Infrastructures and Transport

Smart Mobility Analytic – SMA

Home » Solutions » Infrastructures and transport » Smart Mobility Analytic – SMA

Effective management of passenger and vehicle flows through an artificial intelligence model that reduces queues, waiting times and congestion in airport environments

The Smart Mobility Analytics (SMA) solution uses artificial intelligence systems applied to image analysis analysis to measure and manage the mobility of a cruise terminal. It can be applied to other equivalent airport and transport environments where there are many people who need to move to another location.

Those responsible for these areas are soon faced with a space with large surfaces and a large influx of people and means of transport. The aim of the SMA is to efficiently manage the mobility of passengers and vehicles during embarkation and disembarkation operations. In this way, passengers can be transported to the place where they want to go and the means of transport can be located where this demand is concentrated.

Delonia’s solution provides mobility actors, passengers and carriers with real-time information on mobility-related parameters, enabling better decisions to be made by those involved. In addition, high efficiency in the demand of all parties is achieved without having to use a large number of cameras thanks to the neural network-based algorithms used in the system.




The solution is easily scalable and can be moved to other physical environments and integrated into different IT infrastructures.


It reduces waiting times for people and vehicles at the terminal.


The equilibrium point between transport supply and demand is achieved in less time.



It provides real-time qualitative information to all members of the mobility chain.


Transport vehicles arrive when they are needed and in the right quantity.


Improve the experience of tourists and transport providers in transit through the port.


Make transport demand forecasts adjusted to each landing of passengers and goods.


Passengers decide which type of transportation to use based on the waiting times for each queue.


It reduces the carbon footprint in the space where the solution is used by optimising mobility.


[ba_image_magnifier image_small=”” image_large=”” mg_height=”200px” mg_width=”200px” _builder_version=”4.18.0″ _module_preset=”default” global_colors_info=”{}”][/ba_image_magnifier]

Transport and mobility facilities where the arrival and departure of passengers and goods brings together many mobility agents (people and vehicles) in a small space and in a short time:

  • Ports
  • Airports
  • Railway stations and bus stations
  • City Transport Stations


Video analysis

The images are obtained by means of a system of cameras, with a reduced and optimised number, which triangulates the analysed space, so that the system can analyse the information and transmit it to all the actors in the value chain.

Detection algorithm

With the images obtained and the identification of the elements, the system allows a precise measurement of the flow of people and vehicles (buses and taxis). Detection algorithms based on neural networks offer great precision in processing information.

Mobile app

Through a mobile application, field personnel know the situation of each area without having to travel to it. This makes it possible to assess each situation telematically and without having to waste time travelling.

Identification systems

The system identifies people, vehicles and number plates, as well as monitoring the tracking of both means of transport and humans. Eliminate redundant measures such as passers-by crossing from one place to another.

Managers panel

The terminal managers, through panels, know at all times the status of the queues of passengers and vehicles. In order to take the actions they consider appropriate when resolving each situation.

Forecast model

It has a forecasting model to anticipate future situations based on the information obtained and the history stored in the system. In addition, simulations can be carried out to refine the forecasting model.