Introduction to Life Cycle Assessment

5.1 Principles of LCA

5.1.1 How to Assess Environmental Impact Environmental awareness of our industrialised societies has been developing rapidly for the last several decades. A shift in attitude towards the environment has brought a new term environmentally friendly. Although a commonly-used description, it is not easy in fact to determine which products or, in broader context, which forms of human activity, are environmentally friendly. Another crucial issue is to nd out what can be done to improve the environmental prole of a certain product or process. How can we assess possible benets gained by changing the mode of production, usage or disposal of a product. In other words: how do we determine which one of several alternatives is more, or the most, environmentally friendly. Answers to such questions are important for a sustainable development. It has been proved that when evaluating the environmental friendliness of a product, intuition is not enough. In a survey on green milk packaging a majority would certainly nd a returnable milk bottle much preferable to a disposable milk carton. The reason is that the bottle is recyclable and the cardboard cartons are not. The bottle is therefore expected to lead to signicantly lesser amounts of waste in comparison to cardboard cartons. These two environmental problems recycling and waste production are broadly reported in the mass media. In reality the overall difference between the alternatives is insignicant. The two kinds of packaging contribute to completely different environmental impacts. Admittedly, glass can be reused and recycled, but it is connected with high costs of transportation and cleaning. This is not the case for disposable cartons. In conclusion, the most advisable solution would combine the environmental advantages of the two considered alternatives, namely being recyclable but at the same time
5 introduction to life cycle assessment

light (for instance a square polycarbonate bottle) [Heijungs et al., 1996]. The environment is a complicated network of many unexpected and unexplained interrelationships. Sometimes a solution which appears to be excellent might only shift the problem to another life stage of the product, or to another sort of impact. Life cycle assessment (LCA), is a rather new tool in environmental management, which has the capacity at least in principle to answer these seemingly easy questions: Is a product environmentally friendly? Which product is greener? What is then life cycle assessment and how can it be used. These are the questions we will address in this chapter.

In this Chapter
1. Principles of LCA. How to Assess Environmental Impact. Denitions of LCA. The Goals and Applications of LCA. Developments of LCA. 2. The Qualitative (approximate) LCA. The Red Flag Method. The MET Matrix. 3. Quantitative LCA Methods. The Components of Quantitative Methods. Goal Denition and Scope. The Functional Unit. System Boundaries and the Process Tree. 4. Inventory Analysis and Allocation. Inventory Table. Allocation.

87

5.1.2 Denitions of LCA According to the ISO DIS standards, LCA is dened as a method for analysing and determining the environmental impact along the product chain of (technical) systems. It includes the various types of technical conversions that occur in the manufacturing process. These consist of the change of material chemistry (chemical conversion), material formulation, or material structure; the removal of material resulting in an increase of (primary) outputs over the inputs; and the joining and assembly of materials resulting in a decrease of (primary) outputs over the inputs. This general description has been specied in two widely known denitions of LCA. According to ISO 14040, the formal denition of LCA is as follows:
LCA is a technique for assessing the environmental aspects and potential impacts associated with a product by: Compiling an inventory of relevant inputs and outputs of a product system. Evaluating the potential environmental impacts associated with those inputs and outputs. Interpreting the results of the inventory analysis and impact assessment phases in relation to the objectives of the study. The denition by SETAC (Society of Environmental Toxicology and Chemistry), which was a pioneer in publishing its Code of Practice, states that: Life Cycle Assessment is a process to evaluate the environmental burdens associated with a product, process, or activity by identifying and quantifying energy and materials used and wastes released to the environment; to assess the impact of those energy and materials used and releases to the environment; and to identify and evaluate opportunities to affect environmental improvements. The assessment includes the entire life cycle of the product, process or activity, encompassing, extracting and processing raw materials; manufacturing, transportation and distribution; use, re-use, maintenance; recycling, and nal disposal. Since the late sixties and early seventies much attention has been given to life cycle technology. Over the years life cycle assessment has developed (Box 5.1). It means that the analysis called LCA gathers a number of more or less different methods. What they have in common is the holistic viewpoint on the life cycle and dealing with environmental aspects of all emissions and material consumption resulting from the life cycle. Nevertheless, there is still no internationally accepted methodology of LCA. The international standard now developing is based on the ISO 14040 series. With respect to the way of conducting an LCA we can separate qualitative and quantitative methods [Jensen et al., 1997].
88

