Railway simulation using OpenTrack in Finland Tero Kosonen VR-Track Ltd / Railway Consulting Contents: Overview of the Finnish railway scene All OpenTrack simulation studies done in Finland Previews of some OpenTrack studies Helsinki main station study Pasila Riihimäki capacity increase study Summary 21.1.2008, Kosonen Tero, page 1
Track network in Finland RIIHIM ÄKI KO L ARI KEL L O SEL KÄ Length of track network 5,700 kilometres HYVINKÄÄ T O RNIO RO VANIEM I Electrified track 2,600 kilometres Remote-controlled track 2,500 kilometres VANT AANKO SKI HU O PALAHT I L EPPÄVAAR A KIR KKO NUM M I HEL SINKI KERAVA T IKKURIL A KOKKOLA KEM I RAAHE O UL U KAJAANI YL IVIESKA HAAPAJÄRVI KO NT IO M ÄKI VARTIUS Railway service locations 480 VAASA SEINÄJOKI IISAL M I SIIL INJÄRVI KUO PIO JO EN SUU UIM AHARJU IL O M ANT SI KASKINEN JYVÄSKYL Ä PIEKSÄM ÄKI SÄKÄNIEM I NIIRALA Single track Double track Four track RAUM A PO R I KO KEM ÄKI T URKU O RIVESI T AM PERE T O IJALA RIIHIM ÄKI M IKKEL I PARIKKAL A IM AT RA HEINO L A KO UVO L A IM AT RANKO SKI LUUM ÄKI VAINIKKAL A KO T KA HAM INA 21.1.2008, Kosonen Tero, page 2 HANKO HEL SINKI
Traffic volumes Freight Total amount of freight on rail is about 45 Mio tonnes / year Traffic is quite well distributed around the network Largest traffic volumes are in Eastern Finland (on some sections over 10 Mio tonnes / year) Russian import and transit traffic is about 30 % of the total volume (wood, oil, ore etc.) 21.1.2008, Kosonen Tero, page 3
Traffic volumes Passenger long distance Total amount of long distance passenger trips is about 12 Mio / year The system and most of the traffic is Helsinki-based Northern mainline between Helsinki and Oulu has the largest volumes (about 5 Mio / year Helsinki Riihimäki) Western and eastern main lines have over 1 Mio passengers Traffic to Russia is about 350 000 passengers / year 21.1.2008, Kosonen Tero, page 4
Current capacity problems RIIHIM ÄKI KO L ARI KEL L O SEL KÄ HYVINKÄÄ RO VANIEM I T O RNIO KERAVA KEM I VANT AANKO SKI OULU T IKKUR IL A RAAHE HU O PALAHT I L EPPÄVAAR A VARTIUS KAJAANI KO NT IO M ÄKI KIR KKO NUM M I HE LS INKI KOKKOLA YL IVIESKA HAAPAJÄRVI IISAL M I VAASA SEINÄJOKI SIIL INJÄRVI KUO PIO UIM AHARJU IL O M ANT SI JOENSUU Capacity totally in use Possibilities of adding new traffic limited KASKINEN RAUM A PO R I KO KEM ÄKI O RIVESI TAMPE RE T O IJALA JYVÄSKYL Ä HEINO L A M IKKEL I KOUVOLA PIEKSÄM ÄKI PARIKKAL A IM AT RA IM AT RANKO SKI SÄKÄNIEM I NIIRALA RIIHIM ÄKI LUUM ÄKI VAINIKKALA T URKU HAMINA KOTKA 21.1.2008, Kosonen Tero, page 5 HANKO HE LS INKI
OpenTrack simulation studies in Finland Helsinki main station study 2004 Ilmala passenger depot study 2005 Turku Toijala traffic analysis 2006 Tampere area traffic study 2006 New eastern high speed line study 2006 New Vuosaari harbour study 2006 Master s thesis of OpenTrack calibration 2007 KOLARI KELLOSELKÄ ROVANIEMI TORNIO KEMI OULU RAAHE KAJAANI KONTIOMÄKI YLIVIESKA KOKKOLA HAAPAJÄRVI IISALMI VARTIUS New airport line traffic study 2007 New car-carrier terminal study 2007 Master s thesis of using OT in capacity allocation process 2008 Helsinki Riihimäki capacity increase 2008 Kolari iron ore mine study 2008 VAASA KASKINEN RAUMA PORI KOKEMÄKI TURKU SEINÄJOKI TAMPERE TOIJALA RIIHIMÄKI ORIVESI JYVÄSKYLÄ HEINOLA SIILINJÄRVI KUOPIO MIKKELI KOUVOLA PIEKSÄMÄKI LUUMÄKI HAMINA KOTKA PARIKKALA IMATRA UIMAHARJU JOENSUU ILOMANTSI VAINIKKALA SÄKÄNIEMI IMATRANKOSKI NIIRALA 21.1.2008, Kosonen Tero, page 6 HANKO HELSINKI
