
on industry and therefore bring about ecological change (Kuehr, 2003). Unfortunately, the
general public has yet to realize the ecological effects of this influence; however, increasing
education and awareness will aid in the realization of this.
2.3 Case Study: Green Procurement Guidelines for the University of Manitoba
The University of Manitoba recently completed a draft report entitled The University of Manitoba
Campus Plan – Network Services (Searcy, 2001). As part of the plan, the University
acknowledged the need for stewardship of the physical environment and the development of
green procurement guidelines. As one of the fundamental concepts of green procurement, each
guideline was created based on sound environmental principles that take into account the entire
life cycle of a product. Therefore, impacts related to the extraction and processing of raw
materials, design, manufacturing, packaging, transport, distribution, installation, use,
maintenance, recycling, reuse, and final disposal of the product were all considered in the
development of guidelines (Searcy, 2001). In addition to criteria associated with environmental
31
impacts, it is important to note that traditional performance requirements, such as cost and
availability, were also considered.
Two broad categories of criteria for procurement were identified: procurement of products and
procurement of services. Several green procurement guidelines for office equipment were
included in the plan:
ƒ
Preference must be given to office products which bear ecolabels, are upgradeable, and
contain recycled materials
ƒ
Suppliers which offer a take-back or trade-in program must be sought out and given
preference
ƒ
Products which use minimal packaging and are energy efficient must be given preference
ƒ
Supplies should be bought in bulk in order to minimize wastes from packaging as well as
fossil fuel emissions from transportation
ƒ
Office equipment that is easily dismantled must be selected in order to encourage efficient
reuse, refurbishment or recycling
By following these green procurement practices the University of Manitoba should reduce the
overall impact that it has on the environment (Searcy, 2001). However, for this to effectively
occur, environmental considerations must be integrated with existing purchasing practices and be
consistent with such traditional factors as product safety, price, performance, and availability
(Searcy, 2001). Through the implementation of these guidelines, potential benefits such as
reduced environmental impacts and improved energy efficiency may be realized.
32
3.0 Materials and Methods
The materials and methods of this study were designed to test the following null hypothesis:
The University of Guelph's current practices will be the same as the best-case scenario for
energy consumption and cradle-to-cradle environmental efficiency.
3.1 Survey Design
3.1.1 Quantifying Computer Energy Use in Libraries and Laboratories
Upon our request, the departmental manager at Computing and Communication Services (CCS)
sent out an e-mail survey to all the information technology (IT) managers at the University of
Guelph regarding computer energy use in the libraries and laboratories. Laboratories with less
than 20 computers were not considered because the number of these that exist on campus is too
difficult to determine. Questions in the survey included:
•
Which computer lab(s) are you responsible for?
•
How many hours a day are the computers in this lab powered on for?
•
How many days a week is the computer lab open for?
•
Do the computer systems have a stand-by/energy saving mode that is used if the
computers become inactive? And if so, how many hours a day do you suspect that it is on
stand-by/energy saving mode?
E-mail responses were received from the following IT managers: Crop Science, Graham Hall,
McLaughlin Library, Ontario Veterinary College (OVC) library, and OVC microcomputer
33
laboratory. Responses from IT managers were not received from the following buildings: Hutt,
Mackinnon, MacNaughton, and Powell. A physical count was necessary to estimate the quantity
of computers in Hutt, Mackinnon, MacNaughton, and Powell (Appendix 5). It was assumed that
the length of the computer activity was the same as the posted hours in each lab. Due to the
difficulty of accurately determining the amount of time the computers were on stand-by/energy
saving mode, it was assumed that they were left active for the entire time the laboratory was
open.
There was a slight discrepancy in the number of computers reported by the McLaughlin library’s
IT manager to the number reported on the library’s website. To verify the quantity of computers
in the library a physical count of the library’s student used computers was performed.
From speaking to various IT managers and doing physical quantification of computers on campus
it was discovered that there are a variety of computer systems on campus. There are some very
old CPUs and old CRT monitors, as well as some new CPUs and LCD monitors. To quantify
energy consumption, IT managers were asked about the proportion of CPUs that are new along
with LCD monitors. In the MacLaughlin library computers were counted and LCD and CRT
monitors were noted.
3.1.2 Quantifying Computer Energy Use by Faculty and Graduate Students
An approximation of the current number of faculty and graduate students on campus was
obtained on the University of Guelph’s Resource Planning and Analysis website (University of
Guelph: Office of the President, 2004). Fall 2004 statistics were presented for on-campus full
time and part time graduate students, as well as faculty members.
34
Since a campus wide survey for faculty and graduate students was not temporally feasible for this
study, one building on campus was selected as a model for energy use in all buildings. The
Richards Building was selected as this model for faculty and graduate students on campus
because it is assumed that this building represents standard computer energy consumption for all
buildings. This assumption was based on the hours of operation of the Richards Building (8:30
am - 4:30 pm), which is typical for most buildings on campus. The faculty and graduate staff of
the Richards Building represent 4.23% (120 computers) of the University of Guelph's total
faculty and graduate student population. The results for the Richards Building can be used as a
representation of the amount of time that all faculty and graduate student computers are active.
The Richards Building was also selected because it is the easiest place to obtain data for the
researchers.
Computer energy use was measured using Netscan version 2.4, a program that scans for internet
protocol (IP) addresses that are in use at the time the scan is performed (Netscan, 2005). This
program, therefore, has the ability to track computers that are active in a given IP range. During
this study, a lab technician in the Richards Building provided the IP range of the computers in
that building’s network. This allowed for the location of all active computers on the network in
the Richards Building at the times the scan was performed. Further assistance was given by the
same lab technician to determine which of the IP addresses were indeed computers and not
networked printers or departmental servers, which are also assigned IP addresses. The precision
and accuracy of this program was first tested on a personal network to ensure that it was
effectively picking up all active computers. From this control it was discovered that the program
35
is not capable of distinguishing between offline computers and those in energy saving mode.
Therefore, if a computer is not showing up on Netscan it was assumed to be off, as most energy
saving modes use negligible quantities of energy. This program was run three times a day at
approximately 10:00 am, 2:00 pm and 7:00 pm, for one week (Tuesday March 15, 2005 to
Monday March 21, 2005) to obtain a representative sample of computer usage at different times
of the day.
Assumptions used:
ƒ
The proportion of new CPUs and LCD monitors for the faculty and graduate students are
the same as those found in the MacLaughlin library
ƒ
The Richards Building uses only static IP addresses
ƒ
All computers in the Richards Building are used by faculty and graduate students and
there is one computer per person
The results of the IP scanning period are a realistic representation of the faculty and graduate
student population of the University because:
1. It is assumed that all faculty and graduate students have similar schedules and
responsibilities. Faculty and graduate students from every building on campus will have
to attend conferences, perform lab work and, fieldwork, and will have sick days.
2. It is assumed that the only differences between all faculty and graduate students are the
subject matters with which each individual works.
36
3.2 Statistical Analysis
3.2.1 z–Test Analysis:
Data obtained from the e-mail surveys distributed to departmental managers by CCS were used to
conduct 3 separate one-tailed z–tests. These tests were used to determine if there was a
significant difference between current computer usage by libraries, computer laboratories, faculty
and graduate students and the best and worst-case scenarios. The tests were performed at a 95%
significance level and the z–critical value was 1.96.
For the best-case scenario, the null hypothesis (for statistical purposes only) is:
Ho: current energy consumption by computers on campus (libraries, computer labs and faculty
and graduate students) is equivalent to the best-case scenario of computer energy use.
For the worst-case scenario, the null hypothesis is
Ho: current energy consumption by computers on campus (libraries, computer labs and faculty
and graduate students) is equivalent to the worst-case scenario of computer energy use.
3.2.2 S-Plus Analysis
S-Plus, a statistical software program was used to produce 3 statistical summaries for the raw
data from the monitoring sessions (S-Plus, 2002). Each summary produced values for the mean
computer usage for each of the monitoring times as well as, the standard deviation, the minimum
and maximum percentage of computers on, the first, second and third quartiles and the median
value of the raw data. The first summary included computer usage for Monday through Sunday.
The second summary included computer usage for Monday through Friday and was considered to
37
be a 5 day work week. Finally, a summary of weekend use, Saturday and Sunday, was produced.
All three summaries were compared to examine different use patterns of computer systems in the
Richards Building.
3.3 Environmental Impacts and Green Procurement Strategies
Throughout the course of this study, literature reviews were conducted in order to gain insight
into computer life cycles and various related topics associated with their environmental impacts
and potential green procurement. In order to apply this information to the University of Guelph,
the on-campus Sustainability Coordinator was contacted, and, upon request provided information
regarding the University of Guelph’s current disposal strategies. As the University currently
donates old systems to a local outreach program, the coordinator of that initiative was also
contacted.
