Dealing with uncertainty in railway traffic management and - - PowerPoint PPT Presentation

Challenge the future

Delft University of Technology

• Dr. Rob M.P. Goverde

Department of Transport and Planning Delft University of Technology r.m.p.goverde@tudelft.nl

Algorithmic Methods for Optimization in Public Transport Schloss Dagstuhl, April 24-29, 2016

Dealing with uncertainty

1

… in railway traffic management and disruption management

April 26, 2016

Introduction

• Traffic management of small disturbances

 Errors in traffic prediction and conflict detection  Impact on conflict resolution policy?

• Disruption management

 Uncertain disruption length  Impact of disruption length to contingency plan  Impact of error in disruption length prediction?  What to tell to the traffic controllers and passengers?

2

Uncertainty in railway operations

Outline

• Introduction
• Railway traffic management

 Traffic prediction & conflict detection  Dealing with uncertainty

• Disruption management

 Prediction of disruption length  Dealing with uncertainty

• Conclusions
• References

3

Dealing with uncertainty…

Railway traffic management

Real-Time Traffic Plan (RTTP)*

• Successive routes of trains
• Train orders over routes
• Route setting times

*Quaglietta et al. (2016)

4

ON-TIME real-time traffic management framework

Track Train TMS Real-Time Traffic Plan Traffic State Monitoring Traffic State Prediction Conflict Detection Conflict Resolution Driver Advice Interlocking Train operations

Traffic prediction & conflict detection

• Acyclic precedence graph based on RTTP and timetable
• Nodes: (microscopic) train events
• Arcs: precedence relations (run, dwell, transfer, signal headway, …)

 Arc weights estimated from historical track occupation data conditional

• n actual circumstances*

*Kecman and Goverde (2015ab)

5

Data driven approach*

Traffic prediction & conflict detection

*Kecman and Goverde (2015b)

6

Prediction at 7:13

Track conflict Schiedam Inbound route conflict Rtd Inbound route conflict Conflict resolution

• Slow down IC9216

before Sdm

• Slow down IC 2127

before Rtd

Traffic prediction & conflict detection

*Kecman and Goverde (2015b)

7

Realization (after 8:20)

Track conflict Schiedam Inbound route conflict Rtd Inbound route conflict Inbound route conflict Conflict resolution

• Slow down IC9216

before Sdm

• Slow down IC 2127

before Sdm

Traffic prediction & conflict detection

*Kecman and Goverde (2015b)

8

Prediction at 7:13 and realization

Realized (black) Predicted (colour)

Traffic prediction & conflict detection

*Kecman and Goverde (2015b)

9

Adaptive prediction

Traffic prediction & conflict detection

*Kecman and Goverde (2015b)

10

Prediction errors

Dealing with uncertainty

• Adaptive prediction of train paths
• Model-based predictive control*

*Quaglietta et al. (2013)

11

… in traffic management of disturbances

Disruption management

1st phase: Situational awareness, contingency plan, transition 2nd phase: Operate to contingency plan, prediction disruption length 3rd phase: Return transition to normal timetable

12

Bathtub model

1st phase 2nd phase 3rd phase Failure Traffic intensity Time Disruption length

Prediction of disruption length

*Zilko et al. (2016)

13

Example

• 1 July 2014 in Sloterdijk
• Disruption 51+101=152 min

Mean +/- st.dev.

Prediction by mean 1) 104 min (initial)

Copula Bayesian Network (BN) of track circuit failure

Prediction of disruption length

*Zilko et al. (2016)

14

Conditionalized BN after situational information

Prediction by mean 1) 104 min (initial) 2) 134 min (siuational)

Example

• 1 July 2014 in Sloterdijk
• Disruption 51+101=152 min

Prediction of disruption length

*Zilko et al. (2016)

15

Conditionalized BN after contractor diagnosis

Prediction by mean 1) 104 min (initial) 2) 134 min (situational) 3) 150 min (post diagnosis)

Example

• 1 July 2014 in Sloterdijk
• Disruption 51+101=152 min

Dealing with uncertainty

When prediction gives a wide distribution

• What to tell to the traffic controllers and passengers?

 Entire distribution  Mean and standard deviation  Mean, median or mode (most likely prediction)  Low percentile (optimistic prediction)  High percentile (pessimistic prediction)

• Optimization of traffic control measures during disruption

 Scenario analysis

• Get more and better data to decrease variance

 Improve registration of disruption details

• What exact part failed, how was it repaired?

16

… in disruption management

Conclusions

• Traffic management of small disturbances

 Errors in traffic prediction and conflict detection

• Stochasticty, rare cases, trains or routes without data, …

 Adaptive prediction  Model-based predictive control

• Disruption management

 Uncertain disruption length  Scenario analysis  Pessimistic, most likely, and optimistic predictions

17

Dealing with uncertainty…

References

1. Quaglietta, E., Corman, F., Goverde, R.M.P. (2013). Stability analysis of railway dispatching plans in a stochastic and dynamic environment. Journal of Rail Transport Planning and Management, 3(4), 137–149. 2. Quaglietta, E., Pellegrini, P., Goverde, R.M.P., Albrecht, T., Jaekel, B., Marlière, G., Rodriguez, J., Dollevoet, T., Ambrogio, B., Carcasole, D., Giaroli, M., Nicholson, G. (2016). The ON-TIME real-time railway traffic management framework: A proof-of- concept using a scalable standardised data communication architecture. Transportation Research Part C: Emerging Technologies, 63, 23-50. 3. Kecman, P., Goverde, R.M.P. (2015a). Predictive modelling of running and dwell times in railway traffic. Public Transport, 7(3), 295-319. 4. Kecman, P., Goverde, R.M.P. (2015b). Online data-driven adaptive prediction of train event times. IEEE Transactions on Intelligent Transportation Systems, 16(1), 465-474. 5. Zilko, A.A., Kurowicka, D., Goverde, R.M.P. (2016). Modelling railway disruption lengths with Copula Bayesian Networks. Transportation Research Part C: Emerging Technologies.

18

Cited in presentation