184 PART | V The future of ITS applications
fleet owners to make informed decisions concerning containers and their con-
tents (Sharma et al., 2016; Mahmood et al., 2019) to prevent delays and search
for alternative transportation paths in case of emergency. The same principles
apply in any transportation network and rely on the evaluation of flow over a
static network, with given loads and capacities, based on historic information,
and real-time updates based on information collected by the vehicles that move
on the network and from sensors at various network points (e.g., ports, stations,
tolls, etc.). The freight transportation is even more interesting within urban envi-
ronments, where the information is denser and updates more frequently and at
the same time the fluctuation in traffic conditions, congestions, and other events
that introduce delays in the flow are more frequent (Hesse, 2016). A generic
model behind product-transportation networks is depicted in Fig. 16.4. The net-
work model is created based on historic traffic information and is always evalu-
ated using actual network usage data. Using real-time traffic data and a path/
route recommendation model it is possible to redirect transports and optimize
the network performance.
16.2.2.2 Network traffic and safety management
The ability to dynamically collect traffic data from a transportation network,
for example, an urban-road network, an airport network, or a port allows net-
work operators to make informed decisions or respond fast in case of emer-
gency. For example, an accident can be detected in the road network when the
average moving speed of vehicles in a section is much slower than usual, traf-
fic congestion can be detected in a network node (e.g., a crossroad) when the
incoming flows are much larger than the outgoing ones. In order to respond to
such situations, network operators can reschedule the traffic lights of a junc-
tion depending on current traffic conditions. They can also combine this action
with a traffic rerouting, as it can be pointed by a routing optimization algorithm
(Cao et al., 2017).
FIGURE 16.4 A generic transportation network model.