Once current disposal techniques were established, alternative techniques were researched,
compiled and ranked based on environmental effectiveness and economic feasibility. The current
and possible computer acquisition techniques were also researched via literature reviews and
communication with various IT personnel. In order to understand the potential implications of
the implementation of green procurement at the University of Guelph, other Canadian institutions
that had recently undergone such changes were studied on a case-by-case basis. All information
was analyzed and compiled for the provision of future recommendations.
38
4.0 Results
4.1 Statistical Results
4.1.1 z-Test Statistical Results
Best-case scenario
The following are results for testing if computer energy consumption is equivalent to the bestcase scenario for laboratories, libraries and by faculty and graduate students. The null hypotheses
presented in this section are used for statistical purposes only and do not effect the overall null
hypothesis. For each of the tests below, the null hypothesis is rejected if z > 1.96.
For Laboratories:
Ho: µ = 442.00 kWhr/yr
H: µ > 442.00 kWhr/yr
xavg. = 1137.65 kWhr/yr
σ2 = 1.28 x 10-8
z = (xavg. – 442.00 kWhr/yr / (σ2 / n)1/2
= (1137.65 – 442.00) / (1.28 x 10-8 / 689) 1/2
= 1.61 x 108 kWhr/yr
z > 1.96, therefore we can reject the null hypothesis.
For Libraries:
Ho: µ = 493.00 kWhr/yr
H: µ > 493.00 kWhr/yr
xavg. = 1741.34 kWhr/yr
σ2 = 3.68 x 10-8
z = (xavg. – 493.00 kWhr/yr / (σ2 / n)1/2
= (1741.34 – 493.00) / (3.68 x 10-8 / 289) 1/2
= 1.11x 108 kWhr/yr
z > 1.96, therefore we can reject the null hypothesis.
39
For Faculty and Graduate Students:
Ho: µ = 239.00 kWhr/yr
H: µ > 239.00 kWhr/yr
xavg. = 933.66 kWhr/yr
σ2 = 1.35 x 10-8
z = (xavg. – 239.00 kWhr/yr) / (σ2 / n)1/2
= (933.66 – 239.00) / (1.35 x 10-8 / 289) 1/2
= 1.73 x 109 kWhr/yr
z > 1.96, therefore we can reject the null hypothesis.
Worst-case scenario
The following are results for testing if computer energy consumption is equivalent to the worstcase scenario for laboratories, libraries and by faculty and graduate students. The null hypotheses
presented in this section are used for statistical purposes only and do not effect the overall null
hypothesis. For each of the tests below, the null hypothesis is rejected if z > 1.96.
For Laboratories:
Ho: µ = 2359.00 kWhr/yr
H: µ > 2359.00 kWhr/yr
xavg. = 1137.65 kWhr/yr
σ2 = 1.28 x 10-8
z = (xavg. – 2359.00 kWhr/yr) / (σ2 / n)1/2
= (1137.65 – 2359.00) / (1.28 x 10-8 / 689) 1/2
= - 2.83 x 109 kWhr/yr
z < 1.96, therefore we fail to reject the null hypothesis.
For Libraries:
Ho: µ = 2359.00 kWhr/yr
H: µ > 2359.00 kWhr/yr
40
xavg. = 1137.65 kWhr/yr
σ2 = 3.68 x 10-8
z = (xavg. – 2359.00 kWhr/yr) / (σ2 / n)1/2
= (1741.34 – 2359.00) / (3.68 x 10-8 / 289) 1/2
= -5.47 x 107 kWhr/yr
z < 1.96, therefore we fail to reject the null hypothesis.
For Faculty and Graduate Students:
Ho: µ = 2359.00 kWhr/yr
H: µ > 2359.00 kWhr/yr
xavg. = 933.66 kWhr/yr
σ2 = 1.35 x 10-8
z = (xavg. – 2359.00 kWhr/yr) / (σ2 / n)1/2
= (933.66 – 2359.00) / (1.35 x 10-8 / 289) 1/2
= -6.54 x 108 kWhr/yr
z < 1.96, therefore we fail to reject the null hypothesis.
4.1.2 S-Plus Statistical Results
Three moments, 10:00 am, 2:00 pm and 7:00 pm, were monitored to determine the number of
computers that were on in the Richards Building. 10:00 am is considered to be the morning.
2:00 pm is considered to be the afternoon. 7:00 pm is considered to be the evening and both
10:00 am and 2:00 pm combined are considered to be the afternoon. The number of computers
turned on at a given time were computed as a percent value and then compared by using
statistical software.
41
From the use of the S-Plus statistical software program the following information was determined
(Appendix 6). The average percentage of computers turned on at 10:00 am, 2:00 pm and 7:00 pm
over a 7 day week period are 48.6%, 52.3% and 33.5% respectively. Assuming that these
moments in time are the only time that a computer is on for, the results show that more
computers are turned on throughout the day than in the evening. Also, a higher percentage of
computers are turned on in the afternoon compared with the percent of computers turned on in
the morning.
The difference between the maximum number of computers turned on at one time and the
minimum number of computers turned on at one time result in a value called the range. The
range for the percentage of computers turned on for the 10:00 am, 2:00 pm and
7:00 pm for the 7 day week is 31.8%, 33.3% and 31.7% respectively. The 2:00 pm monitoring
period showed the greatest deviation from the mean (14.5%) while the 7:00 pm monitoring time
had the least deviation from the mean (11.1%). The average percentage of computers turned on
at 10:00 am, 2:00 pm and 7:00 pm over a 5 day work week are 55.9%, 60.7% and 35.3%
respectively. These results are similar to the 7 day week results above, in that more computers
are turned on throughout the day than at night. Also, a higher percentage of computers are turned
on in the afternoon compared with the percent of computers turned on in the morning.
The range for the percentage of computers turned on at 10:00 am, 2:00 pm and 7:00 pm for the 5
day work week is 8.5%, 5.0% and 31.6% respectively. The 7:00 pm monitoring period showed
the greatest deviation from the mean (13.0%) while the 2:00 pm monitoring time had the least
deviation from the mean (2.2%).
42
The average percentage of computers turned on at 10:00 am, 2:00 pm and 7:00 pm over a 2 day
weekend are 30.4%, 31.3% and 28.8% respectively. These results are similar again to both the 7
day week and the 5 day work week in that more computers are turned on throughout the day than
at night. Again, it was also found that a higher percentage of computers are turned on in the
afternoon compared with the percent of computers turned on in the morning.
The range for the percentage of computers turned on for 10:00 am, 2:00 pm and 7:00 pm for the 2
day weekend is 2.5%, 2.5% and 0.8% respectively. The 7:00 pm monitoring time had the least
deviation from the mean (0.6%) while both the 10:00 am and 2:00 pm monitoring times showed
the same deviation value from their means.
The percentage of computers that were observed to be on during all three monitoring times over
the entire week long monitoring period was determined manually, without the use of statistical
software. It was found that 19.2 % of all computers were on throughout the entire monitoring
period. Excluding the 19.2% of computers that were on for each of the monitoring times, 47.5%,
57.5%, 46.7%, 48.3%, 45.8%, 13.3% and 20.0% of computers were turned on at least once for
Monday, Tuesday, Wednesday, Thursday, Friday, Saturday and Sunday, respectively.
4. 2 Energy Consumption
As shown in Table 4, the University of Guelph’s current energy consumption lies between worst
and best-case scenarios, being closer to the best case. It was also found that conservation plans to
reduce energy consumption contributed the most to energy savings as compared to purchasing
43
new computer equipment thus rejecting the first sub null-hypothesis and accepting second sub
null-hypothesis (Figure 5).
4.2.1 Faculty and Graduate Students
From Netscan, it was determined that the average hourly usage for faculty and graduate students
was 9.5 hours per day, for a 5 day work week, without stand-by/energy saving mechanisms
(Appendix 7). It was assumed that 50% of all monitors were LCD as this was the general on
campus trend (Appendix 7). Faculty and graduate students were found to consume the highest
amount of energy on campus (Figure 3, Table 4). However, on a per computer basis, faculty and
graduate students were found to use the least amount of energy (Figure 4, Table 6).
4.2.2 Computer Laboratories
Computer laboratories with more than 20 computers were located in: Crop Science, Graham Hall,
Hutt, Mackinnon, MacNaughton, OVC microcomputer laboratory and Powell (Table 5). The
laboratories hours ranged from 24 hours, 7 days a week, to 9 hours per day, 5 days a week
(Appendix 7). Each laboratory differed in the number of LCD monitors versus CRT monitors.
Some laboratories contained 100% CRTs while other laboratories only contained 10% CRTs.