LCA in Brief
LCA is a technique for evaluating the environmental effects of a product or a service over the entire period of its life cycle (from cradle-to-grave).

Qualitative methods draw conclusions straight from the life cycle. The quantitative methods evaluate the environmental impacts by mathematical processing of the data describing the life cycle. They may even result in the calculation of a single score representing the environmental friendliness of a product.

5.1.3 The Goals and Applications of LCA LCA assess the environmental effects of a product or service or, more commonly, the effects of a change in the production or design of a product or service. The goals and applications of LCA range over a scale from short to long term. It includes:
Short-term process engineering. Design and optimization in a life cycle (type 1). Product comparisons including product design and product improvement. Eco-labelling in the medium and long term (type 2). Long-term strategic planning (type 3). Each goal requires its own type of analysis and modelling. Data requirements can then be specied more precisely, both for case applications and for generic databases. Thus when performing an LCA, all the emissions and the resource consumption which enter or leave a life cycle are translated into the environmental problems that they potentially may contribute to. The two terms environmental effects and life cycle both need to be properly understood. Environmental effects are the consequences of a physical interaction between a system studied and the environment. In practical use all environmental effects are represented by several categories of environmental problems. The most commonly used are: Resource depletion Global warming Ozone depletion Human toxicity Ecotoxicity Photochemical oxidation Acidication Eutrophication Land use Others (including solid waste, heavy metals, carcinogens, radiation, species extinction, noise).
introduction to life cycle assessment 5

The other term, crucial to understanding the holistic approach of the life cycle assessment, is the life cycle itself. It encompasses all the processes required to full the function provided by a product or service (Figure 5.1), [Stachowicz, 2001; Walz, 2000]. At present LCA is used for the following elds of application:

Infrastructure Process industry Energy production Transportation Heavy industry Consumer goods Livelihood

