Matching supply and demand in large networks with focus on railways - PowerPoint PPT Presentation

Matching supply and demand in large networks with focus on railways Dr. Andreas Oetting 3 rd International Conference on Applied Infrastructure Research Berlin, 9 th of October 2004 Dr. Andreas Oetting 09.10.2004 1

lines and nodes model the infrastructure Dr. Andreas Oetting 09.10.2004 2

model design scenarios, travel time, characteristics of supply transport demand long-distance short-distance freight traffic passenger traffic passenger traffic Planungswelt ProSim passenger and freight trains matching supply and demand on one network Dr. Andreas Oetting 09.10.2004 3

capacity and waiting times as characteristics of supply line 1 node line 2 ? + + = capacity capacity capacity line 1 node line 2 + + = waiting time waiting time waiting time sum of waiting time line 1 node line 2 Dr. Andreas Oetting 09.10.2004 4

waiting times scheduled waiting times waiting times waiting times during operation • can be measured waiting times • are perceived by the user • may be processed together with other characteristics of supply • have an impact on route choice, transport demand, costs and benefits • can be communicated politically • quantify the quality of supply Dr. Andreas Oetting 09.10.2004 5

calculation of waiting times example example network element IC RE ICE an increasing number of trains leads to higher waiting times and a lower level of supply desired departure time waiting time minimum waiting headway time IC waiting time RE number time of trains Dr. Andreas Oetting 09.10.2004 6

waiting time and capacity The calculation of waiting times is a condition for the determination of the capacity waiting times desired quality practical capacity Dr. Andreas Oetting 09.10.2004 7

waiting times in nodes input data waiting times are produced • infrastructure • load-dependant in 3 areas A A B B C C platform zone switching zones • flows in a node • load-independant A B A B (e.g. locomotive change) C C Dr. Andreas Oetting 09.10.2004 8

modelling infrastructure and network usage input data • matrix of flows • infrastructure and flows to A A A B C from B B A — — 3 C C B — 2 4 — — C 1 • conflicts between flows flow interfering flows A B 1 2, 3 2 1, 4 3 1 4 2 C Dr. Andreas Oetting 09.10.2004 9

calculation of waiting times for the switching zone waiting time for the switching zone • formula for waiting times • input data ⋅ 2 2 N z ∑ • number of trains N = ∑ t = t t N w − ⋅ Z N z w , Zug w Züge • coefficient of variation of the distances Züge t w waiting time between arriving trains z minimum headway • minimum headway N number of trains Z length of time segment • queuing theory • equivalent number of trains  ⋅    n z Z ∏ ( GI / GI / 1 / ∞ ) ∏ − ρ = − ρ ( 1 ) 1  j  ⇔ = ⋅ − − n 1 1     j V V z Z     j j size of arrivals ρ occupation of switching zone by flow holding area n number of trains of a flow service j index for interfering flows number of service lines v index for equivalent interfering flow Dr. Andreas Oetting 09.10.2004 10

future new constructions of DB complement the high speed network Fulda Frankfurt Hanau Würzburg Mannheim Dr. Andreas Oetting 09.10.2004 11

the comparison of infrastructure and network utilisation identifies bottlenecks 1 2 5 109> 0 < > 25< 25> 22< 18< 18> 100< 22> 18< 101> 19> 1 3 1 1 100< < 4 1 101> 42< > 44> 102< 18< 81< 18> 97> 84> 98< 101> 33< 113< 30> 63< 63> 15 4 < 8 < 181< 103> 3 174> 1 < 6 1 5 151> 102< 23< 0 > 7 > 47< 110< 5 3 30> 8 > 115> 1 6 4 110> 68< 92< 5 < 5< 9 8 < 1 < 23< 9< < 7 23> 3 21< 7 3 4 3 7> 6 21< 2> < 116< 2 69< 2 162< > > 42> 73> 0 > 7 2 21 3 2 93 Dr. Andreas Oetting 09.10.2004 12