However overall, approximately 50% of monitors on campus were found to be CRT monitors
(Appendix 7). Energy consumed by computer laboratories as a whole was only exceeded by the
energy consumed faculty and graduate students (Figure 3, Table 4). This trend was noticed on a
per computer basis as well (Figure 4, Table 6).
4.2.3 Libraries
The MacLaughlin and OVC computers were found to be on for 24 and 12 hours per day,
respectively (Appendix 7). MacLaughlin library was found to contain 45% CRTs and 55% LCD
44
monitors, while the OVC library contained 100% CRTs (Appendix 7). The libraries were found
to consume the least energy of the three groups (Figure 3, Table 4). However, on a per computer
basis the libraries were found to consume the most energy (Figure 4, Table 6).
Table 4: University of Guelph's Current Energy Consumption in Comparison with the Worst-Case and BestCase Scenarios
Current
Best-Case
Worst-Case
Total
Total
Total
energy
energy
energy
Cost ($)
Cost ($)
Cost ($)
use/year
use/year
use/year
(kWhr/yr)
(kWhr/yr)
(kWhr/yr)
Laboratories
Faculty and Graduate Students
Libraries
783,844
2,645,992
503,246
62,708
211,679
40,260
304,290
677,893
142,598
24,343
54,231
11,408
1,625,158
6,684,612
681,670
130,013
534,769
54,534
Table 5: Energy Consumption in University of Guelph Buildings
Current
Location
Crop Science
Graham Hall
Hutt
MacNaughton
McKinnon
OVC Microcomputer lab
Powell
Thornbrough and Renyolds
MacLaughlin Library
OVC Library
Faculty and Graduate Students
Total
Total
energy
use/year
(kWhr/yr)
50,450
36,036
160,393
87,567
68,234
65,520
22,113
293,530
470,486
32,760
2,645,992
3,933,082
Cost ($)
4,036
2,883
12,831
7,005
5,459
5,242
1,769
23,482
37,639
2,621
211,679
314,647
Best-Case
Total
energy
use/year
(kWhr/yr)
8,907
6,362
17,304
16,795
18,322
20,452
5,938
210,210
132,372
10,226
677,893
1,124,782
Cost ($)
713
509
1,384
1,344
1,466
1,636
475
16,817
10,590
818
54,231
89,983
Worst-Case
Total
energy
use/year
(kWhr/yr)
82,555
58,968
160,393
155,676
141,523
117,936
82,555
825,552
622,702
58,968
6,684,612
8,991,441
Cost ($)
6,604
4,717
12,831
12,454
11,322
9,435
6,604
66,044
49,816
4,717
534,769
719,315
45
Figure 3: Current Energy Consumption
300
Power Usage
Cost
250
$212
250
200
200
150
150
100
100
78
$63
50
50
$40
50
Cost (Cost/yr/Computer x 102)
Power Usage (kW hr/yr/computer x 103)
265
0
0
Faculty and Graduate
Labs with 20 computers or
Students
more
MacLaughlin Library
Table 6: Energy Consumption per Computer in University of Guelph Buildings
Location
# of
Computers
Crop Science
Graham Hall
Hutt
MacNaughton
McKinnon
OVC Microlab
Powell
Thornbrough and Renyolds
MacLaughlin Library
OVC Library
Faculty and Graduate Students
Total
35
25
68
66
60
50
35
350
264
25
2834
3812
Current
Total
energy
Cost ($)
use/year
(kW/yr)
1,441
1,441
2,359
1,327
1,137
1,310
632
839
1,782
1,310
934
14,513
115
115
189
106
91
105
51
67
143
105
75
1,161
Best-Case
Total
energy
Cost ($)
use/year
(kWhr/yr)
254
254
254
254
305
409
170
601
501
409
239
3,652
20
20
20
20
24
33
14
48
40
33
19
292
Worst-Case
Total
energy
Cost ($)
use/year
(kWhr/yr)
2,359
2,359
2,359
2,359
2,359
2,359
2,359
2,359
2,359
2,359
2,359
25,946
189
189
189
189
189
189
189
189
189
189
189
2,076
46
Figure 4: Current Energy Consumption per Computer
2000
$139
1741
160
Cost
140
1600
120
1400
100
$91
1138
1200
$75
934
1000
800
80
60
600
40
Cost (Cost/yr/Computer)
Power Usage (kW hr/yr/computer)
1800
Power Usage
400
20
200
0
0
MacLaughlin and OVC
Labs with 20 computers or
Faculty and Graduate
Libraries
more
Students
Figure 5: Energy Savings for Conservation plans and New Computer Equipment
$1,400
$1,321
$1,150
$1,200
24hrs/7days
8hrs/5days
$1,000
$800
$734
$563
$600
$400
$225
$200
$196
$125
$96
$0
Old CPU and CRT
New CPU and CRT
Old CPU and Flat
New CPU and Flat
screen
screen
47
5.0 Discussion
5.1 Statistical Analysis
5.1.1 Discussion for z -Test
No situation can ever be better than the best-case scenario and no data can ever be worse than the
worst-case scenario. The z-tests preformed in this paper were used to determine if computer
energy use on campus is actually running at a best or worst-case scenario.
Each of the best-case scenario z-tests resulted in z-values larger than their critical value. This
results in the rejection of the null hypothesis for all of the best-case scenario tests. Since the
statistical null hypothesis is rejected there is sufficient evidence to conclude that computer energy
use by computers in the laboratories, libraries and by all faculty and graduate students does not
run at a best-case scenario.
The z-tests performed to determine if computers on campus are following a worst-case scenario
resulted in a failure to reject the statistical null hypotheses. This resulted in sufficient evidence to
conclude that computers on campus do not run at the worst-case scenario.
These results show that the University of Guelph’s current energy use by computers does not
follow a best-case scenario. However, the results do show that energy consumption by computes
on campus does not represent a worst-case scenario.
48
5.1.2 Discussion for S-Plus
Using the statistical summaries of the S-Plus output for the analysis of Monday through Sunday
(7 day week), Monday through Friday (5 day work week) and Saturday and Sunday (weekend), it
was found that the percentage of computers on during the day exceeded the percentage of
computers on at night. For example, during the 7 day week, 16% more computers were turned on
in the morning and 19% more computers were turned on in the afternoon than in the evening.
However, the percentage of computers on in the morning was always less than the percentage of
computers on in the afternoon. The average number of computers on for each time period is
lower during the weekend period and higher during the work week period. The lower
percentages of computers on in the evenings and over the entire weekend is most likely attributed
to the fact that the Richards Building’s hours of operation are Monday through Friday from 8:30
am to 4:30 pm.
Only three different times over a 24 hr period were monitored in this experiment to determine
when a computer was on. In the raw data, it was found that 19.2% of computers were turned on
at each of the three monitoring times for all seven days. This is a possible indicator that these
computers were turned on for 24 hours through each of the days of the week. However, this
cannot be tested with confidence unless the computers were monitored every second of the day
for the 1 week monitoring period.
Without the use of statistical software it was determined that computers are used at least once
everyday of the week in the Richards Building. This shows that computers are important in an
academic setting. If an experiment were conducted to monitor computer use over a 24 hour
49
period for a number of consecutive years, it would be possible to determine if a significant
dependence for computer use increased with time. From the above analysis, results show that all
computers in the Richards Building do not run for 24 hours a day, 7 days a week, therefore
computers are not running at a worst-case scenario.
5.2 Energy Consumption
The University of Guelph is currently operating between the best and worst-case scenarios, with
an overall tendency towards the best-case scenario. The University has approximately 50% LCD
monitors already in use, and is gradually increasing this percentage to keep pace with
technological demands. Furthermore, there are already some conservation plans in use, since
most computers are not on for 24 hours a day, 7 days a week.
5.2.1 Faculty and Graduate Students
Faculty and graduate students were found to use the most computers on campus, and therefore
consumed the most energy. However, the energy consumed per computer by faculty and
graduate students was lower than energy consumed by computers in the laboratories and libraries.
This is could be attributed to the fact that laboratories and libraries leave their computers active
for longer established hours than faculty and graduate student. Also, the libraries and
laboratories facilitate the frequent use of computers by many users; therefore these computers
may not enter a low power mode as often.
5.2.2 Computer Laboratories
Laboratories were found to be mid-range in terms of overall and per computer energy
consumption, when compared with libraries and faculty/graduate students (Figure 4). This is
50
most likely due to the fact that each laboratory has different hours of operation. Another reason
that laboratories fell in the mid-range of energy consumption is because these computers are less
active than the library computers but more active than those of faculty and graduate students.
5.2.3 Libraries
The libraries consume less energy compared to the laboratories and faculty and graduate students.
