Developers at ITMO University have presented an artificial intelligence capable of analyzing the current public transport system in a few hours and suggesting how to make it more efficient. The new service, called ConnectPT, will reduce the time and cost of designing urban routes, according to the university's press service.
According to Sergey Mityagin, Director of the Institute of Design and Urban Studies at ITMO University, the main advantage of ConnectPT is its simplicity and accessibility. "This is a light and inexpensive tool that formulates and argues its recommendations for improving routes on a minimum of initial data," he noted. The system uses open sources such as OpenStreetMap, which contains information about residential and non-residential buildings, roads, stops, places of attraction for people, and population density. Based on this data, the AI generates route options for different modes of transport and selects the optimal ones in terms of demand and connectivity.
Today, urban planners use complex mathematical models to predict passenger traffic. However, building them requires collecting a large amount of data — from statistics from telecom operators to information on the use of travel documents. Such studies can cost tens of millions of rubles and take years. ITMO's development offers an alternative: the assessment of existing routes takes only a few hours, and recommendations are formed without the need for expensive research.
ConnectPT takes into account the interests of all participants in the transport system. For residents, it can reduce travel time and the number of transfers, for carriers — remove unprofitable routes, and for the city — eliminate transport "voids" and connect enclaves without public transport. The algorithm is universal and can analyze data from any city or region of Russia.