Box 5.1 The History of Life Cycle Assessment

Several tools for assessing environmental impacts The roots of Life Cycle Assessment reach back to the 1970s. At the time methods such as integral environmental analysis, ecobalances, resource and environmental prole analysis etc were used. Over the years the experiences of working with these tools fed into the Life Cycle Assessment method around 1990. At a conference organised in 1991 by the Society of Environmental Toxicology and Chemistry (SETAC) it was agreed that the proper name for LCA should be Life Cycle Assessment (rather than Analysis). In 1997 the International Organisation for Standardization published its rst standard for LCA , ISO 14040. LCA is one of several tools for assessing the environmental impact of a product or service. Others include Environmental Impact Assessment (EIA), Ecological Risk Assessment (ERA), Material Flows Analysis (MFA) as well as several more economic tools such as Cost Benet Analysis (CBA). LCA is (normally) by far the most comprehensive (all inclusive) of these. 1970s and 1980s studies on beverage packaging The rst study considered to be an LCA was made in 196970 for Coca-Cola by Midwest Research Institute in the US. This was one of several studies on packaging and waste. The question was which is better to manufacture and use: beverage (steel or aluminium) cans, plastic bottles, rellable glass bottles or disposable containers. Coca-Cola asked for an all inclusive study of the energy, material and environmental costs of the entire chain from resource extraction to manufacture, use and nally waste of the containers. The researchers conducted a so-called Resource and Environmental Prole Analysis (REPA) for all the alternatives. The rst European study was done by Ian Boustedt at the Open University in the UK in 1972. Inspired by the Coca-Cola study and its development he constructed a case on milk bottles to be used in a text book. The work tted well with the at times rather heated public debate on the pros and cons of returnable and non-returnable bottles for milk, beer etc. Later in the 1970s a German study was done on meat trays, and in Sweden on PVC bottles for Tetrapak, a very large company producing containers for beverages. The result came out in 1973, at the time when plastic bottles could be incinerated in the rst waste incineration plants (in Malm) and the impact of HCl, produced in the incineration, could be estimated. The 1990s - LCA methodology develops In the coming several years hundreds of similar studies were published, especial in the USA, UK, Germany and Sweden. Most often the databases used were made public, which was very important for the possibility of conducting more studies. In the mid 1980s rst Switzerland, completed in 1990, and later Denmark and Sweden published very large packaging studies, in early 1990 and 1991. These early studies were mainly concerned with the energy and material used and waste released during the life cycle. It was not until later that the impact assessment was included in a more serious manner. The rst scientic conference on LCA was organised by the Society of Environmental Toxicology and Chemistry (SETAC) in 1991. Methodology was a main topic to discuss. Especially the reporting of environmental impacts was rather simple. Working groups were organised to develop recommendations for industry. LCA of products was now considered a more important tool for environmental improvement than just minimising the emissions from production. The main impact may be elsewhere, not least in the waste phase. Into the 2000s Standardisations and Code of Practices Efforts were made to make LCA data publicly available and to develop software for calculations. There were also strong forces requiring a standardisation of the methodology to make different studies comparable, especially to make it less easy to use LCA for promoting specic products. The rst Code of Practice for LCA was published by SETAC in 1993. This work on methodology, especially concerning impact assessment, is still ongoing. The ISO standards for LCA have been published since 1997 (ISO 14040) and later 1998, 2000 and 2002. In 2005 the European Commission introduced its European Platform on Life Cycle Assessment as a component in the IPP Directive. Its initial phase, 2005 to 2008, intends to reduce costs for LCA studies, to harmonise quality control, and to produce consistent data basis. The intention is to establish LCA as a reliable support in decision making and promote its acceptance in governments. Source: Baumann and Tillman, 2004. LR

5 introduction to life cycle assessment

89

Transport
Trade Product usage

Manufacturing process

RESOURCES RAW MATERIALS ENERGY

Recycle or reuse

Raw material acquisition

Waste managemet

AIR, WATER AND SOIL EMISSIONS SOLID WASTE NON-MATERIAL EMISSIONS

ning issues is more difcult. Radical changes in technologies, legislative constraints or policy goals need to be anticipated. Future developments of key technologies and entire economic sectors will thus affect the outcome of the LCA. They have to be dened carefully [Frischknecht, 1997]. Current LCA integrates over time and space. A desegration of these two parameters are needed to get a more precise result. First environmental impact is depending on location. For example acidication is very different in different places. Desegregation of the inventory of impact as to location is a matter of practicability. It has been done in several studies, to allow for a differentiated impact assessment. Desegregation in time is needed to allow for a differentiated impact assessment. One reason for attening out time in current practice is that LCA is supposed to support decision making and affect future decisions, while for an actual system a substantial part of the processes have already taken place. For example, the factory which is bringing out a new car next year will itself have been set up some 10 years ago. The decisions in car design will not inuence past decisions but only exert inuences on production facilities yet to be built.

n Tra

spo rt

rt spo

Tra n

Tr a n spo rt

Tra

ort nsp

5.2 The Qualitative (approximate) LCA

Figure 5.1 Product life cycle [Stachowicz, 2001; Walz, 2000].

