| Front page | | Contents | | Previous | | Next |

Prioritisation within The Integrated Product Policy

7 Database development

7.1 Introduction
7.2 Adding physical units
7.3 Further disaggregation
7.4 User's guide to the LCA database
7.5 How to use the database for hybrid LCA
     7.5.1 Tiered hybrid analysis
     7.5.2 Embedded hybrid analysis
7.6 Prioritising future data collection for the LCA database

7.1 Introduction

The database developed in this project provides a set of background data for lifecycle assessment of products used and/or produced in Denmark. The IO-based background data can be used to fill gaps in LCAs where specific process data are missing, and at the same time provide a basis for prioritising future data collection.

The database is provided in formats compatible with the EDIP LCA-database administrated by the Danish LCA Centre (i.e. SimaPro and GaBi data formats). In the GaBi version, only terminated data per product group is provided, since the GaBi software does not yet handle the endless loops which are an inherent feature of IO-data. However, GaBi-users may analyse the full data in the free demo version of SimaPro, which can be downloaded at www.simapro.com/simapro.

The database is provided both as an attributional version where all the links between industries contribute proportionally to the result and a market-adjusted version where market constraints are taken into account as described in Chapter 2.9.

Requests for access to the database should be directed to the Danish LCA Centre <info@lca-center.dk>.

This chapter has been prepared with contributions from Niels Frees from the Danish LCA Centre, who reviewed the database and suggested improvements for the user-interface.

7.2 Adding physical units

The supply-use tables (Danmarks Statistik 2003b) provide data as to which commodities are produced by each industry (unfortunately only in Danish language). For some of these commodities, also data in physical units are provided. For each industry and final consumption group, the 15 most important commodities (from an economic perspective) have been placed in the comment field of the database (see Chapter 7.4). For the industries, the calculated basic price per physical unit has been added, when available.

The information can be applied in two ways:

• As an additional information to the name of the industry or final consumption group, to check that the right industry or group is chosen for a specific commodity ^[9].
• To estimate the monetary value that should be applied for a given physical amount of product (see also Chapter 7.4).

The latter application is complicated by the large number of products typically produced by each industry, out of which physical data are only provided for some. Also, physical data does not necessarily mean mass, but could also be “pieces” (e.g. of footwear), “pairs” (e.g. of trousers), “m²” (e.g. of glass plate), etc. This makes it impossible to provide an overall price/mass relationship for most industries. A further complication is that mass balances or similar quality control techniques have not been applied to the data in physical units, which means that there may be undiscovered errors in the price information provided.

In most cases it is possible to estimate the missing mass flows from the price/mass relationships of similar imported or exported commodities, as we have done to construct the mass balance for the textile industry (see Chapter 6.4). However, this is a cumbersome task, which could not be performed for all industries within the limits of the current project.

Thus, to find the monetary value that should be applied to the output from an industry for a given quantity of a specific commodity (for which the price is not already provided in the comment field), we recommend using the price/mass relationship given in the export commodity statistics (www.statistikbanken.dk; look under “External trade” for Table FORS2: “Supply of goods, by BEC (Broad economic categories), trade/production and quantity/value”).

Note that the prices shown for industry outputs are basic prices, i.e. without taxes and wholesale and retail profits.

7.3 Further disaggregation

Compared to the database that was used for the prioritisation presented in Chapter 1, the published database has been further disaggregated on a number of points:

