Estimated Data Production for Urban Goods Transport Diagnosis - Sustainable Urban Logistics: Concepts, Methods and Information Systems - page 123

Weighting of driver trips:
2 biases were found in the driver sample:

• For each management mode, there was a distortion between the number of
movements generated by the different types of activities of establishments
delivered by the drivers and the number of movements generated by all the
movement slips described in the establishment sample.
• Drivers working for the firm surveyed answered more frequently than third party
drivers: the rate of questionnaires sent back differed according to the closeness
of the driver and the establishment delivered.

In order to amend these biases, the following weighting of a round of the type
(m,a) was used:

r tour m ; a ¼ Nbmv m ; a = nbmv round m ; a

is the weight (in weighted number of rounds).
Where:

• a: branch of activity (industry, craftsman, wholesale, retail, large store, tertiary
services, warehouses; agriculture);
• m: is the management mode in five classes: own account carried out by the
establishment as consignee, own account not carried out by the establishment as
consignee, own account carried out by the establishment as forwarder, own
account not carried out by the establishment as forwarder, third party (carriers).

Nbmv m ; a : the weighted number of movements calculated in the urban area for
the (m,a) establishments, nbmv round m ; a is the total number of stops of the (m,a)
drivers' trips in the driver sample.
If nbstop: number of stops of a round for a driver of type (m,a), the weight (in
number of movements) of this driver is:

nbmvdriver m ; a ¼ nbstop r tour m ; a

The size of the establishment sample and the quality of the responses guaranteed the
correct estimation of the average number of movements generated by each activity
stratum. The number of no-answers was high regarding the weight of the payload
(40 % of NA) and the origin-destination of the goods. The number of movements
obtained by computing the sum of the frequencies of movements in the data sheets was
20 % lower than the actual total described by the establishment interviewed.
Based on the pertinence of the weighting the expansion of these surveys made it
possible to estimate several mean ratios (Patier 2001 ; Patier and Routhier 2009a , b ;
Routhier 2013 ). According to the city, the share of urban goods traffic (UGT) in
the total traffic (in vehicles*km car units) varies from 9 to 15 % of trips, 13-20 %
of vehicles*km, 15-25 % of vehicles*km in terms of road occupancy rates.
One of the main contributions of the surveys consisted in providing knowledge

on urban management rules (the links between activities, operating mode,