5.1.4 Developments of LCA There are many shortcomings in the applications of LCA techniques. Up to the present time, the main focus of research in LCA has been on developing LCA as a tool rather than a system [Hauschild and Wenzel, 1998]. Hence, there has been an emphasis upon assessment of potential environmental effects. Relatively little attention has been paid to, for example, the process of dening alternatives for consideration in an LCA, or choice of weighting factors. Further research could allow us to develop LCA as a system capable of assessing the environmental impact per function, system needed or money spent. Such a new approach would allow us to search for conditions for reaching maximum cost-effectiveness with respect to the environment for a function or product. Then the ratio of system cost/environmental impact could be maximized. A method to do this will be discussed later, primarily developed and tested to support Life Cycle Management (LCM) of capital assets [Stavenuiter, 2002]. Todays LCA approaches are valid only for incremental changes in the product of interest and dened geopolitical regions. To use LCA to provide answers to long-term plan90

5.2.1 The Red Flag Method Qualitative LCA methods do not use systematic computational procedures to assess the environmental prole of the system under study. They analyse the life cycle of a product in environmental terms directly on the basis of emissions released and the consumption of raw materials. Assessing the seriousness of the impacts directly from the impact table requires thorough training and extensive knowledge. A decisive role is played by relevant experiences of the expert carrying out the evaluation. The red ag method (RFM) may serve as an example of a qualitative method. There are a number of companies working with RFM, for instance Philips. The rst step is, as usual, preparing an impact table. This gathers all emissions and material consumption during the whole life cycle of a product. Then, the items which are harmful to the environment are redagged. Red ags can occur along with emissions of CFCs (chlorouorocarbons), toxic substances, greenhouse gases, etc. or where scarce materials are consumed. The red-agged process or product should then be given special attention and if possible excluded from the life cycle of the product. Even though this approach is fairly easy, there is a major obstacle. The red ags many times are placed in nearly each process or life stage without, any distinction between small and large quantities of unwanted emissions. In practice not all these
introduction to life cycle assessment 5

stages can be removed or changed. In these cases the red ag method does not provide a sufciently qualied evaluation and is not useful. A piece of the impact table for production of 1kg of EPDM rubber with ags is shown in Table 5.1.

Impact assessment Gole definition and scope

Improvement assessment

5.2.2 The MET Matrix Another qualitative method for assessing life cycle of a product is the so-called MET matrix (materials, energy and toxicity). A MET analysis consists of ve stages. The rst is a discussion of the social relevance of the products functions. Then the life cycle of the product under study is determined and all the relevant data is gathered. Next the data is used in which is the core of the MET matrix method: completing the matrix (Table 3.1). The processes in the life cycle are then entered in the matrix divided into three categories: material consumption, energy consumption, and emissions of toxic substances. As in the case of Red Flag Method, completion of the MET matrix can be done only with an aid of environmental experts. Finally, when the most signicant environmental problems are identied, possible steps to improvement of the product or service should be outlined. The qualitative methods in general have poor reproducibility. The reason is that they require support provided by experienced environmental experts, and that experts often come to different conclusions. The scientic support for making reproducible and reliable judgements is so far lacking.

Inventory analysis

Figure 5.2 Interrelation of LCA phases [Hillary, 1995].

5.3 Quantitative LCA Methods

5.3.1 The Components of Quantitative Methods There are a number of different quantitative LCA techniques. These are in practice applied as a group of methods which use classication, characterisation, normalisation and weighting. The most important are:
Eco-points Eco-indicator EPS system MIPS concept

The methodological framework of all the LCA techniques is based on ISO standards 14040-43. A complete LCA consistent with ISO standards consists of four interrelated phases (compare with the denition of LCA given by ISO): 1. Goal denition and scope. 2. Inventory analysis. 3. Impact assessment with four sub-phases: classication, characterisation, normalisation, weighting. 4. Improvement assessment. Interrelations among the LCA phases make LCA an iterative process (Figure 5.2), [Hillary, 1995]. The calculation and evaluation procedure is repeated until the analysis reaches the required level of detail and reliability. The rst step in an LCA is a raw assessment to determine critical points in the life cycle and nd directions for further studies. Such a quick analysis is called screening. Sometimes it is enough to answer all the questions asked in the goal denition. Goal denition and scope is crucial for all the other phases. These include gathering data, that is building a model of the life cycle, choosing appropriate environmental effects to consider (local, global?), and drawing conclusions to answer the questions asked at the beginning of the project. Nevertheless, sometimes a previously established goal of the study needs
91

