Introduction
A product system can be modeled as a complex sequence of unitary operations that communicate with each other and with the environment through input and output flows. Inventory analysis is the stage during which an analogical model of reality is constructed, capable of approximating as faithfully as possible all potential exchanges between the individual process units belonging to the actual productive (and destructive) chain. No evaluations or judgments are made concerning the meaning of the different inputs and outputs, that is, the environmental effects they may cause: the objective of an inventory is, in fact, to provide objective data, which can only be processed and commented on later in order to draw evaluations and useful indications at the decision-making level (Life Cycle Impact Assessment and Life Cycle Interpretation Phase).
An LCA study inventory must also guarantee reliability; to this end, its drafting must be carried out following a very precise and defined procedure. Only in this way can the results of different inventories be compared. The ISO standards of the 14040 series and, in particular, ISO 14041 provide this code, making the drafting of a life cycle inventory less subjective today than it was in the past.
Information Collection Methods
Assessing the reliability of the data collected during the inventory phase is an important preparatory stage, which also allows for saving precious time in subsequent phases. The evaluation of data quality can be diagrammed through an iterative process in which the collected data are verified, validated, and modified as necessary. The first step is to create a detailed flowchart of the operations that contribute to forming the analyzed system. This diagram will help the technician analyze all the various process units without risking overlooking any of them. For the most common production processes, most of the detailed information can be drawn from technical literature, which is sufficient to achieve the initial goal, i.e., to "get a hold" of the fundamental process units in order to later discuss them effectively with plant operators. At this stage, at least the following two objectives must be set:
- Determine the level of detail the company can provide evidence for, as this will determine the level of detail to which the entire analysis can be pushed;
- Verify that the prepared flowchart corresponds to the real process.
For example, if we want to analyze an aluminum can production plant in order to evaluate energy consumption, it would be pointless to divide the process into individual components if there is only one electricity meter inside the facility. Conversely, many valuable pieces of information would be lost if the production cycle analysis were considered as a single system where aluminum enters and cans exit. Since it is a model of a real system, the flowchart will necessarily be an approximate representation of this system. The main problem then lies in obtaining the most faithful and effective approximation. Its quality will not depend so much on the completeness of the representation, but on the presence of all process units deemed significant. In order to make the overall system more identifiable and easier to interpret, often voluntary exclusions of subsystems are necessary, such as those related to minor components with negligible flows of materials or energy.
Noteworthy are the supplies of centralized services, common to many process industries. This is the case, for example, with the cogeneration of electricity and steam, which is often attributed to a separate plant from which many process units draw, or in a completely similar way, that of the so-called "general services" (overheads) such as compressed air production or wastewater treatment. Obtaining accurate and reliable data for such ancillary activities is as important as for the main process. The final flowchart will be used as a basis for questionnaires that will be subsequently distributed to the various competent company figures for data and information collection.
The data to be used for building the inventory should, as much as possible, come from field measurements, hence the term "primary data" (site data). If it is not possible to collect data directly, it is necessary to use derived or secondary data, i.e., data obtained from literature sources (e.g. scientific articles) or specially prepared databases (e.g. EcoInvent). The distinction between primary and secondary data is actually only apparent, as most secondary data are in turn derived from field measurements. A typical example of secondary data includes those relating to the energy mix of a country (Residual Mix) where the analysis is conducted, emissions produced by a truck used for transportation, and so on. When using secondary data, it is always advisable to check their source, publication date (using data collected more than a decade ago is not recommended), and in any case to compare them, if possible, with those from similar publications. Finally, regarding the information reference period, it is advisable to collect data for at least 12 months of activity, a period that usually coincides with the company's financial statement reference period.
