changeand’safety:decisionmaking’ from’data’william/presentation/rruka-2012.pdf · 0.7 0.4...
Post on 09-Oct-2020
3 Views
Preview:
TRANSCRIPT
Change and Safety: Decision-‐Making from Data
Anna Holloway, Safety Knowledge and Planning, RSSB London William Marsh, School of Electronic Engineering and Computer Science, Queen Mary, University of London
• Incident data risk estimate decision: many applications • Local prediction needed • Local prediction possible with data and knowledge • Bayesian network: example application
Boarding and Alighting Accidents
1.9
1.2 1.2 0.9
1.3
0.5 0.7
0.4 0.7 0.7
2.1 1.9 1.9
2.6 2.2
0.6 0.7 1.1
0.9 1.3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
2006
/07
2007
/08
2008
/09
2009
/10
2010
/11
2006
/07
2007
/08
2008
/09
2009
/10
2010
/11
2006
/07
2007
/08
2008
/09
2009
/10
2010
/11
2006
/07
2007
/08
2008
/09
2009
/10
2010
/11
Fall between train and platform
Caught in train doors
Other alighting accident
Other boarding accident
FWI
Shock & trauma Minor injuries
Accidents to passengers geEng on and off trains
From 2011 Annual Safety Performance Report
How to reduce Boarding / AlighLng harm StaLon staffing (local) StaLon Design (local) Train design (more global)
Incident data – many examples Safety: SPADs, broken rails, bridge strikes, …. Reliability: signal failure, staff absence, ….
Decision Making is Local
Not enough data for local decision making directly
Just Data for Decision Making?
0.E+00
2.E-07
4.E-07
6.E-07
8.E-07
1.E-06
1.E-06
VAL
WK
K
WA
F W
MW
W
WR
W
GV
W
RP
WFJ
W
EE
WLP
W
MS
WC
L W
HD
W
RU
W
SB
WES
W
YB
W
HM
W
HT
WB
D
WW
L W
DD
W
MI
WN
E W
IH
WSF
W
IV
WO
L W
DL
WSR
W
OF
WPL
W
RS
WM
D
FWI/
expo
sure
0
200
400
600
800
1000
1200
0 1 2 3 4 5 6 7 8 9 10
11-1
2 13
-14
15-1
6 17
-18
19-2
0 21
-25
26-3
0 31
-35
>35
Num
ber
of S
tatio
ns
Number of Incidents
Normalised observed harm (FWI) – some staLons: spikey
DistribuLon of incident count: many zeros
Model Concept: Data and Knowledge
Knowledge: causal analysis of incidents
EsLmated effect of causes on incidents
• ORR Station Usage • TSDB • DfT – Significant
Steps Research • DFT National Travel
Survey • SRM Normalisers • MET Office • APRS • T763 dispatch data
How railway is used Prevalent of
causes
Incident data Presence of
causes (e.g. ice, crowding)
If ice is a cause of falls then we expect iciness to occur in incident reports more o\en than we expect from the
prevalence of icy condiLons
Causes
Incidents
How the railway is used
Profile: region, station, …
Distinguish individual and aggregate risk
Model Queries
5.2E-‐08 5.3E-‐08 5.4E-‐08 5.5E-‐08 5.6E-‐08 5.7E-‐08 5.8E-‐08
Individual FWI
Aggregate FWI (propor4onal)
!"#$%&
!'("#$%&
!')"#$%&
!'*"#$%&
!'+"#$%&
!'!"#$%&
!',"#$%&
!'-"#$%&
!'%"#$%&
!'."#$%&
,"#$%&
/01&
/23&
405
&40"
&401&
406
&403
&407
&485
&481&
489
&483
&458
&45:
&451&
459
&45;
&4<8
&4<"
&4<:
&4<=
&4<>
&4<7
&4"0
&4":
&4"=
&4"7&
4??&
4?2&
4?1&
4@0
&4@6
&4@7
&4@4
&4:5
&4:"
&4:2&
4:=
&4:9
&4:A
&4:B
&42<&
42C&
42=
&42/&
4D<
&4D@
&4DD
&415&
41"&
41@
&41=
&
ValidaLon Strength of different effect SensiLvity to approximaLons Consistency with data
Other applicaLons
Tools to support model building
Future Work
Thank You Thanks to the Safety Knowledge and Planning and Safety Intelligence groups at RSSB for their generous collaboration
• Incident data risk estimate decision • Local prediction • Data and knowledge • Bayesian network: example application
top related