Table 5.1 Impact for production of 1kg of EPDM rubber. No 1 2 3 4 5 6 7 8 9 10 11 12 Substance Baryte Nickel Sand Acetone Cd He Methane As K Na Cr Compartment Raw materials Unit g mg g g g mg g mg mg g g g Amount 7.07 23.9 4.39 954 978 110 9.94 1.24 555 26.5 460 72.6

O O O O O O

Raw materials Raw materials Air emissions Air emissions Air emissions Air emissions Water emissions Water emissions Water emissions Soil emissions Soil emissions

Oil biodegradable

5 introduction to life cycle assessment

to be changed to some extent, for instance when unforeseen obstacles arise (insufcient or unavailable data) or additional information arrives. The last step, the improvement assessment phase, is performed in accordance with the goal of the study and on the basis of results from the impact assessment phase. This, in turn, is achieved by applying the computational procedure to the data in the inventory table.

5.3.2 Goal Denition and Scope In the goal denition and scope phase the unambiguous and clear description of the goal of the study and its scope must be developed. The product (or service) to be assessed is dened, a functional basis for comparison in case of comparative analysis is chosen and, in general, the questions to be answered are established. The scope of the study sets requirements to the desirable level of detail. The main issues to consider in this stage are:
Purpose of the study: Why is the analysis being performed? What is the end use of the LCA? To whom are the results addressed? Specify the product to be investigated (functional unit). Scope of the study: depth and breadth (system boundaries). As far as the LCA end use is concerned there are several basic possibilities: Product or process improvement. Product or process design. Publication of information on the product. Granting of an eco-label. Exclusion or admission of products from or to the market. Formulation of company policy (purchasing, waste management, product range, how to invest the money).

dene a function of a product rather than a product itself. The measure of performance which the system delivers is called a functional unit. The functional unit provides a reasonable point of reference when comparing different products. Two products, A and B, may have different performance characteristics even though they full the same function. An illustrative example is the comparison of different kinds of milk packaging, already discussed above. Two possible alternatives are: a milk carton and a returnable glass bottle. A glass bottle can be used ten or more times, whereas a milk carton can be used only once. On the other hand, a milk carton does not need washing and additional transportation. When comparing one carton and one bottle we could conclude that carton is the environmentally best choice. If the functional unit of the two packages is established, however, the analysis are not distorted by unfair assumptions. Considered for example, that the packaging for 10 litres of milk could be a functional unit. In this case we have to com-

Production of fertilizers Production of pesticides

Production of other inputs energy

Production of rapeseed Harvest Rapeseed Transport Processing Scrap Rapeseed oil

Straw and roots

The intended audience is especially important to consider when preparing the presentation and communication of results. An LCA may be addressed to scientists, environmentalists, NGOs, the public (media, consumers). The manner of presenting the results should be tailored to meet their needs.

Production of methanol Methanol

Transport Estrification Biodiesel (RME) Transport Fuel station Combustion in engine

Glycerol

5.3.3 The Functional Unit An LCA of a product must have clearly specied functions to be assessed. If, for instance, the product is a washing machine, it is important to describe its performance characteristics. These state what minimum quality standards the washing machine must meet: the degree of cleanliness and the degree to which clothes should be dried, how long the machine should work and how frequently it is to be used, the amount of clothes that can be washed at one time, etc. That is, it is important to
92

Figure 5.3 Process tree of the production and use of biodiesel [Hillary, 1995]. introduction to life cycle assessment 5

pare 10 milk cartons to 1 bottle and 9 washings (assuming 9 return trips of the bottle). Another example of a functional unit is when one wish to compare different anti-corrosive paints used for protecting a metal surface. The functional unit in this case might be the amount of paint which covers a certain surface for a certain time , e.g. 1 m2 painted for 2 years. We then compare the different properties of the paints, the lifetime of coating, and the ability to cover a specied surface instead of the amount of paint.