However, they consume the most energy per computer because, although the libraries have a
large majority of LCD monitors (50% LCD, 50% CRT), they are on for one of the longest time
intervals. Although the OVC library currently applies conservation strategies, it only comprises
of 9% of total library computers, and does not compensate for the MacLaughlin library
computers being on 24 hours a day 7 days a week.
Although the cost savings for the best-case scenario were large there was no consideration for
initial cost of procuring new LCD monitors and CPUs. This lack of consideration will likely
lower the cost savings.
5.3 Computer Equipment Purchasing Guidelines
In addition to investigating computer usage at the University of Guelph, the procurement and
disposal practices were also examined. This was carried out in order to assess the University’s
environmental impact, and to determine if green procurement and environmentally sound end-oflife management practices have been put into place.
51
In terms of computer acquisition strategies, there are currently no green procurement procedures
being used by the University in guiding computer purchasing decisions. Current policies are
based upon pricing structure, and are derived from contracts with the set of preferred vendors
listed below (Loo, 2005):
•
Apple Canada
•
Audcomp Computer Systems
•
Computer Hardware Services Inc.
•
Dell Canada
•
First Avenue Information Systems Inc.
•
Kerr Norton
•
Onward Computer Systems
This list is comprised of both manufacturers and re-sellers. Manufacturers can often be evaluated
in terms of their environmental performance, as certain companies document their dedication to
environmental stewardship (Dell 2005, Apple 2005). Resellers often sell products from a wide
variety of manufacturers, therefore, it is often difficult to determine if products meet certain
environmental criteria.
One main problem that affected this study was the lack of centralized data concerning computer
procurement at the University. A budget for computer information technology is allotted in the
form of a lump sum, which is to be managed over a period of several years. This money is
allocated by Purchasing Services to specific departments as funding is required. There are
currently no records kept by Purchasing Services about computers purchased, their
manufacturers, and quantities of computers acquired. Without such records, and the potentially
52
varying environmental policies between vendors, the task of comparing current computer
procurement strategies with desired green procurement, could become problematic.
As purchasing decisions are made by individual departments, it is important to investigate how
departments acquire their computers, as there are currently no green procurement guidelines in
place to direct departments in their purchasing. Most purchasing decisions were based primarily
upon pricing structure, with any green procurement strategies taken only as a secondary
consideration (Loo, 2005).
5.4 Computer Equipment Disposal Guidelines
In Canada, only 50% of high-tech electronic equipment is reused or recycled (Globe and Mail,
2001). The remainder is either disposed of in landfills or incinerated, thus releasing toxic
substances into the environment. Currently in Canada, there is no legislation that requires
producers to take-back and properly recycle used, un-wanted equipment (Globe and Mail, 2001).
Unfortunately, this trend is strongly reflected by the University of Guelph’s computer recycling
system.
While there is no official computer reuse strategy being implemented on campus, computers that
are no longer wanted by one department can be received by another department. It is only when
computers are unable to run the necessary software required by any of the departments, that they
are slated for end of life management.
53
The University of Guelph’s current end of life management strategy for used computer systems
was initiated in 2002; however, it is still small and practically unknown. The procedure is
facilitated by the Recycling and Waste Co-ordinator who requires small pick-ups (less than 12
units) to be brought to MacKinnon loading docks (MacKinnon 009), while larger pick-ups
(greater than 12 units) can be gathered directly from their source (Maurice, 2005). CRT monitors
are not accepted for pick-up, and for larger pick-ups, exact quantities of the various materials
available must be expressed prior to their removal. A complete out-line of the current system at
the University of Guelph is available in Appendix 4.
The University currently donates all used PCs to a company in Fergus Ontario named Production
Works Co-op, which operates out of the community living centre of Guelph-Wellington
(Maurice, 2005). At this site, approximately 30 developmentally challenged community
members are employed to dismantle component parts, which are then sold to two local recycling
companies: Hi Tech Recycling (Canada) Ltd and Joseph & Company (Dano, 2005). The
dismantling of computers and other electronics (no CRT monitors are accepted) is carried out in a
manner that is safe for the workers as well as the environment. Through Production Works Coop, people with intellectual disabilities are also provided with valuable training and experiences
in the workplace and business community. Production Works Co-op also ensures that both
companies which purchase its materials carry out recycling processes using the most
environmentally sound methods possible (Dano, 2005). The members of Production Works Coop are also the workers, owners and decision makers of the company. The Co-op charges the
University no fees except for the cost of transportation at a rate of CDN $ 0.38 per kilometre
(Dano, 2005). Unfortunately, while most materials are sent elsewhere for further recycling,
54
Production Works Co-op must deal with wastes in the form of plastics; which are subsequently
disposed of in landfills (Dano, 2005).
Both Hi tech Recycling and Joseph & Company are local; based in Toronto and Kitchener,
respectively. At Hi Tech Recycling, 300 to 500 computers are recycled everyday, and their
metals, such as aluminum, copper and silver are recovered and valued between CDN $ 1.59- 2.50
per machine (Globe and Mail, 2001).
The main problems associated with the disposal of used computer systems at the University of
Guelph, stem from insufficient knowledge and communication. An overall campus-wide lack of
knowledge regarding proper disposal methods and their locations, directly relate to a lack of
communication within and between departments regarding these topics. There are a few areas in
which the University of Guelph’s current computer disposal system can be improved. Firstly,
because there is no streamlined computer disposal policy, there is no organizational structure to
facilitate proper disposal through this program. Secondly, there is no way to evaluate the success
of this program, as there are no records regarding quantity of donations to Production Works Coop. The only estimate obtained regarding current volumes equate to one van load every two
months (Dano, 2005). The increasing environmental problems associated with the disposal of
electronic equipment, indicates that there is a growing need to maximize waste diversion to
recovery services such as Production Works Co-op.
55
6.0 Recommendations
6.1 Purchasing Computer Equipment
The findings of this report have provided important insight into current computer procurement
practices at the University of Guelph as well as, strategies that have been developed at other
institutions such as the University of Manitoba. From this, a number of recommendations were
formulated that could be implemented by the University of Guelph. These recommendations
would work towards achieving benefits in terms of cost savings in the form of improved energy
efficiency, as well as the minimization of environmental impacts.
Green Procurement Guidelines
The development of a set of green procurement guidelines for computers at the University of
Guelph is needed. Examples of these guidelines are readily available from a variety of
organizations and can be adapted to fit within the context of the University’s purchasing policy.
The University of Guelph should set purchasing guidelines that include:
ƒ
Giving preference to products with ecolabels that are: energy efficient, upgradeable, and
contain recycled materials
ƒ
Buying products from suppliers which offer a take-back or trade-in program
ƒ
Purchasing products in bulk in order to minimize wastes from packaging which use
minimal packaging
ƒ
Encourage efficient reuse, refurbishment or recycling by the purchasing equipment that is
easily dismantled
56
Implementation of these guidelines through the Purchasing Services department would allow for
a uniform adoption in all areas of the University, whether it is administration, computer
laboratories, faculty, graduate students or staff. Implementation at a departmental level would
also be valuable, as it would raise the overall awareness of procurement guidelines in the
University community.
The evaluation of the current set of preferred vendors for conformity to the green procurement
guidelines is needed. In the case of nonconformity, the economic feasibility of phasing out these
vendors in favour of vendors agreeing to operate within green procurement guidelines must be
determined.
Lastly, the University should establish records of computer purchases, including the number and
types of purchased computers, such that there can be documentation of the use of green
procurement strategies, and an evaluation of its success can be made.
As the University of Guelph faces current budget challenges, the need to conserve and reduce
expenditures is apparent. The implementation of the suggested green procurement strategies may
act as significant aids in reducing these challenges, while further contributing to the University’s
excellent reputation as an environmentally conscientious institution.
6.2 Energy Saving Strategies
The results of this study have indicated that conservation plans were found to contribute the most
to energy savings compared to replacing current computer equipment with new computer
57
equipment. The following recommendations will outline how the University of Guelph could
increase their energy savings in computer laboratories, libraries and by faculty and graduate
students.
6.2.1 Computer Laboratories and Libraries
Monitors
Computer systems consist of two main components that draw power, the monitor and the actual
computer system itself. To reduce the amount of power consumed by monitors, we recommend
that power management options, which are available for the majority of monitors, be activated.
The power management setting of the monitor should be set so that if the system is inactive for
ten minutes or longer, the monitor will go into the low powered sleep mode. Activating monitor
power management can reduce power consumption by 60 to 90 W, a savings of up to $55 per
computer each year (US EPA, 2005). The activation of power management for a computer
monitor causes practically no problems for the system or network (US EPA, 2005). The screen
saver options for computer monitors should not be activated for two reasons. Firstly, screen
savers do not reduce the amount of power consumed by the computer system (Hewlett-Packard,
2005). Secondly, if the computer has a low power setting option which has been activated,
complex screen saver graphics may require enough processing power to bring the computer out
of sleep mode (Nordman, 1997).
