MATSim scenario for Krasnoyarsk

1 November 2016, 19:40

The scenario was built in 2015 as part of the preparation for a major sporting event – the 2019 Winter Universiade. The city of Krasnoyarsk has over 1 million inhabitants, with 0.3 million more living in the satellite towns of the agglomeration. The network data was extracted from the OSM and consists of 11,970 nodes and 17,552 links. 24 hours of a normal working day were simulated.

A number of scenarios were built as part of working on a comprehensive plan for sustainable transport development and on the transportation concept of the Universiade. One scenario covers current situation, another considers population growth and planned road infrastructure for the year of 2019. There is also a scenario with a redesigned PT network including dedicated bus lanes and proposed BRT corridors for 2019. Moreover, a scenario with an additional demand during the peak day of the Universiade was built.


Although a travel diary survey with ca. 1,600 participants was conducted, the prime source of data for the initial demand modeling was geolocated smartcard validations in public transport. According to the analysis of the proportion of cash transactions and validations for a number of transit lines, ca. 63% of all PT trips are paid via smart cards. For each personal ID the starting times and locations such as home, work/study, and shopping activities were extracted using time/location clusterization and validation chain analysis.

Scoring parameters were obtained from the survey results using a logit model.


The resulting plans were then scaled according to the modal split derived from the survey, assuming that plans of car users do not differ much from those of PT users. The simulated network load compared to the traffic counts shows that such an assumption delivers a plausible result.


One of the main objectives of the model was to plan the optimal design of the future rapid transit network. This was achieved using a minibus module designed by Andreas Neumann. Using the demand data for the year of 2019, an evolutionary algorithm for synthetic minibus operators was used to determine the most efficient lines and corridors for the rapid transit. The surviving 8 lines were then evaluated and transformed into 5 proposed BRT lines.