5.3.4 System Boundaries The next vital task in the goal and scope denition step is to dene system boundaries. The necessity of dening system boundaries results from the fact that the main technique applied in any LCA is modelling. A function fullled by the product is represented by a model of the complex technical system. This consists of subsequent processes required to produce, transport, use and dispose of a product. The model is graphically illustrated by a process tree (a process ow chart) (Figure 5.3) and is used in the inventory analysis. Moreover, models of environmental mechanisms are created to translate inows and outows from the life cycle into the environmental impacts they may contribute to. For example, SO2 emissions could increase acidity. This, in turn, can cause soil and water impairment, inuence the quality of the ecosystem, deteriorate the living condition for animals and plants, etc. Such models are the basis for the impact assessment phase. A model, by its denition, is a representation of reality but at the same time it is a simplication of reality. It means that the reality must be distorted to some extent in a model. On the other hand, one cannot avoid this problem. The system without simplications is too complex to analyse. If a product system should include all the processes from cradle-to-grave one has to follow each inow or outow. This include, for example, crude oil, solar energy, iron ore from the environment, and all nal waste released to the environment, i.e. emissions to air, water, soil, radiation. As a result the process tree would be practically endlessly branched. Product systems are usually interconnected in a complex way, and it is impossible to isolate a single life cycle of a product without coming up against life cycles of other products. Thus e.g. in an LCA on glass bottles, trucks are used for transportation, so life cycle of a truck should be involved into the LCA. In the life cycle of the truck, steel is used to produce many parts of the vehicle, coal is needed to produce steel, steel is transported by trucks, etc. This phenomenon is called endless regression.

To avoid such a problem the boundaries of the system must be dened. The system under study has to be separated from the environment as well as from other products and systems. The typical question when dening the system boundaries is whether to include the production of capital goods or not. In a majority of LCAs capital goods, e.g. equipment of a workshop, are neglected. This assumption does not lead to important distortions of the nal LCA outcome. In some cases, however, neglecting capital goods signicantly underestimates environmental burdens. This applies to, for example, electricity production. It has been shown, that the production of capital goods constitutes about 30% of the total environmental impact resulting from an average generation of electricity. Another common problem is presented by agricultural areas, which can be seen as a part of nature or as a part of the production system. For instance pesticides can be treated as emissions if agricultural areas are a part of nature. Otherwise (when agricultural areas are seen as a part of an economic system) only the part of pesticides, which leaves, a eld somehow (evaporate or are accidentally sprayed outside) are perceived as emissions. The rest, which are not released to the environment, remains a part of a system. A similar problem which substances should leave the life cycle concerns dumping waste. It can be regarded as nal waste released to the environment or the start for long-term waste processing. To narrow down the system boundaries, one uses cut-off rules. Thus if the mass or economic value of the inow is lower than a certain percentage (a previously set threshold) of the total inow it is excluded from further analysis. The same applies when the contribution from an inow to the environmental load is below a certain percentage of the total inow. Carefully and properly specied goals and scope help to develop the model of the product in such a way that the simplications and thus distortions have only an insignicant inuence on the results. This is vital for getting reliable answers from an LCA. This challenging task undoubtedly depends to some degree on subjective decisions and requires a lot of experience.

5.4 Inventory Analysis and Allocation

5.4.1 Inventory Table The inventory phase is the core of an LCA and is a common feature of any LCA. During this phase all the material ows, the energy ows and all the waste streams released to the environment over the whole life cycle of the system under study are identied and quantied. The nal result of the inventory analysis is an inventory table. The inventory phase has four separate sub-stages:
93

5 introduction to life cycle assessment

Constructing a process ow chart (so-called process tree). Collecting the data. Relating the data to a chosen functional unit (allocation). Developing an overall energy and material balance (all inputs and outputs from the entire life cycle) an inventory table.