For extended periods of non-use, such as over night or weekends, computer monitors which will
be not be in use for these periods should be turned off completely. The reason for this is that
even in a sleep mode the monitor continues to draw power, and when aggregated over an entire
year this could constitute a significant amount of energy use and costs (Lebot, 2000).
58
CPUs
As of 1996, approximately 70% of all computer sold have power management capabilities
(Nordman, 1997). However, presently this figure is most likely higher. The implementation of
power management for CPUs can reduce overall power consumption by an additional 40 to 90
W, a savings of about $45 per CPU each year (US EPA, 2005). It is recommended that these
power management options be activated, since they are feasible. However, unlike computer
monitors, the CPU can have some difficulties with low power settings. According to Norman
(1997) CPUs on a network will receive routine messages from the server, which requires the
client CPU to respond. If the CPU is in a low power setting and is unable to respond to the
server, it could cause network difficulties. Newer CPUs can handle these operations by having a
smart network card, which can send a message back to the server without bringing the whole
system out of the low powered state. Another characteristic of newer CPUs is that they can
partly awaken, handle the operation, and then resume a low power setting (Nordman, 1997).
These CPU power management options work best if the CPU is a Pentium IV running Windows
2000 or XP and the administrative updates are retrieved from the network by client machines (US
EPA, 2005).
6.2.2 Faculty and Graduate Students
Many of the recommendations made for computers in the laboratories and libraries can be
applied to faculty and graduate students, as well as, other staff at the University of Guelph.
Additionally we feel that the computers used by faculty and graduate students can be subjected to
more detailed power management strategies since these computers have fewer users. We
recommend that the CPUs and monitors be turned off during extended periods of non-use over an
hour (as long as the computer is not needed to perform a task). This will increase energy savings
59
because, as has been previously stated, even in low power settings CPU and monitors will
continue to draw power (Kawamoto, 2002).
6.3 Computer Equipment Disposal Methods
One basic recommendation for the remedy of the problem of poor interdepartmental
communication is the streamlining of all departments, laboratories and offices with respect to
proper computer disposal methods. If all faculty and staff are organized to dispose of their
computers in the same fashion, all used on-campus computers could be effectively dealt with in
the same environmentally conscientious manner. This could easily occur through increased
promotion of the University’s current system via e-mail, written word and awareness campaigns.
Keeping detailed records of working computers, as well as those which have been removed for
disposal, and their current locations, would allow IT and office managers to further track their
machines and divert them into the on-campus disposal system, rather than allow them to simply
“disappear” in storage.
When the Recycling and Waste Management Coordinator first began the current computer
disposal system, it quickly became apparent the large quantity of on-campus computers that were
simply being stored rather than properly disposed of. Computer systems from the 1980s were
being and, continue to be, brought in for disposal, thus implying the need for an efficient
streamlined system (Maurice, 2005). The coordinator of Production Works Co-op has also
experienced the inconsistencies of the donation quantities from the University. The company has
suggested the need for a standard on-campus system as well, thus allowing for regular, rather
than sporadic pick-ups to be made (Dano, 2005).
60
The streamlining and expansion of the University of Guelph’s current on-campus computer
disposal system would not only result in the increase of profit and productivity for the workers at
Production Works Co-op, but ensure that all of the University’s discarded computers would be
dealt with in the same ecological manner. Unfortunately, this scenario still does not provide an
adequate means of monitor disposal. There are, however, certain new options that can be
explored in order to divert these toxic components from the landfill.
It is recommended that on-campus computers, which are still fully functional (both CPU and
monitor) be donated to such programs as Computers for Schools, so that they may continue to
provide service for libraries, school and non-profit organizations. Direct contact could even be
established between the University and local schools/organizations so that direct donations could
be made. Perhaps a website or community notice board could even be implemented to allow
faculty and staff to post the availability of functional computer systems for donation.
For computer systems, which are no longer functional, Logic Box Distribution Inc. appears to be
the best choice for recycling and refurbishment. The company accepts unlimited quantities of
electronic waste from any source, free of charge. In addition to monitors, Logic Box accepts
desktops, laptops, printers, keyboards, mice and other computer peripherals, as well as other
electronic equipment (BAN, 2005). Logic Box is also one of the three Canadian recycling
companies which adheres to the “Electronic Recycler’s Pledge of True Stewardship”, thus
ensuring it employs environmental and socially just practices (BAN, 2005). Currently, the
company stocks from 15 000 to 20 000 different products, and the goal is to increase this number
61
to 100 000 in order to satisfy consumer demand (Chezzi, 2002). Logic Box recently launched a
project in cooperation with several southern Ontario municipalities in order to efficiently and
effectively collect residential electronic waste (Chezzi, 2002). Unlike Production Works Co-op,
Logic Box does not provide any pick up services; however, used equipment can be delivered in
person or by train (BAN, 2005).
There are many aspects of computer recycling that are still elusive and currently somewhat
unattainable. The complete recycling of CRTs and plastic casings pose various environmental
problems due to high toxicities and difficulty in the separation of their chemical components
(Brennan et al., 2002). While some firms are researching the possibilities of using computer
plastics to produce fuel or sulphuric acid, these are simply preliminary theories, which require
extensive research.
The best option for effective computer disposal at the University of Guelph would make use of
various firms to ensure that all computer components are diverted from landfills in the greatest
amounts possible for the longest periods of time. Streamlining the disposal of unwanted, oncampus computers so that all are gathered in a central, well-known location requires promotion,
education, communication and record keeping by those who are involved. Computers, which are
still functional, should be donated to local schools and organizations as to prolong their lifespan.
Systems which are no longer functional should be donated (without CRT monitors) to the local
initiative Production Works Co-op in Fergus Ontario to support the workers, ensure that materials
are being handled in an environmentally conscientious manner, and cut down on fossil fuel
emissions during transportation. Monitors, and any other peripherals not accepted by Production
62
Works Co-op should be sent to Logic Box Distribution Inc. in Mississauga Ontario for either
refurbishment and re-selling, or recycling.
The catalyst for occurrence of this chain of events must manifest itself in the form of widespread
on-campus education and awareness programs regarding the various hazards associated with
landfill disposal, as well as the environmentally friendly methods available for disposal. The
proposed system can only be a success if a campus-wide effort is made to coordinate the disposal
of computers across all departments and offices. This requires communication, and above all, the
realization that positive actions will induce positive change, both on campus and within the
computer industry (Kuehr and Williams, 2003).
63
7.0 Conclusion
This study examined the University of Guelph’s current performance with regards to computer
acquisition, energy consumption, and disposal. In order to achieve this, the null hypothesis,
which states that the University of Guelph’s current practices will be the same as the best-case
scenario for energy consumption and cradle-to-cradle environmental efficiency, was tested.
With regards to the computer energy consumption at the University of Guelph, statistical
evidence was found to reject the null hypothesis. The results from our study were proved to be
statistically significant by our z-test, indicating that the University is not operating at the bestcase scenario, and in fact falls mid range between best and worst-case.
Testing the null hypothesis with regards to the green procurement and disposal of computers
proved to be more challenging because it was qualitative in nature. However, from the
information collected, there was sufficient evidence to reject the null hypothesis. Currently, there
are no green procurement strategies or cradle-to-cradle considerations in place at the University
of Guelph. Departmental computer acquisition is primarily based on price, with little
consideration for environmental standards. An initiative is in place to donate unwanted systems,
however, there is no streamlining of this effort within, or between, departments. The lack of a
sufficient means of currently disposing of CRT monitors is further grounds to reject the null
hypothesis.
While the University of Guelph is currently not operating within a best-case scenario, there is
room for optimism. This study placed the University's computer energy consumption at mid64
range between best and worst case; an encouraging result which indicates that the University is
on the path to improvement. By adhering to the listed recommendations, direct savings in the
form of reduced energy costs can be achieved. Suggested green procurement policies would give
incentives to industry to improve their environmental standards, as well as extend the lifespan of
functional, yet unwanted systems. Streamlined management of the University’s computer
wastes, especially with regards to CRT monitors, would help to alleviate the flow of hazardous
materials entering landfills, and reduce the risk of contamination by lead and other heavy metals.
65
References
Apple and the Global Environment. 2005. [Online] http://www.apple.com/environment/
Accessed 31 Mar 2005.
Anonymous. 2003. PC Computer Notes and hardware Tutorials. [Online]
http://www.pccomputernotes.com/pcterms/glossaryc.html. Accessed March 18, 2005.
