BASIS OF REPORTING
The accuracy of the data depends on the method of measurement, the calculation
procedure and whether estimates have been used. Sampling method and frequency can
also affect accuracy.
Because Heineken did not start recording accidents involving contractors until
2000, estimates had to be made for 1999. In the case of some multi-purpose production
units (for example, breweries incorporating soft-drink plants), total figures for such
parameters as energy consumption and waste water volume have been allocated on the
basis of estimates. Because the COD, nitrogen and phosphorus contents and concen
tration of suspended solids are not measured or are not measured at the same frequency
by all production units, they have had to be estimated in many cases. The quantity of
refrigerant is difficult to establish because it is used in dynamic systems in which it can
occur in both the liquid and gaseous phases. Refrigerant losses are determined on the
basis of the quantities added to replenish systems, which is less accurate than direct
measurement of actual losses (systems are not topped-up at regular intervals and may
be replenished one year and not another, giving rise to variations in the reported data).
Where (H)CFCs are used in air-conditioning systems in non-production departments,
losses have in some cases been allocated to production. The volume of residual waste
generated by production units outside Europe is estimated in many cases.
At a number of production units, waste is removed from the site in containers of a given
volume, and inaccuracies can arise in translating volume to weight. In the absence of local
legislation in some countries outside Europe, the definition of hazardous waste is not
always clear. In some cases, hazardous waste is safely recycled and is no longer desig
nated as hazardous.
After each reporting round, the safety and environmental data, measured as accurately
as possible, are aggregated and presented in the SHE report. Where material errors have
been found in the 2000 data, the aggregated data have not been revised. Where such
errors affect trends, this is mentioned in the text or data table. The comparability of the
data depends on the extent to which estimates have been used in determining the
performance indicators. Where estimates have been used in interpreting trends, this is
stated in Chapter 6.