• The consumption group Non-durable household goods n.e.c. has been further disaggregated into 10 product groups: Brooms and brushes, Matches, Carbondioxide cartridges, Metal articles n.e.c., Paper articles n.e.c., Pesticides, Plastic articles n.e.c., Polishes, Solvents, and Textile articles n.e.c.
• Domestic cooling equipment has been split out from Domestic appliances n.e.c. as reported in Chapter 6.3.
• The foreign textile and apparel industries have been further subdivided to distinguish different fibre types, as reported in Chapter 6.4.
• The Danish industries Textiles and Wearing apparel have been subdivided as reported in Chapter 6.4.
• Service activities n.e.c. has been subdivided into Laundries and dry cleaners, Hairdressers and other beauty shops, and Funeral services, as reported in Chapter 6.4
• Fringe benefits in the form of free PC and free car has been moved from being a commodity output from the employing industry to be an output of the industries supplying these goods. For the supplying industry, this means that the fringe benefit parts of the supplies to industries are now recorded as supplies directly to private consumption. Fringe benefits in the form of canteens, free housing and free airfares are already in the original NAMEA supplied from the industries producing these commodities directly to the corresponding final consumption groups.
• Service (labour) outputs of commodity producing industries (“Lønarbejde” in the commodity statistics) have been eliminated (as labour does not carry any environmental impact) and are instead recorded as direct wages of the service-receiving industries. The reduced output of the service-supplying industries (which is matched by an equivalent reduction in their wage expenditure) implies that the environmental impact intensity increases for the remaining commodity outputs of these industries.
• Recycling of jewellery has been separated from the consumption group Jewellery, clocks and watches.
• In the market-adjusted version of the database, recycling of waste and scrap is subdivided into separate recycling processes for each material type. Each recycling process is remodelled to supply a recycling service to the scrap supplying industries. In this way, emissions of the supplying industries are no longer assigned to scrap as a commodity, but rather the opposite: the emissions of the recycling industries are assigned to the scrap supplying industries. In return, the new recycling processes provide emission credits to the supplying industries equal to the value of the supplied scrap, which is assumed to reflect the amount of primary material that is replaced by the supplied scrap. The remodelling reduces the turnover of the supplying industries by the original value of the traded scrap, which implies that their emission intensities increase. This is a reflection of the new situation where emissions are no longer assigned to scrap as a commodity. The remodelling also means that the value originally paid by industries receiving scrap is moved from the recycling industry to the industries supplying the corresponding virgin material. Therefore, the emission intensities of the scrap receiving industries also increase.

These improvements were chosen among a long list of possible improvements, based on two criteria:

• Issues that were identified as important during the work with the database when performing the prioritisation presented in Chapter 1 and the more detailed applications presented in Chapter 6.
• Product groups where our uncertainty assessment reveiled the largest absolute data uncertainty.

We also sought to find industries where the existing process-based LCA-data in the EDIP database could be used to disaggregate the IO-based data. However, the areas where the EDIP database has large detail does not in general match the areas identified as having priority from the above criteria.

7.4 User's guide to the LCA database

This sub-chapter gives a short introduction to the database as available in SimaPro. For further guidance on how to use the functionalities in the SimaPro software, please consult the SimaPro user manual, which is included in the free demo version.

Figure 7.1 below shows the main menu of the project opened in SimaPro. In the LCA explorer, processes are chosen in the left menu. The Danish input-output data for industry outputs (i.e. cradle to gate) are placed under Material – Input Output – Danish production and the data for final consumption (i.e. including wholesale, retail and use stages) under Use – Input Output – Danish consumption.

Click here to see Figure 7.1.

Figure 7.1. The main menu of the LCA database in SimaPro

Under Danish production, the following sub-menus appear:

• Analyzing DK contains the overall data on production and consumption as applied in the prioritisation, in total and per DKK.
• Danish constrained production contains the constrained processes, which were included in the prioritisation (see Chapter 2.9), but which are of little relevance when using the database for life cycle assessments.
• Foreign production contains the terminated processes used for imported products (see Chapter 2.8).

Under Danish consumption, the following sub-menus occur:

• grouped as needs contain the data corresponding to the need groups from Chapter 1.2.4,
• of which domestic production contain the purchases from domestic production as well as all the retail trade and most of the wholesale trade, even for products imported for final use,
• of which imported production contain the final use purchases from foreign producers.

In the right menu in Figure 7.1, the process Advertising from Danish production is chosen. In the bottom, below the list, a comment field shows the commodities included in the chosen process, and the price information when available.

The attributional version of the database (i.e. without the market adjustments of Chapter 2.9) is provided as a separate project in the database, with the same structure as explained above.

SimaPro (also in the free demo version) offers several different ways to analyse a product system and its environmental impact. Using the Calculate Network option (F10) the links between the selected process and the other processes is provided in a graphical form, see Figure 7.2.

Click here to see Figure 7.2.

Figure 7.2. Example of network (with cut-off 3.2%) showing important processes contributing to the 140 g CO₂-equivalents from the lifecycle of 1 DKK of products from the industry “Beverage, DK” (accumulated emission values in the lower left of each box).

It is possible to cut off minor contributing processes in the graphical presentation of the network. For example, in Figure 7.2, a cut-off of 3.2% was selected to limit the number of processes shown in the diagram. The cut-off relates only to what processes are shown in the diagram, not to cut-off's in the calculation (i.e. the underlying calculation still includes all contributing processes).