Basel Action Network (BAN). 2005. Electronic Recycler’s Pledge of True Stewardship. [Online]
www.ban.org/E-wastes/Pledge/Locations.html#logic-mat. Accessed March 15, 2005.
Baul, H. 2002. Error: dumping does not compute. Alternatives Journal. 28:2.
Bechara, E. J. H. 2004. Lead poisoning and oxidative stress. Free radical Biology and Medicine.
36: S22 Suppl. 1.
Brennan, L.B., Isaac, D.H., and Arnold, J.C. 2002. Recycling of Acrylonitrite-Butadene-Styrene
(ABS) and High-Impact Polystyrene (HIPS) from waste computer equipment. Journal of
Applied Polymer Science. 86: 572-578.
BSD Global. 2002. Business and Sustainable Development: A Global Guide - Green
Procurement. Developed by the International Institute for Sustainable Development.
[Online] <www.bsdglobal.com/tools/bt_green_pro.asp> . Accessed January 10, 2005.
Chezzi, D. 2002. A need for e-cycling-Where do old computers go when they die? [Online]
http://www.rebootcanada.ca/sections/macleans.html. Accessed March 17, 2005.
Computers for Schools (CFS). 2005. Computers for schools. [Online]
http://www.schoolnet.ca/cfs-ope. Accessed March 29, 2005.
Curran, M. 1996. Environmental life-cycle assessment. New York: McGraw Hill. pp.432
Dano, S. Personal communication via e-mail. March 16, 2005.
Darnerud, P., Eriksen, G.; Johannesson, T.; Larsen, P. and M. Viluksela. 2001. Polybrominated
diphenyl ethers: Occurrence, dietary exposure, and toxicology. Environmental Health
Perspectives. 109 (supplement 1): 49-68
Dell. 2005. Dell Recycling: national Cristina Foundation. [Online]
http://www.cristina.org/dsf/dell.nfc. Accessed January 13, 2005.
Dell and the Environment. 2005. [Online]
http://www1.ca.dell.com/content/topics/global.aspx/corp/environment/en/index?c=ca&cs=
CADHS1&l=en&s=dhs. Accessed 31 Mar 2005.
66
Domingo, J.L. 2004. Human exposure to polybrominated diphenyl ethers through the diet.
Journal of Chromatography. 1054: 321- 326
Fava, J.; Consoli F.; Denison R.; Dickson K.; Mohin T.; Vigon B. (Editors). 1993. A conceptual
framework for life-cycle impact assessment. Pensacola FL: Society of Environmental
Toxicology and Chemistry (SETAC).
Geiser, K. 2001. Materials Profile of the Computer and Semiconductor Industries. Lowell
Center for Sustainable Production. Lowell: University of Massachusetts.
Globe and Mail. 2001. Discarded computers hazardous in landfill. [Online]
http://www.hightechrecycling.com. Accessed March 28, 2005.
Gungor, A. and Gupta, S.M. 1999. Issues in environmentally conscious manufacturing and
product recovery: a survey. Computers and Industrial Engineering. 36: 811-853.
Handy and Harman Electronic Materials Corporation. 1999. Composition of a Personal Desktop
Computer. [Online] http://www.svtc.org/hightech_prod/desktop.htm. Accessed March, 28,
2005.
Osako, M., Kim, Y-J., and Sakai, S. 2004. Leaching of brominated flame retardants in leachate
from landfills in Japan. Chemosphere. 57: 1571-1579.
Kawamoto, K.; Koomey, J.G.; Nordman, B.; Brown, R.E.; Piette, M.A. 2002. Electricity used
by office equipment and network equipment in the US. Energy. 27(3): 255–269.
Keafe, C. 2002. Where haves meet wants [Online] http://www.dealerscope.com Accessed March
28, 2005.
Klatt, S. 2003. Recycling Personal Computers. In Kuehr, R. and Williams, E (Eds.). Computers
and the Environment: Understanding and Managing Their Impacts. London: Kluwer
Academic Publishers.
Kuehr, R. 2003. Managing PCs through Policy: Review and Ways to Extend Lifespan. In Kuehr,
R. and Williams, E (Eds.). Computers and the Environment: Understanding and Managing
Their Impacts. London: Kluwer Academic Publishers.
Lee, C-H.; Chang, C-T.; Fan, K-S.; and Chang, T-C. 2004. An overview of recycling and
treatment of scrap computers. Journal of Hazardous Materials. B114: 93-100.
Lebot, B.; Meier, A.; and Anglade, A. 2000. “Global Implications of Standby
Power Use” In
The Proceedings of ACEEE Summer Study on Energy Efficiency in Buildings. Lawrence
Berkeley National Laboratory Report No. LBNL-46019: June, 2000.
Loo, L. Personal communication. March 15, 2005.
67
Maurice, G. Personal communication via e-mail. March 16, 2005.
Martinez-Villegas, N.; Florez-Velez, L.M.; and Domingo, O. 2004. Sorption of lead in soil as a
function of pH: a study case in Mexico. Chemosphere. 57: 1537- 1542.
Nagel, C. and Meyer, P. 1999. Caught between ecology and economy: end-of-life aspects of
environmentally conscious manufacturing. Computer and Industrial Engineering. 36: 781792.
Needleman, H. 2004. Lead poisoning. Annual Review of Medicine 55: 209-222.
Netscan. 2005. SoftPerfect Network Scanner. [Online] http://www.softperfect.com/.
Accessed March 9, 2005.
Nordman, B.; Piette, M.A.; Kinney, K.; and Webber, C. 1997. User Guide to Power
Management for PCs and Monitors. [Online]
http://eetd.lbl.gov/EA/Reports/39466/39466.PDF .
Accessed March 25, 2005.
Norfold, L. and Hatcher A. 1990. Electricity use in Information Technologies. Annual
Rev. Energy. 15:423-453.
NRCan (Natural Resources Canada). 2002. Office Equipment – Fact Sheet. [Online]
http://oee.nrcan.gc.ca/energystar/english/participants/library/pdf/F180_NRCAN_FS_Offi
ceEquip.pdf Accessed January 23, 2005.
Searcy, C. 2001. Green procurement Guidelines for the University of Manitoba. [Online]
http://www.i-clps.com/gp/greenprocurementpdf. Accessed January 31, 2005.
Secretariat of the Basel Convention. 2005. The Basel Convention Introductory Text. [Online]
www.basel.int/pub/basics.html. Accessed March 17, 2005.
Silicon Valley Toxics Coalition (SVTC). 2005. Clean Computer Campaign. [Online]
http://www.svtc.org/cleancc/greendesign/ecolabel.htm. Accessed March 17, 2005.
S-PLUS. 2002. S-PLUS Software: Version 6.1. Lucent Technologies Inc.
United States Environmental Protection Agency (US EPA). 2005. Energy Star Computers.
[Online]
http://www.energystar.gov/index.cfm?fuseaction=find_a_product.showProductGroup&pgw
_code=CO. Accessed March 15, 2005.
United States Department of Energy. 2005. Energy Use of Some Typical Home Appliances.
[Online] http://www.eere.energy.gov/consumerinfo/factsheets/ec7.html. Accessed on
March 14, 2005.
68
University of Guelph. 2004. Office of the President. 2004 Facts and Figures. [Online]
http://www.uoguelph.ca/info/facts/. Accessed March 11, 2005.
University of Guelph. 2005. From the president more Budget Challenges lie ahead for U of G.
[Online] http://www.uoguelph.ca/atguelph/05-01-26/newsalastair.shtml. Accessed March
28, 2005.
Williams, E. 2003. Environmental Impacts in the Production of Personal Computers. In Kuehr,
R. and Williams, E (Eds.). Computers and the Environment: Understanding and
Managing Their Impacts. London: Kluwer Academic Publishers.
Williams, E and Sasaki, Y. 2003. Strategizing the End-of-life Handling of Personal Computers:
Resell, Upgrade, Recycle. In Kuehr, R. and Williams, E (Eds.). Computers and the
Environment: Understanding and Managing Their Impacts. London: Kluwer Academic
Publishers.
Zhang, S. and Forssberg. E. 1999. Intelligent Liberation and classification of electronic scrap.
Powder technology 105: 295-301.
