Strategic Directions in Port State Control

1
Strategic Directions in Port State Control

• Benefits of
• Assessing Ship Risk

Chris Barnes Maritime Operations
www.amsa.gov.au
2
Annual Shipping Activity in Australia
3
Very LargeGeographical Coverage
4
(No Transcript)
5
The Challenge

• 21,000 arrivals by 3,700 foreign-flag ships each
  year at 70 ports (some locations difficult to
  access, others experiencing rapid growth)
• Finite resources
• 40 Marine Surveyors at 14 Major Ports
• Wide range of responsibilities
• PSC/FSC, cargo inspections, marine
  qualifications, ISM audits
• Need to apply resources effectively
• Maritime Industry is second most
• incident-prone industry on earth

6
Risk Management Concerns

• AMSA had comprehensive records of more than
  20,000 PSC inspections
• Knew a lot about ships inspected, but
• Did not have data to provide a national overview
  of industry operating patterns and of ships not
  inspected
• Profile and nature of Risk represented by
  shipping industry was largely unknown
• Did not have a clear view as to whether the
  inspection effort was correctly focussed on
  higher risk ships

7
Strategies

• Redevelop Information Systems to record port
  arrivals
• Allocate objective risk indicators to arriving
  ships
• Use ship risk as a guide in selection for
  inspection
• Adopt performance measures that reflect risk
• Seek to focus safety surveillance effort on
  higher risk
• ships and, if appropriate, undertake inspections
  at
• most convenient port

8
Review of Best Practice

• Considered strategies in use by Paris MoU and
  USCG
• Degree of statistical analysis applied by Paris
  MOU was not clear
• Factors other than Flag appeared likely to be
  highly relevant as indicators or lack of
  seaworthiness
• European experience with certain Flags was not
  repeated in Australia
• USCG Qualships 21 program appeared to have some
  limitations
• AMSA had detained several Qualships

9
Ship Risk Assessments

• Undertook preliminary statistical analysis
  in-house
• Results and subsequent trial were very promising
• Commissioned full statistical analysis of data to
  identify and rank ship characteristics with
  predictive value in relation to likelihood of
  being unseaworthy

10
Statistical Analysis

• Must be objective and thorough
• Aim to test many ship characteristics to
  determine which ones are valid indicators of
  seaworthiness
• Having found those factors that prove to be
  useful indicators of probability of detention
• rank them in order of importance and
• estimate their relative importance

11
Methodology

• Consultants used Logistic Regression techniques,
  to see which of many possible ship
  characteristics were statistically significant
  indicators of the probability of a ship being
  found to be unseaworthy at a PSC inspection.
• ranked these factors by importance and indicated
  relative importance

p 1-p
Logit(p) Loge
12
Ship Risk Models

• As bulk carriers represented 62 of ships and 40
  of port visits, these ships were analysed
  separately from other ship types
• These 2 models were used in AMSAs database to
  allocate a Risk Factor to all ship arrivals
• Testing with rolling timeframes showed
  variations and trends from year to year
• model based on most recent 5 years data proved
  best

13
Specific Findings in 2002 Analysison Indicators
of Seaworthiness
Bulk Carriers Ship Age Previous Insp No.
Defs Flag Gross Tonnage Inspection time
gap Whether 1st Insp
Other Ship Types Ship Age Ship type Whether 1st
Inspection Previous Insp No. Defs Time since
Special Surv Gross Tonnage Flag, Recognised Org
Importance 1 1 2 2 3 3 4 4 Could be ignored
Some of the least important factors could be
ignored in the final model, as they made little
contribution.
14
How Risk Indicators were Used

• Arriving ships were allocated a Risk Factor to
  assist in selection for PSC
• The Risk Profiles of ship arrivals were
  identified, nationally and for each port
• These risk profiles were used to consider level
  of Surveyor resources needed at each AMSA Office
• Adopted Key Performance Indicators based on Ship
  Risk
• Focus of inspections on higher risk ships
• Priority for inspection based on risk

15
Port ArrivalsScreen
16
Risk Profile Trends over Time
17
How Useful?Higher Risk Factor gtgtgt deficiencies
are more likely
18
How Useful?Higher Risk Factor gtgtgt more
deficiencies found
19
How Accurate?
20
How Successful?
21
Risk Profile by AMSA Office
22
Benefits

• Recording all port arrivals and identifying ship
  risk greatly improved AMSAs ability to regulate
  the Maritime Industry in an effective manner
• Provided an overall view of the foreign-flag
  fleet (ships inspected and those not inspected)
• Ship risk profiles and trends, overall and by
  port
• Growth rates and risk profiles by port (to assist
  planning)
• Much better understanding of industry
• One third of foreign flag ships visiting in a
  year did not visit in the previous year
• Over one quarter of ships make only a single
  visit in a year (little opportunity to inspect)
• More than half of world fleet of Capesize bulk
  carriers visit Australia each year.