Examples of finished studies Helsinki main station study 2004 Study was made for analyzing significant traffic increase planned for 2006 (from 967 trains/day to 1040 trains/day) Train amounts had increased in 1995 2003 over 70 % would there be capacity left for additional increase? For initial data we had only preliminary future timetable structures, all the rostering and track usage had to be planned during the project Three different traffic scenarios were simulated Existing production timetable 2003 (for comparison) 2006 traffic planned with existing operation principles 2006 traffic planned with alternative operation principles This combination made possible: Analyzing the traffic increase without operation principle change Would there be advantages in changing them 21.1.2008, Kosonen Tero, page 7
Examples of finished studies Helsinki main station study 2004 track usage schedule Lähtevät 30 31 32 3 34 35 36 37 38 39 40 41 42 43 4 45 46 47 48 49 50 51 52 53 54 5 56 57 58 59 9:00 1 2 3 4 5 6 7 8 9 10 1 12 13 14 15 16 17 18 19 20 21 2 23 24 25 26 27 28 29 30 31 32 3 34 35 36 37 38 39 40 41 42 43 4 45 46 47 48 49 50 51 52 53 54 5 56 57 58 59 10:00 1 2 3 4 5 6 7 8 9 10 1 12 13 Pendo 3 # IC # # # Taajam a # # # # # # Kaupunki # # # # # # # # # # # # Pika Saapuvat 30 31 32 3 34 35 36 37 38 39 40 41 42 43 4 45 46 47 48 49 50 51 52 53 54 5 56 57 58 59 9:00 1 2 3 4 5 6 7 8 9 10 1 12 13 14 15 16 17 18 19 20 21 2 23 24 25 26 27 28 29 30 31 32 3 34 35 36 37 38 39 40 41 42 43 4 45 46 47 48 49 50 51 52 53 54 5 56 57 58 59 10:00 1 2 3 4 5 6 7 8 9 10 1 12 13 Pendo 2 # IC # # # Taajam a # # # # # # Kaupunki # # # # # # # # # # # # # # # # Pika ## # Raide 30 31 32 3 34 35 36 37 38 39 40 41 42 43 4 45 46 47 48 49 50 51 52 53 54 5 56 57 58 59 9:00 1 2 3 4 5 6 7 8 9 10 1 12 13 14 15 16 17 18 19 20 21 2 23 24 25 26 27 28 29 30 31 32 3 34 35 36 37 38 39 40 41 42 43 4 45 46 47 48 49 50 51 52 53 54 5 56 57 58 59 10:00 1 2 3 4 5 6 7 8 9 10 1 12 13 001 9457 9065 9442 002 003 004 005 006 007 008 009 010 225 9335 9434 9743 9320 9459 9438 9326 9923 9745 9436 9746 624 309 v 9461 9328 9752 9463 9747 9067 3 309 306 v 9924 106 52 53 666 9927 53 9069 9749 9071 353 9754 222 203 9751 308 108 9465 Ilmala Ilmala 011 012 013 8678 8471 8454 8454 8327 8481 8322 109 8485 014 015 116 8325 8475 8331 8337 016 017 018 8684 8692 8701 8703 8711 8055 8061 8713 8717 8721 8723 8727 8731 019 8053 8707 8067 8071 8077 120 Saapuvat 31 32 3 34 35 36 37 38 39 41 42 43 4 45 46 47 48 49 30 40 50 21.1.2008, Kosonen Tero, page 8 51 52 53 54 5 56 57 58 59 9:00 1 2 3 4 5 6 7 8 9 10 1 12 13 14 15 16 17 18 19 20 21 2 23 24 25 26 27 28 29 30 31 32 3 34 35 36 37 38 39 40 41 42 43 4 45 46 47 48 49 Pendo IC # # 50 51 52 53 54 5 56 57 58 59 10:00 1 2 3 4 5 6 7 8 9 Taajam a ## ## ## ## # ## ## ## Kaupunki # # # # # # # # # # # # # # # # # # # Pika 10 1 12 13 Lähtevät 30 31 32 3 34 35 36 37 38 39 40 41 42 43 4 45 46 47 48 49 50 51 52 53 54 5 56 57 58 59 9:00 1 2 3 4 5 6 7 8 9 10 1 12 13 14 15 16 17 18 19 20 21 2 23 24 25 26 27 28 29 30 31 32 3 34 35 36 37 38 39 40 41 42 43 4 45 46 47 48 49 50 51 52 53 54 5 56 57 58 59 10:00 1 2 3 4 5 6 7 8 9 10 1 12 13 Pendo #
Examples of finished studies Helsinki main station study 2004 simulation model Line length Total track length 11.5 km 85.0 km Number of stations 13 Number of signals 225 Number of routes 680 Number of elements 1280 Number of train movements per simulation run - timetable 2003 350 - timetable 2006 380 - timetable 2006 improved 370 Duration of a simulation (from 4 to 10 am) ~ 15 min Number of output files per simulation run ~ 8200 21.1.2008, Kosonen Tero, page 9