69
Appendix 1: Composition of a personal desktop computer
Based on a typical desktop computer, weighting ~70lbs (Handy and Harman Electronic Materials
Corp., 1999)
Content
Recycling
(% of total
Efficiency
Use/Location
Name
weight)
plastics
lead
22.9907
20% includes organics, oxides other than silica
6.2988
5% metal joining, radiation shield/CRT, PWB
structural, conductivity/housing, CRT, PWB,
80% connectors
aluminum
14.1723
germanium
0.0016
0% semiconductor/PWB
gallium
0.0013
0% semiconductor/PWB
structural, magnetivity/(steel) housing,CRT,
80% PWB
iron
20.4712
tin
1.0078
70% metal joining/PWB, CRT
copper
6.9287
90% conductivity/CRT, PWB, connectors
barium
0.0315
0% getter in vacuum tube/CRT
nickel
0.8503
structural, magnetivity/(steel) housing,CRT,
80% PWB
zinc
2.2046
60% battery, phosphor emitter/PWB, CRT
tantalum
0.0157
indium
0.0016
vanadium
0.0002
terbium
0
0% capacitors/PWB, power supply
60% transistor, rectifiers/PWB
0% red phosphor emitter/CRT
0% green phosphor activator, dopant/CRT,PWB
beryllium
0.0157
0% thermal conductivity/PWB, connectors
gold
0.0016
europium
0.0002
0% phosphor activator/PWB
titanium
0.0157
0% pigment, alloying agent/(aluminum) housing
ruthenium
0.0016
80% resistive circuit/PWB
cobalt
0.0157
structural, magnetivity/(steel) housing,CRT,
85% PWB
palladium
0.0003
95% connectivity, conductivity/PWB, connectors
manganese
0.0315
structural, magnetivity/(steel) housing,CRT,
0% PWB
silver
0.0189
antinomy
0.0094
99% connectivity, conductivity/PWB, connectors
98% conductivity/PWB, connectors
0% diodes/housing, PWB, CRT
70
bismuth
0.0063
0% wetting agent in thick film/PWB
chromium
0.0063
0% decorative, hardener/(steel) housing
cadmium
0.0094
battery, blu_green phosphor emitter/housing,
0% PWB, CRT
selenium
0.0016
niobium
0.0002
0% welding allow/housing
yttrium
0.0002
0% red phosphor emitter/CRT
rhodium
0
50% thick film conductor/PWB
platinum
0
95% thick film conductor/PWB
mercury
0.0022
0% batteries, switches/housing, PWB
arsenic
0.0013
0% doping agents in transistors/PWB
silica
24.8803
70% rectifiers/PWB
0% glass, solid state devices/CRT,PWB
71
Appendix 2: Eco-labelling Comparison
Desktop Computers
Blue Angel
Germany*
Recycling
Product take back
Most plastic and metal parts bust be
recyclable
No glued or welded connections
No metal inlays
TCO
Sweden
X
X
X
X
X
X
X
No metallic varnish, paint, or lacquer
X
Must use at least 2% recycled glass
in monitors
X
Design
Modular design
Up-gradable
Needs no special tools to
disassemble
Has one or more empty slots
European Union
(Austria, Belgium,
Nordic Swan
Demark, Finland, France, (Norway,
Finland,
Germany, Greece,
Ireland, Italy,
Sweden,
Denmark,
Luxembourg, The
Iceland)
Netherlands, Portugal,
Spain, Sweden, UK)
X
X
X
X
X
X
X
X
X
X
X
X
Packaging
No chlorinated solvents
No chlorine based plastics
No lead or cadmium based
stabilizers
X
X
X
Energy (all 4 labels have slightly
different criteria with regards to
wattage and allowable time before
modes change)
Has sleep mode
Has deep sleep mode
X
X
Must meet energy Star requirements
X
X
X
X
X
X
Guarantees
72
Unit guaranteed for 3 years
Spare parts availability - 5 years
Monitor - 1 year
X
X
X
X
User Instructions
Must explain take back procedures
X
X
Must include hazmat explanations
Must explain energy saving features
X
X
X
Must explain up-gradability options
X
X
Must explain availability of spare
parts
X
X
Plastics
No cadmium
No lead
No chlorine based plastics
No brominated flame retardants (on
parts >25g)
X
X
X
X
X
X
X
X
X
No chlorinated flame retardants (on
parts >25g)
X
X
Parts must include CAS# of flame
retardant used
X
No halogenated polymers or
halogenated organic compounds
Displays
No cadmium
No mercury
Must declare amount of mercury
used in background lighting
X
X
X
X
X
X
X
Manufacturing Process
No chlorinated solvents
X
Must adhere to Montreal Protocol
X
No CFC, HFC, HCFC
No carbon tetrachloride or 1, 1, 1 trichlorethylene
Printed Circuit Boards
No polybrominated biphenyls,
polybrominated diphenyl ethers or
chlorinated paraffins
X
X
X
X
73
Capacitors
No PCBs
X
* Blue Angel specifies no substances may be added to the plastics, which in TRGS 905, 900 or in the MAK-valueList1 as amended, are classified as
a. Carcinogenic according to EC Category Carc.Cat.1, Carc.2, or Carc.Cat3 or according to the MAK
classification III1, III2 or III3;
b. Mutagenic according to EC Category Mut.Cat.1, Mut.Cat.2 or Mut.Cat.3 or M1, M2 or M3;
c. Teratogenic according to EC Category Repr.Cat1, Repr.Cat.2 or RE/F1, RE/F2 or RE/F3
(SVTC, 2005)
74
Appendix 3: Electronics Recycler’s Pledge of True Stewardship
We, the signing and registered recycling company, agree to uphold the following:
I. We will not allow any hazardous E-waste* we handle to be sent to solid waste (non-hazardous
waste) landfills or incinerators for disposal or energy recovery, either directly or through
intermediaries.
II. Consistent with decisions of the international Basel Convention on the Control of
Transboundary Movements of Hazardous Wastes and their Disposal, we will not allow the export
of hazardous E-waste we handle to be exported from developed to developing countries** either
directly or through intermediaries.
III. We will not allow any E-waste we handle to be sent to prisons for recycling either directly or
through intermediaries.
IV. We assure that we have a certified, or otherwise comprehensive and comparable
“environmental management system” in place and our operation meets best practices.
V. We commit to ensuring that the entire recycling chain, including downstream intermediaries
and recovery operations such as smelters, are meeting all applicable environmental and health
regulations. Every effort will be made to only make use of those facilities (e.g. smelters), which
provide the most efficient and least polluting recovery services available globally.
VI. We agree to provide visible tracking of hazardous E-Waste throughout the product recycling
chain. The tracking information should show the final disposition of all hazardous waste
materials. If there is a concern about trade secrets, an independent auditor acceptable to parties
concerned can be used to verify compliance with this pledge.
VII. We agree to provide adequate assurance (e.g. bonds) to cover environmental and other costs
of the closure of our facility, and additionally to provide liability insurance for accidents and
incidents involving wastes under our control and ownership. Additionally we will ensure due
diligence throughout the product chain.
VIII. We agree to support Extended Producer Responsibility (EPR) programs and/or legislation in
order to develop viable financing mechanisms for end-of-life that provides that all legitimate
electronic recycling companies have a stake in the process.
IX. We further agree to support design for environment and toxics use reduction programs and/or
legislation for electronic products.
* Following best interpretation of the definitions of the Basel Convention, “hazardous electronic waste” will for the purposes of this pledge
include circuit boards, CRTs as well as computers, monitors, peripherals, and other electronics containing circuit boards and/or CRTs. It will also
include mercury and PCB containing components, lamps and devices. The definition of “hazardous electronic waste” will not include
75
nonhazardous wastes such as copper unless it is contaminated with a Basel hazardous waste such as lead, cadmium, PCBs, mercury etc. The
definition of “hazardous electronic waste” includes non-working materials exported for repair unless assurances exist that hazardous components
(such as CRTs or circuit boards) will not be disposed of in the importing country as a result. The definition of “hazardous electronic waste” does
not include working equipment and parts that are certified as working, that are not intended for disposal or recycling, but for re-use and resale.
The term 'hazardous e-waste' as used in this Pledge does not pertain to, nor is synonymous with any current legal US definitions of 'hazardous
waste', but is meant for the purposes of this Pledge only.
** Following the definitions of the Basel Convention and its Basel Ban Amendment, developing countries are any country not
belonging to either the European Union, the Organization of Economic Cooperation and Development (OECD) or Liechtenstein. For a
complete list of OECD and EU countries see http://www.ban.org/country_status/country_status.html and find countries shaded in gray.
(BAN, 2005)
76
Appendix 4: Computer and Electronics Recycling
The disposal of obsolete computer equipment presents a growing problem to the IT industry. The
rate of production far outstrips the rate of recovery of waste materials in this sector. It is possible
to recycle outdated or non-functional computer equipment, but it must be done through a special
program. Electronic waste cannot go in the regular blue recycling bins around campus.
The University of Guelph runs a depot program for on-campus departments and users of
computer/electronic equipment. Items are sent via the Material Handlers to Production Works in
Fergus, a division of the Guelph-Wellington Association for Community Living, who
disassemble the equipment for recovery.