Figure 7.2 shows that to obtain a net-production of 1 DKK beverage, the actual (gross) production is in fact 1.08 DKK. Similar results can be found for other processes. The explanation is that not only the required output (1 DKK worth of beverage) is produced, but also beverage used in other processes of the upstream network. These feedback loops are shown by the arrows leaving at the top of the beverage industry feeding back into itself and other processes in the network.

The environmentally most important inputs to a process can be found by following the thickest streams of Figure 7.2. In this example, Grain farms are dominant, but also electricity and wholesale trade have large contributions to global warming. Other impact categories may of course be chosen, giving further information.

Already from a diagram as Figure 7.2 it can be seen e.g. that energy processes are not very costly compared to their environmental burden; itself an interesting observation. In the example of Figure 7.2, the electricity contributes with 10 % of the Global Warming Potential, but represents only 1 % of the economical cost of the product.

Having selecting several processes in the SimaPro LCA explorer, the Calculate Compare option (F9) allows comparative presentations as already presented in earlier Chapters of this report (e.g. Figure 1.3. comparing the 11 need groups, and all the figures in Chapter 6). The contribution from individual processes within the product systems can be shown as in Tables 6.3 and 6.4. The results from these analysis options can be presented both as tables and figures and are of course also available for a single product.

When using the data, it should be remembered that thay have been derived by following the monetary flows from buyer to seller. This means that services provided for free will not be included, even when they have environmental relevance. For example, the data for “Freight transport by road, DK” and “Car purchase and driving…” have inputs of vehicles, fuels, maintenance, accessories, and road and bridge tolls, but do not include the general road infrastructure since this is provided “for free” by the society, which means that it is placed under “Public infrastructure” drawing mainly on the industry “Civil engineering” (Anlægsvirksomhed).

7.5 How to use the database for hybrid LCA

There are several ways in which the IO-database may be combined with more specific process data. The two main approaches are:

• Tiered hybrid analysis
• Embedded hybrid analysis

The name hybrid analysis refers to the combination of process-based LCA and Environmental IOA.

7.5.1 Tiered hybrid analysis

The typical (and simple) application of IO-data in a process LCA is to start from one or more specific processes that are better documented in terms of emissions and inputs than the corresponding industries in the IO data. To this “foreground” kernel, the IO data are simply added, linking each input to the process-based system with the corresponding best fitting final use group or industry output in the IO-database. Downstream processes like recycling or waste treatment may also be added. In this way, the IO data are used to complete the upstream and downstream parts of the product system not covered by specific process data.

A very simple hybrid application starts from one single foreground process (for example a specific industry site), identifies in the IO-database the final use group or industry output that best covers this process, makes a copy of this IO-based process, and use the more precise data of the foreground process to replace the less precise data in the IO-based process (leaving the IO data as a proxy for those parts of the foreground data which are not adequate or complete). The resulting hybrid process may then be used in a direct comparison (benchmarking) with the original IO-based process, or it may be used in further modelling of a more complex foreground system.

The advantage of this approach is that it is simple. The drawback is that there are no links back from the IO data to the foreground processes, i.e the upstream IO data do not take advantage of the added information available in the foreground processes. This also means that knowledge does not accumulate in the database, i.e. when applying the IO database for another foreground system, the added information from first foreground system is not automatically linked into the new product system.

This is the main reason for the development of embedded hybrid analysis (see Chapter 7.5.2), where these drawbacks are overcome. However, embedded hybrid analysis is more demanding and therefore appeals more to the advanced user that wishes to make several LCAs while continuously improving the underlying database.

7.5.2 Embedded hybrid analysis

This more advanced hybrid approach utilises the common matrix nature of process-based and IO-based data, by embedding the process-based data in the IO-matrix. This is the approach used when subdividing industries in the IO-database as described in Chapter 2.6 and 7.3.

The first step of this approach is identical to that of the tiered approach (Chapter 7.5.1): The starting point is an identification of the IO-database the final use group or industry output that best covers the process for which more specific data are available. A copy of this IO-based process is then made, and the more precise data of the foreground process is used to replace the less precise data in the IO-based process.

Two additional steps are then needed to embed the new process in the IO-database:

• The original IO-based process is modified by subtracting the inputs, outputs and emissions now represented by the new hybrid process. In order to do this, the relative production volume of the two processes needs to be known. The production volumes for the processes in the IO-database can be found as the inputs to the processes under Analyzing DK (see Chapter 7.4).
• The output of the new hybrid process is linked as inputs to all the processes that it supplies. This can be the same processes and proportions as for the original IO-based process, or it can be a different distribution when the specific process is supplying a specific segment of the market. The original supplies from the IO-based process are reduced with the amounts now supplied by the new hybrid process.

