Information Technology Reference
In-Depth Information
The simulation component comprises two main models: (1) a demand-side model
implemented as a travel request generator and users' behaviour simulator, and (2) a
supply-side simulator that simulates the functioning DRT services, including the deci-
sions made by the ODSS and the vehicles operations.
Both of these models are based on a micro-simulation event-driven approach. The
main demand-side events are: trip reservation, trip cancelation, user arrival to stops
(origin and destination), user no-show and non-reserved shows at stops. The main
supply-side events are: trip planning events (such as hourly and daily time-scheduled
planning of advanced reservations, real-time acceptances or rejections and re-
planning), vehicle departure and arrival from/to stations, stops and strategic waiting
and interface points.
Travel requests are generated based on socio-economic characteristics of the resi-
dent population (from Census), from domiciliary questionnaires and local authorities
interviews, as well as acquired knowledge about the main attracting poles for visiting
(workplaces, schools, hospitals, markets, general services, transfer stops to inter-
urban transport services, etc.).
The analyst component is used to properly analyse and evaluate DRT options. It is
powered by a BI type framework that starts to extract relevant data from the historic
main database, transform such data and load it into a proper database system for
multi-specialized analyses. It comprises different methodologies: simple and ad-
vanced statistical reporting and inference techniques, data mining and operational
research inference and prospective models.
Thus, it is important to incorporate reliable predictive systems capable of evaluat-
ing beforehand the impact of alternative or small changes in strategic or tactical plans.
This can be done by integrating simulation tools whose goal is to inspect the behav-
iour of a complex system under different scenarios. This referred framework is used
to determine the system performance to evaluate the alternative specifications for a
certain area. It allows choosing the better specifications and the better working rules,
as illustrated in Figure 4. This figure shows how the analytic component of the system
generates new alternatives to be simulated and how their performance is compared
with the previous solutions already tried.
An iterative approach is used between the simulator and the analyst components.
At each iteration, a solution option is simulated and evaluated by performing an as-
sessment process. And, for each iteration (option) a large set of simulation runs (days)
are realized and their operational details are stored into the main database of the op-
erational decision support system (ODSS). The number of days is set in order to infer
statistics (e.g., mean values) within a given high level of accuracy.
The result of the assessment process will provide guidelines and required feedback
to adjust system resources and operating parameters, as illustrated by feedback arrows
in Figure 3. Additional analysis is performed to assess solution viability and
sustainability, encompassing several dimensions such as: social, economic and
environmental.
