Skip to main content



Reporting period: 2018-05-01 to 2019-04-30

Public transport is crucial to the liveability of any city. It plays a key role in the reduction of congestion and pollution and in the increase and improvement of mobility.
Most transport operators depend on ticket revenue to sustain their operations. This pay-per-use method is commonly guaranteed with the usage of travel tickets that are canceled or validated when accessing the transport network. However, some passengers do not buy or validate a travel ticket. This practice, known as fare-dodging or fare evasion, is a serious problem in many public transport systems around the world. Not only does fare evasion have a negative impact on operators’ economic sustainability, but it also creates a feeling of unfairness among paying commuters.

A common way of reducing this problem is the deployment of random mass ticket inspections: stopping every single passenger in the vehicles or in the premises of the transport company, to check whether they carry a valid pass. These mass inspections interfere with the overall passenger flow and affect the customer experience, which makes them an inconvenient solution especially during rush hours. Many operators rely on the installation of ticket barriers, also called fare gates, at the entrance and/or exit of their premises, to keep fare evasion low. Yet fare evaders have learned to dodge fare gates with practices like tailgating (passing close behind a paying passenger), jumping over fare gates, entering through exit doors and other methods to avoid paying.
Under the TRAINSFARE project framework, AWAAIT is developing DETECTOR, an automatic real-time video analytics system that enables selective controls for tackling fare evasion.

Using Artificial Intelligence (AI) algorithms, DETECTOR analyses the images captured by a camera above ticket barriers and sends an alert to the smartphones of ticket inspectors when it detects fare evasion. This way only fare dodgers are intercepted, without disrupting the passenger flow and causing unnecessary checks, even during the busiest hours.

This pioneering system exerts a strong deterrent effect (offenders intercepted shortly after entry, for other passengers to see), effectively reduces fare evasion and facilitates the job of ticket inspectors.

DETECTOR enables the use of leaner ticket inspection teams that can move faster around the network, as well as a better traveling experience for paying passengers. Furthermore, ticket inspectors that have tested the system are eager to adopt it.

After successfully proving the technology and business opportunity in the earlier Phase 1 project, AWAAIT’s main objective in TRAINSFARE is to scale up and internationalize DETECTOR. Firstly, the company aims to evolve the current platform to a new generation with beyond the state-of-the-art AI techniques and methodologies. Secondly, AWAAIT aims at introducing the system across public transport operators worldwide. Finally, AWAAIT pursues to become a world class player in the development of industrial solutions that use AI.
During the first 24 months of the 36-month TRAINSFARE project, AWAAIT has performed activities in three main business areas: Dissemination, R&D&I and Human Resources.

Being a small and young company, AWAAIT's main initial challenges were to enhance the company's brand image, expand the awareness of its product offering and devise and execute the right market development strategy for sustainable growth.
AWAAIT activated its presence in social networks, redesigned its website and created a blog on fare evasion in public transport.
In parallel, AWAAIT participated in professional forums and exhibitions focused on public transport. The most relevant events were the ITS World Congress (Montreal, October 2017), the Smart City Expo World Congress (Barcelona, November 2017), the Mobile World Congress 2018 (Barcelona, February 2018), the IT-TRANS (Karlsruhe, March 2018), the Transport Research Arena (TRA, Vienna, April 2018), InnoTrans (Berlin, September 2018), Calypso Smart Ticketing & Digital Services Forum (Budapest, April 2019) and diverse UITP (International Association of Public Transport) meetings and events.
Presence in such events was essential for increasing AWAAIT’s brand awareness, for extending its contact network in the public transport sector and, most importantly, for the generation and follow-up of market development opportunities. DETECTOR has been tested in 3 European cities, and 2 additional European operators are in line to perform pilot tests at their premises. Outside the EU, another test and a demo are being held at two mass transit operators in the Americas. These tests are a direct result of AWAAIT’s communication and dissemination efforts.

The main technical objective of the project is to achieve a scalable, industrialized and fully marketable version of DETECTOR (Version 2), a leap improvement over the initial Version 1 which was a working prototype finetuned for AWAAIT’s first client's requirements and operational conditions.
In the second 12-month period, the development strategy has been mostly focused on building a scalable version of DETECTOR, with a strong focus not only on extending the analytical reach but also on the maintainability and generalizability of the system. During this period, AWAAIT has been successful in convincing additional operators to test DETECTOR at their premises. The testing will be performed in the third and last 12-month period and will help DETECTOR to become strongly generalizable and ready for fast scalability.

Human Resources
AWAAIT’s team size is growing with the project. It has grown from an initial team of three (3) people to a team of ten (10). Currently, 6 of the staff members are dedicated to R&D activities.
DETECTOR has already been recognized as a pioneering system in the combination of Artificial intelligence and mobile technologies with the aim of tackling fare evasion in public transport.

By the end of the project, we expect that DETECTOR can help in the economic sustainability of public transport globally and in creating a fairer and a more secure environment in the places where it is deployed.

We also expect it to be also a successful example of the introduction of spearheading technology (AI-Machine Learning in this case) in the field of public transport, hopefully promoting the consideration and adoption of subsequent innovation in this field.