Environmental Engineering Reference
In-Depth Information
Then, an iterative procedure searches an operation within the radius thus
defined whose characteristics are compatible with those of the transaction to which
it is coupled. For example, a shop downtown, served by a professional carrier
using a vehicle less than 3.5 tons on a round of 10-20 stops, will be linked to
another delivery of similar characteristics (carried out by a professional carrier
using the same type of vehicle taking a route of the same size category) and
located within a range of distance imposed by the corresponding function. The O-
D matrix deduced takes the following form:
T ij ¼ t ij ¼ X
v
t ij ð v Þ
where t ij is the overall number of TL trips (outbound and return) and the overall
number of approaches, return and connection trips for LTL rounds between z i and
z j . Gradually, all the transactions are matched to build an O-D matrix including all
areas, and the number of operations detailed according to their characteristics.
The O-D matrix is then broken down by time of day in order to distinguish traffic
problems at morning peak, evening peak or off-peak hours. Finally, this O-D
matrix can be linked to a traffic assignment model to include urban goods
movements in car and public transport traffic and make an overall diagnosis of
urban transport (Nicolas 2010 ).
5 ECM Estimation
The ECM estimation module is developed in a similar way to the IEM module,
according to a two-step procedure of movement generation and distance calcula-
tion on an inferential statistics basis (Gonzalez-Feliu et al. 2012a ). However, the
data required were not collected in parallel to the development of the method;
instead a standard personal trip survey was used. Indeed, although there is no
database of pickup and delivery trips specific to urban areas in France, a standard
database can be consulted for shopping trips, namely the National Urban Personal
Trip Survey database. It can be used to characterize several cities in terms of
shopping trip behaviour. Nevertheless, not all the cities present the same data
granularity (for instance, the city of Lille has detailed data from Monday to Friday
including the nature of purchased goods, and aggregated data for Saturday trips.
Lyon has data from Monday to Saturday some years, but the most recent survey
covered only trips from Monday to Friday and only the type of shop is recorded,
but not the nature of purchased goods). In any case, such surveys can form the
foundation for model construction, since specific surveys should be seen as
redundant by public authorities and their costs can be higher than those of the data
collection campaigns of existing databases.
Search WWH ::




Custom Search