Please note: due to employee safety concerns, monitors/televisions/CRTs are not accepted for
recycling by Production Works. Working monitors in good condition may be acceptable for reuse
if clearly labelled as such.
Disposal Procedures
1) Take advantage of our unstaffed drop-off times Tuesday and Thursday between 2:30pm and
4:30pm at the MacKinnon loading dock (located on Trent Lane). Recommended for smaller
loads, and as a quick way to get the equipment out of your area. Please notify the Sustainability
Coordinator (ext. 58129 or recycle@pr.uoguelph.ca) if you have made a drop-off during these
times.
2) For larger loads that can wait around longer, you can arrange for pick up by calling or emailing
the Sustainability Coordinator at ext. 58129 or recycle@pr.uoguelph.ca, with the following
information:
a) type(s) of equipment
b) exact quantity of equipment
c) precise location of equipment
Example: 4 inkjet printers, 5 keyboards, 5 towers/hard drives and 1 box of miscellaneous parts in
room 054 FACS Building
For secure destruction of sensitive data, Valu-Shred, a company based in Mississauga, will take
larger quantities of properly packaged equipment. More information is available through their
website, www.valushred.com, and by our previous contact person, Mickey Dobran, 1-905-6726597 or cell phone 1-416-268-5249.
77
Computer Recycling Program - University of Guelph
Standard Procedure Information Sheet for Transportation Services
Note: We do not accept monitors and television sets (CRTs). All customers are notified of this policy, and
encouraged to make their own arrangements through local reuse outlets.
1 - Unstaffed drop-off for on-campus depot. MacKinnon (068) loading dock off Trent Lane.
Tuesdays and Thursdays, 2:30-4:30.
• Customers with small loads are encouraged to drop off materials during these times and
notify the Recycling and Waste Management Coordinator (RWMC) by phone or email.
• RWMC ensures removal to storage depot (009 MacKinnon)
2 - Small Pick-ups (under 12 units) - delivered to 009 MacKinnon
• RWMC places Work Order for Material Handlers
• Customer must specify amount and type of waste equipment, and exact location.
• Equipment is removed from customer location, and brought to on-campus depot in 009
MacKinnon.
• Material Handlers note numbers and types of equipment picked up on Work Order
• When depot is full, RWMC places Work Order for equipment in depot to be delivered to
Production Works Co-op.
3 - Large Pick-ups (12 units and over) - delivered directly to Production Works Co-op
• RWMC places Work Order for Material Handlers
• Customer must specify amount and type of waste equipment, and exact location.
• Equipment is removed from customer location, and delivered directly to Production
Works Co-op.
• Material Handlers note numbers and types of equipment picked up on Work Order
• A phone call 1 day in advance to Production Works is preferred, once the scheduled date
for pick up is set by Transportation Services
Contacts
Gillian Maurice (Recycling and Waste Management Coordinator)
ext. 58129 ; recycle@pr.uoguelph.ca
Brian Robinson (Custodial Services Supervisor)
ext. 58178 ; cell 220-8594 ; brian@pr.uoguelph.ca
Production Works Co-op (division of Community Living Guelph Wellington)
contact: Shelley Dano (519) 787-1539 ext. 51; sdano@gwacl.on.ca
EMJ Datasystems Ltd. (Backup location for computer recycling)
contact: Mike Hall (519) 837-2444
78
Appendix 5: Sampling Sites
6
10
8
9
5
4
3
2
7
1
8
1. CropScience
2. Graham Hall
3. Hutt
4. MacNaughton
5. McKinnon
6. OVC Microlab
7. Powell
8. Thornbrough and Renyolds
9. MacLaughlin Library
10. OVC Library
79
Appendix 6: S-Plus Statistical Summaries for all Monitoring Times
in Richards Building
***
Summary Statistics for data in:
Min:
1st Qu.:
Mean:
Median:
3rd Qu.:
Max:
Total N:
NA's :
Std Dev.:
***
10:00
29.17000
42.08500
48.59856
55.00000
55.41650
61.01690
7.00000
0.00000
12.70023
14:00
30.00000
45.41500
52.26143
59.17000
61.25000
63.33000
7.00000
0.00000
14.47911
Monday through Sunday
19:00
26.67000
28.75000
33.45143
30.00000
30.83000
58.33000
7.00000
0.00000
11.06908
Summary Statistics for data in:
Monday through Friday
10:00
14:00
19:00
Min: 52.500000 58.330000 26.6700
1st Qu.: 55.000000 59.170000 30.0000
Mean: 55.870000 60.666000 35.3320
Median: 55.000000 60.000000 30.8300
3rd Qu.: 55.830000 62.500000 30.8300
Max: 61.020000 63.330000 58.3300
Total N: 5.000000 5.000000 5.0000
NA's : 0.000000 0.000000 0.0000
Std Dev.: 3.138264 2.156323 12.9702
***
Summary Statistics for data in:
Min:
1st Qu.:
Mean:
Median:
3rd Qu.:
Max:
Total N:
NA's :
Std Dev.:
10:00
29.170000
29.795000
30.420000
30.420000
31.045000
31.670000
2.000000
0.000000
1.767767
14:00
30.000000
30.625000
31.250000
31.250000
31.875000
32.500000
2.000000
0.000000
1.767767
Saturday and Sunday
19:00
28.3300000
28.5400000
28.7500000
28.7500000
28.9600000
29.1700000
2.0000000
0.0000000
0.5939697
80
Appendix 7: Energy Consumption Raw Data
Table 1: Current Energy Use
Location
Crop Science
Graham Hall
Hutt
MacNaughton
McKinnon
OVC Microlab
Powell
Thornbrough
and Renyolds
MacLaughlin
Library
OVC Library
Faculty and
Graduate Students
Total
# of
Computers
CRT
Flat
Average
use (hrs)
Standby
(hrs)
Days
open
35
25
68
66
60
50
35
35
25
68
66
60
50
35
0
0
0
0
0
0
0
13.5
13.5
24
13.5
13.5
12
9
10.5
10.5
0
0
0
12
0
7
7
7
7
6
7
5
Total energy
use/year
(kW/yr)
50,450
36,036
160,393
87,567
68,234
65,520
22,113
350
35
315
12
12
7
293,530
23,482
264
119
145
24
0
7
470,486
37,639
25
25
0
12
12
7
32,760
2,621
2834
1417
1417
9.5
0
7
2,645,992
211,679
3812
1935
1877
156.5
57
74
3,933,082
314,647
Cost ($)
4,036
2,883
12,831
7,005
5,459
5,242
1,769
Table 2: Best-case Scenario for Energy Use
Location
Crop Science
Graham Hall
Hutt
MacNaughton
McKinnon
OVC Microlab
Powell
Thornbrough and
Renyolds
MacLaughlin
Library
OVC Library
Faculty and
Graduate Students
Total
# of
Computers
CRT
Flat
Average
use (hrs)
Standby
(hrs)
Days
open
Total energy
use/year
(kWhr/yr)
Cost ($)
35
25
68
66
60
50
35
0
0
0
0
0
0
0
35
25
68
66
60
50
35
6.75
6.75
6.75
6.75
6.75
7.75
4.5
6.75
6.75
6.75
6.75
6.75
7.75
4.5
5
5
5
5
6
7
5
8,907
6,362
17,304
16,795
18,322
20,452
5,938
713
509
1,384
1,344
1,466
1,636
475
350
0
350
12
9
7
210,210
16,817
264
0
264
9.5
9.5
7
132,372
10,590
25
0
25
7.75
7.75
7
10,226
818
2834
0
2834
8
5
677,893
54,231
3812
0
3812
83
64
1,124,782
89,983
72
81
Table 3: Worst-case Scenario for Energy Use
Location
Crop Science
Graham Hall
Hutt
MacNaughton
McKinnon
OVC Microlab
Powell
Thornbrough and
Renyolds
MacLaughlin
Library
OVC Library
Faculty and
Graduate Students
Total
# of
Computers
CRT
Flat
Average
use (hrs)
Standby
(hrs)
Days
open
35
25
68
66
60
50
35
35
25
68
66
60
50
35
0
0
0
0
0
0
0
24
24
24
24
24
24
24
0
0
0
0
0
0
0
7
7
7
7
7
7
7
Total energy
use/year
(kWhr/yr)
82,555
58,968
160,393
155,676
141,523
117,936
82,555
350
350
0
24
0
7
825,552
66,044
264
264
0
24
0
7
622,702
49,816
25
25
0
24
0
7
58,968
4,717
2834
2834
0
24
0
7
6,684,612
534,769
3812
3812
0
264
0
77
8,991,441
719,315
Cost ($)
6,604
4,717
12,831
12,454
11,322
9,435
6,604
82