100 % 90 % 80 % 70 % 60 % 50 % 40 % 30 % 20 % 10 % Examples of finished studies Helsinki main station study 2004 results 0 % 001 002 003 004 005 006 007 008 009 010 011 012 013 014 015 016 017 018 019 21.1.2008, Kosonen Tero, page 10 Occupation 2003 Occupation 2006 classic Occupation 2006 imroved Average occ. 2003 Average occ. 2006 classic Average occ. 2006 improved
Examples of finished studies Helsinki Riihimäki-line capacity increase 2008 Riihimäki Objective of the study was to determine necessary infrastructure improvements in order to increase commuter traffic amount For initial data we had preliminary timetable structures and suggestions of the infrastructure improvements Rostering and track usage schedules for Riihimäki had to be planned during the project Helsinki main yard was left out from the model because of its complexity The simulation process was done in two different companies VR-Track and Ramboll Finland Update of an existing model Järvenpää Pasila Kerava Tikkurila Hyvinkää New model Jokela Vuosaari 21.1.2008, Kosonen Tero, page 11 Helsinki
Examples of finished studies Helsinki Riihimäki-line capacity increase 2008 - model Riihimäki Hyvinkää Palopuro New bypass sidings Nuppulinna Haarajoki Järvenpää Kerava Tikkurila Pasila Vuosaari 21.1.2008, Kosonen Tero, page 12
Examples of finished studies Helsinki Riihimäki-line capacity increase 2008 - results Occupations / morning peak hours to south 90 % 80 % 70 % 60 % Occupation > 60 % Limited recovery from perturbed situations % 50 % 40 % 30 % 20 % Alt B Alt A 2007 10 % 0 % Riihimäki Paavola Hyvinkää 21.1.2008, Kosonen Tero, page 13 Palopuro Jokela Nuppulinna Järvenpää Kyrölä Kytömaa Kerava Savio / Tikkurila Tikkurila Malmi / Oulunkylä Pasila
Examples of finished studies 250,0 Helsinki Riihimäki-line capacity increase 2008 - results Perturbated traffic - alternative A Riihimäki 200,0 sum of delays (min) 150,0 100,0 50,0 0,0 1 2 3 4 5 6 7 8 9 10 11 number of perturbated run Random generated delay at start Delay at end Pasila Vuosaari 21.1.2008, Kosonen Tero, page 14
Examples of finished studies Helsinki Riihimäki-line capacity increase 2008 results With simulation we were able to define following infrastructure improvements By-pass sidings should be longer than first suggested Tikkurila station proved to be a bottleneck changes to layout should be made Also minor changes to other station layouts should be carried out Some additional improvements suggested before the simulation could be dropped out Certain improvement requests to OpenTrack were made during the project Ability to define track occupation observing intervals under way? DelEnd-file observation point should be changed done 21.1.2008, Kosonen Tero, page 15
Experiences from OpenTrack Summary With OT we have been able to study and analyze wide variety of rail traffic problems starting from small travel time analysis ending to complex dead-end track yard studies After implementing OT our scope of simulation studies has widened quite much Simulation itself is still only a tool, the user has to know exactly what he/she is doing to get correct results Until now we have focused on double track studies considering the situation in Finland, single track studies are inevitable Future challenges Single track section modelling with heavy traffic? Construction works modelling with the machine movements? 21.1.2008, Kosonen Tero, page 16