23
Performance MeasuresInspection Ratesby risk
group - 2006
Port visit basis
Unique ship basis
Risk Group Visits Eligible Insps Port
Visit Eligible Ships Ship Insp
Visits Insp Rate Ships Inspected Rate A
high 6526 1329 1080 81.3 852 817 96
B medium high 3767 1220 722 59.2 737
605 82 C medium 5318 1899 750 39.5
891 637 71 D low 5182 2381
525 22.0 944 479 51 Totals 2079
3 6829 3077 45 3424 2538 74
24
Inspection Priority Based on Risk
Eligible Visit No. A high B medium high C medium
low D low All risk risk
risk risk
ships





0 1.9
0.6 1.1 0.8
1.2 1 86.1 70.9 56.2 49.5
69.0 2 9.8 21.5 24.3 21.0
18.0 3 1.6 6.0 11.5 14.7
7.2 4 0.6 1.0 3.7
5.7 2.3 5 0.0 0.0
2.1 4.4 1.3 6 0.0 0.0
0.8 1.0 0.4 7 0.0 0.1
0.0 1.7 0.3 8 0.0
0.0 0.1 0.4 0.1 9 0.0
0.0 0.0 0.8 0.1
10 0.0 0.0 0.0 0.2
0.0 Totals 100 100
100 100 100
25
Recent Developments

• Clear and significant benefits of Statistical
  Risk Management in several ways prompted AMSA to
  adopt further strategies
• Update and extend PSC statistical analysis
• Adopt inspection rate targets based entirely on
  risk assessments
• Revise related performance measures

26
Aims of Further Analysis

• Analyse more recent data
• See if risk indicators are still valid or whether
  one or more new indicators should be adopted
  and/or some dropped
• See if there are objective links between ship
  operator and seaworthiness
• Undertake detailed analysis of deficiencies to
  see what relationships or trends can be
  identified
• Identify links between deficiency types and ship
  types over time
• Possibility of PSC inspection checklists tailored
  to ship type and risk profiles
• Consider deficiencies by nature (eg Operational,
  Structural/Equipment, ISM or Human Factor

27
Findings of Updated Analysis

• Some change with indicators of risk
• evolutionary rather than major changes
• A few ship operators are clearly high or low
  risk
• Limitation is that there are many operators, but
  relatively few have had sufficient inspections to
  allow a statistically-valid risk assessment
• Less than 5 of operators can be graded as being
  high or low risk most are average risk

28
Ship Operator Risk

• Analysis allowed AMSA to determine the risk
  profile of a given Operators fleet of ships
• This produced an expected detention rate for that
  operators fleet which could then be compared to
  the actual detention rate for that operator
• Where actual detention rates were clearly much
  more or much less than the expected rate the
  operator could be categorised as high or low risk
  respectively.

29
New Inspection Rate Targets

• Success of objective statistically-based risk
  assessments justified change to a full risk basis
  for setting inspection rate targets
• 5 Priority Groups have been specified
• SH for single hull tankers - 100
  inspection rate
• P1 where Risk Factor is gt5 80 rate
• P2 where Risk Factor 4 or 5 60 rate
• P3 where Risk Factor 2 or 3 40 rate and
• P1 where Risk Factor 0 or 1 20 rate
• Ships below 5 years of age eligible every 12
  months
• if they have no deficiencies

30
Distribution of Foreign-Flag Ships
by Inspection Priority
Priority 1
16
Priority 2
Priority 4
11
47
Priority 3
26
31
2007 Risk Indicators
Bulk Carriers Ship Age Previous Insp No.
Defs Inspection time gap Recognised
Org. Flag Whether 1st Insp Gross Tonnage
Other Ship Types Ship Age Ship Type Gross
Tonnage Flag Previous Insp No. Defs Inspection
time gap Whether 1st Insp Recognised Org Time
since Special Survey
Importance 1 2 2 3 3 3 4 4 No longer useful
32
Analysis of PSC Deficiencies

• Hundreds of deficiency descriptions across 25
  categories
• Many deficiency types are little used
• Some are recent additions with little history
• Some are specific to certain ship types
• AMSA grouped the many deficiencies into four main
  types to assist with analysis
• Operational
• Structural/Equipment
• Human Error
• ISM

33
Incidence of Main Deficiency Categories
Operational
Human Factor
Structural
ISM
Average Deficiencies per Inspection
25
gt25
Ship Age at Inspection
34
Deficiency Analysis

• Aim is to be able to identify probability of
  particular deficiency types occurring
• according to the age type of a given ship
• other characteristics could also be relevant
• Need to be careful that Surveyors do not become
  too narrowly focussed and still look for other
  deficiency types, however.

35
Summary

• Recording Port Arrivals and Assessing Ship Risk
  of detention has resulted in major positive
  outcomes
• Much better understanding of Maritime industry
  operating patterns
• Improved focus on higher risk ships
• Demonstrated rapid PSC response for higher risk
  ships
• Better geographical positioning of resources
• Able to produce wide variety of useful reports
• In short many strategic benefits

36
Questions?