|
|
|
A5. Industrial networks
Probably the simplest way to represent the components of an extended system is to draw a flow chart showing how the sub-systems are inter-linked. Some care is needed in the construction of these flow charts.
Figure 3(a), for example, shows a simple linear sequence of operations that might be used to represent the sub-systems employed to produce and deliver electricity. While indicating the general form of the system it is an over-simplification of reality.
Figure 3.
Linear system and simple network used to describe electricity production. See text for
fuller explanation.
Figure 3(b) is closer to actual practice and illustrates how the simple linear sequence of Figure 3(a) must be replaced by a more complex network of flows. Figure 3(b) is still a simplification but it demonstrates the essential point that it is impossible to calculate
the energy to produce electricity, for example, without first having values for the energy to produce coal and steel. But it is impossible to calculate the energy required to produce coal and steel without first having a value for the energy to produce electricity. The same problem applies to all other inputs and outputs.
This inter-dependence of energies, and other parameters, can be overcome by iteration. That is, when examining energy, a value is assigned to the electricity production energy and this is used to calculate values for the energy requirements of steel and coal production. The calculated values for steel and coal are then used to produce a revised value for electricity production which is in turn used to revise the steel and coal figures. This cyclic process is repeated until the changes produced by the further iterations fall below the level of accuracy desired. How quickly the values converge will depend upon how close the chosen starting value is to the correct value. A similar procedure can be used for the calculation of the other parameters.
In much of the earlier work reported in the literature, especially in the early days when most of the calculations were carried out manually, iteration was seldom employed Instead, production systems were usually simplified into linear sequences so that the network problem did not arise. Such simplifications can give rise to significant errors.
A6. The nature of extended systems
The systems employed in all LCI calculations trace the production of all materials and fuels back to their extraction from the earth. An extended system therefore consists of three main groups of operations as shown in Figure 4. These are the main production sequence, the production of ancillary materials and the fuel production industries.
Figure 4.
Three main groups of operations making up an LCI system
The main production sequence is usually the easiest to identify. For example, suppose that the system is defined as the use of polyethylene bottles to pack milk. The main operations that would need to be included in this production sequence are as shown in Figure 5. For simplicity, some of the transport operations that are needed within this production sequence have been omitted from Figure 5. This should not be interpreted as meaning that such operations are unimportant; frequently the inputs and outputs from transport operations dominate the final result. In most published work there is seldom any difficulty identifying the main production sequence since the operations to be included are usually self-evident. Frequently however, transport operations are omitted.
The identification of ancillary materials such as pallets, labels, bags, glue, etc. is usually easy since this information is collected along with the data for the unit operations in the main production sequence. In a proper analysis, the production sequence for each of these ancillary materials must also be traced back to the extraction of raw materials from the earth in the same way as for the bottle system shown in Figure 5.
Figure 5.
Example of the main production sequence that would be needed for an LCI of a plastic
bottle system.
It is clear therefore that the analysis of even simple systems frequently demands data from a wide variety of different industries, some of which are remote from the main production sequence. It is this which is one of the most time consuming elements of complete analyses. Some analysts attempt to take a short cut by saying that because the mass input of some of these ancillary materials is small, their contribution to the overall system is negligible. In some instances this may be true but it is important to demonstrate this rather than simply assuming it to be so.
The amount of detail included for these ancillary materials is one of the commonest causes of differences between different analyses of apparently the same system. Frequently the information given about what is included or excluded is inadequate so that the validity of the final result is difficult to judge.
|
|
|