7 Sensitivity of data
The previous chapters have analysed the weight induced energy savings of 100 kg weight reduced vehicle examples. Vehicle specifications have been defined and discussed for exemplification. Apart from the vehicles themselves, especially the use of vehicles (driving pattern and annual performance) can be very different, depending on the respective countries and uses. These differences are also influenced by the stan- dard of infrastructure and exist therefore especially between industrialised and develop- ing countries. The existence or absence of motorways and well paved roads and the general level of traffic have an influence on the driving pattern as well as the annual per- formance. These uncertainties significantly affect the results in respect to weight induced energy savings.
Though there are no hard data for a global average of vehicles and their use, this report has tried to identify typical vehicles and usages. These vehicles may be close to the average situation at least for the industrialised countries of Europe and North America, where the most vehicles exist and good data are available. To estimate the discrepan- cies between the defined typical vehicles and a supposed average vehicle an uncer- tainty for the parameters is estimated.
The life-time energy and CO2 savings depend on two parameters of equal impor- tance, but with different uncertainties. Both parameters will be increased and de- creased by a fixed percentage. The sensitivity to these uncertainties will be calculated based on the assumption of a standardised normal probability distribution of the uncer- tainties. The calculation thus follows the Gauß formula c = (a2 + b2) for c = a · b.
The sensitivity of life-time energy and CO2 savings have been calculated from the un- certainties as described above. Since scientific chargeable data exist mainly for pas- senger cars from tests as well as simulations, the uncertainty of specific energy sav- ings is estimated to be around 10 % for passenger cars. For all other vehicles 20 % un- certainty is assumed. For the life-time performance of passenger cars 15 % and for all other vehicles 20 % uncertainty are estimated (Tab. 29).