Both these steps, but especially the latter, are rather cumbersome if performed within SimaPro, since every input and output needs to be accessed separately. For the advanced user, it is therefore preferable to perform these operations in the original matrix structure, e.g. in a spreadsheet software, utilising the advantage that operations in a spreadsheet can be performed on entire rows and columns. The entire database can be exported to Excel with the “Export to matrix” function of SimaPro. The two embedding steps may then be performed by adding a row and column representing the new hybrid process, performing the additions and subtractions described above, and re-import the adjusted matrix into SimaPro or any other matrix calculation tool. Import of matrices to SimaPro is performed via a CSV-file, which can e.g. be generated by a macro in Excel.

The advantage of the embedded approach to hybrid LCA is that the adjustments made will automatically be available for all future applications of the database. It is therefore the approach preferred by database developers and advanced users that perform several LCAs using the same underlying database.

7.6 Prioritising future data collection for the LCA database

The underlying IO data from Statistics Denmark as well as many of the emissions data are published on an annual basis. Thus, it would be possible to update the database annually. The delay in availability of statistical data already imply that consistent data sets are at least 4 years old when published, which could be an argument for annual updates. However, due to the relatively large amount of work involved in performing all the necessary adjustments described in Chapter 2, a less frequent update could be considered. An update at least every 5 years should be considered imperative due to the developments in technology and consumption patterns.

It should also be considered that the costs of regular updating of the entire database could instead be used to improve the detail of the database, both by subdividing industries based on more detailed statistics and other datasources (as described in Chapters 2.6 and 7.3) and by adding more emissions or environmental impact categories (as suggested for toxicity in Chapter 2.5.1 and for other impact categories in Chapter 2.10.5).

An uncertainty assessment performed on the database may be used as a guiding tool to determine the most cost-effective way of maintaining the database. That maintenance action should be chosen which give the largest reduction the overall uncertainty of the results form using the database.

As mentioned in Chapter 2.11, one of the most important causes of uncertainty in the database is the high aggregation level of industries. The industries may therefore be ranked according to the absolute uncertainty with which their environmental impacts are determined, thus providing a prioritised list of the industries where a further disaggregation could reduce the overall uncertainty the most.

For the database used in the prioritisation (i.e. prior to the improvements described in this chapter) we obtained such a list, the upper part of which is shown in Table 7.1.

The uncertainty data shown in Table 7.1 include the uncertainty from using modified US American data to represent all imports to Denmark. Thus, on an individual industry level, this geographical aspect of data quality is included in the suggested prioritisation in Table 7.1. However, at a more general level, it should be considered how the work on the Danish NAMEA could be embedded within a Global NAMEA, e.g. through the database network initiated by the UNEP Life Cycle Initiative (www.uneptie.org/pc/sustain/lcinitiative).

Neither age of data nor the importance of missing environmental exchanges are included in the uncertainty assessment described above. For an overall data collection strategy, these two aspects of uncertainty need to be included, e.g. by estimating the variation of IO and emission data in NAMEAs from a number of years and estimating the bias caused by environmental exchanges not currently included in the database.

Table 7.1. Absolute uncertainty (standard deviation) of the overall environmental impact (expressed in person-equivalents) from different product groups, calculated before the database improvements reported in this chapter.

Once these two additional estimates are available, it will be possible to judge whether uncertainty is reduced most by more frequent updating of the database, by disaggregating specific industries (including improving the representativeness of foreign industries), or by adding more environmental exchanges.

In practice, other concerns than uncertainty reduction may guide the data collection. Typically, funding will be available for data-generating projects within specific industries, and such data should of course be integrated in the database as quickly as they become available, disregarding the position of that industry in the overall prioritisation suggested by the uncertainty assessment.

Footnotes

[9] Note that the same commodity may be produced by several industries. For example, the commodity “Plastic sheets n.e.c.” (“Plader, ark, film mv a plast ian”) is mainly coming out of the industry “Rubber products, plastic packing etc.” but a smaller amount is also coming from “Basic non-ferrous metals”, which may at first sight look like an error. However, such occurrences can often be explained as the re-sale of surplus purchase of raw materials, in the example given, it could be surplus plastic originally purchased to laminate aluminium articles.

| Front page | | Contents | | Previous | | Next | | Top |

Version 1.0 February 2005, © Danish Environmental Protection Agency