To develop a life cycle it is best to start from the product itself and then follow all upstream and downstream life stages.
Table 5.2 Selected items in an inventory table for the production of 1kg of PVC derived from SimaPro. No Substance 1 2 3 4 5 6 7 8 9 10 22 23 24 25 26 36 37 38 39 40 41 42 60 61 62 63 64 Air Barrage water Baryte Bauxite Bentomite Clay minerals Coal Crude oil IDEMAT Dolomite Energy (undened) Cl2 CO CO2 CxHy Dust Acid as H+ BOD Calcium ions Cl COD CxHy Detergent/oil Mineral waste Plastic production waste Slag Unspecied Occupied area as industrial area Compartment Unit Raw material Raw material Raw material Raw material Raw material Raw material Raw material Raw material Raw material Raw material Air Air Air Air Air Wastewater Wastewater Wastewater Wastewater Wastewater Wastewater Wastewater Solid waste Solid waste Solid waste Solid waste Non material (land use) g kg mg mg mg mg g g mg MJ mg g kg g g mg mg mg g mg mg mg g mg g mg m
2

Total 220 99 82 440 32 9 135 400 2 113 2 2.3 2 19 29 48 850 47 37 76 26 49 42 440 9.4 9 400

This makes the LCA work systematic. Possible upstream stages are: extraction and production of raw materials, production of components (intermediates, semi-nished products, different parts), production of auxiliary materials (such as solvents, catalysts, etc.) and eventually production of the product itself. Among downstream stages are: use of the product, waste handling, processes of recycling and reuse if needed. Additionally, between all these processes, usually transport is needed and similarly the production of the energy carriers (electricity, steam) occurs along with almost all processes and life stages. A result of this step is illustrated by the process tree of the production of biodiesel (Figure 5.3), [Hillary, 1995]. On the basis of such a process tree, more detailed data is collected as required by the previously dened goal and scope (required level of details). All these actions have the same goal, namely to obtain a list of all inputs (materials consumed) and outputs (emissions) connected with the life cycle of the product. The data should be quantitative and are used to build an inventory table. An example of an inventory table for production of PVC is shown in Table 5.2. Note that this example of an inventory table is signicantly abridged (it contains 27 out of 64 items). To obtain such a table one should link the data describing the processes involved to produce the functional unit (e.g. how much CO2 is released in conjunction with the production of 10 milk cartons).

5.4.2 Allocation Very often a process fulls two or more functions or gives two or several of usable outputs. They are multi-output processes. Then we have to determine which part of the total emissions and material consumption should be attributed to each specic product. The same applies to multi-input processes. Petrol production can serve as an example of a multi-output process. It provides several products in fractional distillation of crude oil: not only petrol but also kerosene, diesel oil, and mazout. The question is how to divide emissions and resource consumption over the petrol itself. An example of a multi-input process is a plastic bag. When performing an LCA for a plastic bag, we assume that at the end of its life cycle it is incinerated. However, there are many other products incinerated at one time. To what extent is the bag responsible for chemicals emitted from the incineration plant? The problem of how to divide emissions and material consumption between several product or processes is called allocation. Several methods have been developed to deal with allocation. Substitution of allocation no allocation in fact. As allocation always require more or less subjective decisions, ISO recintroduction to life cycle assessment 5

94

ommends to avoid allocation if possible. This can be done by extending the system boundaries i.e. by including processes that would be needed to make the same by-product in the conventional way. As an example, we can imagine a process in which a usable quantity of steam is produced as an additional output. It can be used to avoid the production of steam by more conventional means. This is an additional gain resulting from the process associated with the analysed product. This fact should be reected in the main products environmental prole. Then the environmental load of the avoided steam production may be subtracted from the overall environmental burden of the process. In this way one can calculate the part of emissions and material consumption that the main product is responsible for, and the rest is ascribed to the steam. The material consumed and emissions released, from the traditional way of producing steam are subtracted. It is not always easy to say how the steam would be produced alternatively, i.e. what a conventional method of steam production actually is. Another typical example is electricity production in conjunction with waste incineration. The main purpose of this process is waste utilisation, but the simultaneously generated electricity is an additional benet. Allocation based on natural causality in other words depending on ones common sense. In cases of combined waste incineration, SOX emissions should be allocated in relation to the S-content of different products, i.e. the more sulphur a certain product contains the more it is responsible for emissions of sulphur oxides. If a fraction of waste does not contain any sulphur, one may say it is not responsible for releasing sulphur oxides. Regrettably, there are plenty of examples of allocation problems, which this principle cannot solve. Allocation based on physical parameters such as mass, energy, etc. Let us consider two usable products in a sawmill: wooden boards, as a main product, and sawdust as a by-product. When performing an LCA of wooden boards, an allocation problem will arise. An appropriate part of the environmental impact of the boards themselves can be derived directly from mass balance between outputs. If, for example, 40% of the total mass of the wood processed gives the sawdust, one can ascribe 40% of the environmental load to sawdust. Another example is naphtha cracking. Note that if this rule were applied in case of allocating steam, it would lead to ambiguous results since the mass of the steam is incomparably smaller. Allocation based on economic values (prices). This principle is analogous to the previous one except that here economic values are the criteria. If, in the example of the sawmill, the sawdust contributes 20% of the value generated by the sawmill, one can allocate 20% of the environmental load to this
5 introduction to life cycle assessment

Allocation Techniques
Substitution of allocation. Natural causality allocation. Allocation based on physical parameters. Allocation based on economic values. Arbitrary allocation.

by-product. Usually the main product is the most valuable, and has the highest price. By applying this method, the product for which the process is carried out is the most responsible for the total environmental burden. Prices, however, tend to change in time. Consequently the economic situation may inuence an LCA although the environmental prole of a process itself remains the same. Arbitrary allocation. The contribution of each co-product in the overall emissions and material consumption can be also imposed arbitrarily, e.g. equally for each product, 100% of emissions to one product, or any other random distribution.

95

Study Questions
1. Give your own denition of LCA. 2. For what is LCA needed? 3. Which are the differences between quantitative and qualitative LCA methods. 4. What is a goal denition and scope of an LCA? 5. How should you dene a functional unit? Will we obtain the same LCA results for product treated as different functional units? Give examples. 6. Which are the difculties to decide on systems boundaries? Give example of cut off rules. 7. Dene system boundaries for a simple product. 8. Write a simple process tree. 9. Which are the methods for allocation, both upstream and downstream? 10. Make an inventory analysis for a spoonful of tea.

Internet Resources
Society of Environmental Toxicology and Chemistry

http://www.setac.org http://www.setac.org/htdocs/who_intgrp_lca.html
ISO organisation

http://www.iso.org/
PR Consultants Life Cycle Assessment

http://www.pre.nl/life_cycle_assessment/default.htm
US Environmental Protection Agency Life-Cycle Assessment LCAccess

http://www.epa.gov/ORD/NRMRL/lcaccess
Life Cycle Assessment Links

http://www.life-cycle.org/

Abbreviations
CBA EIA EM EPS ERA LCA LCM MET MFA MIPS NGO RFM Cost Benet Analysis. Environmental Impact Assessment. Environmental Management. Environmental Priority Strategies. Ecological Risk Assessment. Life Cycle Assessment. Life Cycle Management. Materials Energy Toxicity. Material Flows Analysis. Material Input Per Service unit. Non Governmental Organisations. Red Flag Method.

UNEP environmental management tools Life-Cycle Assessment

http://www.uneptie.org/pc/pc/tools/lca.htm
European Environment Agencys guide to approaches, experiences and information sources of LCA

http://reports.eea.eu.int/GH-07-97-595-EN-C/en

96

introduction to life cycle assessment 5